Adaptive deep brain stimulation for treatment of chronic pain

Abstract

Devices, systems, software, and methods are provided for treating chronic pain in an individual using deep brain stimulation (DBS). An implantable brain stimulation and neural recording device is used to identify personalized pain biomarkers to automate brain stimulation using adaptive DBS algorithms in response to pain-associated brain states. Methods are also provided for identifying effective personalized brain stimulation sites for an individual through a systematic, multi-site brain mapping trial and tailoring real-time adaptive DBS programs to an individual to provide significant pain relief.

Description

ADAPTIVE DEEP BRAIN STIMULATION FOR TREATMENT OF CHRONIC PAINCROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims benefit of U.S. Provisional Patent Application No. 63 / 544,894, filed October 19, 2023, which application is incorporated herein by reference in its entirety.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under Grant No. UH3NS115631 and Grant No. UH3NS109556 awarded by the National Institutes of Health. The government has certain rights in the invention.BACKGROUND OF THE INVENTION

[0003] Chronic pain affects up to 20% of the US population (Dahlhamer et al. (2018) MMWR Morb Mortal Wkly Rep 67(36):1001 -1006); however, many pain syndromes are refractory to all available therapies. As chronic neuropathic pain syndromes are associated with brain-wide circuit changes (Apkarian et al. (2013) Curr Opin Neurol 26(4):360-7, Apkarian et al. (2004) J Neurosci 24(46):10410-5, Baliki et al. (2012) Nat Neurosci 15(8):1 117-9, Hashmi et al. (2013) Brain 136(9):2751 -68), electrical brain stimulation is an appealing strategy to treat refractory pain. Despite over 60 years of study (Mazars et al. (1960) Rev Neurol (Paris) 103:136-8), brain stimulation for pain has yielded highly variable and inconsistent benefits across patients with only one randomized, controlled trial to date (Frizon et al. (2020) Neurosurgery 86(2):191-202).

[0004] Prior attempts of deep brain stimulation (DBS) for pain have used a one-size-fits-all approach by targeting the same brain regions across patients despite known interindividual variation in brain anatomy and physiology. DBS for pain could be significantly improved by optimizing both stimulation location and real-time electrical dosing in response to brain biomarkers of ongoing symptom state.

[0005] Traditional DBS targets for pain relief have included the periaqueductal / periventricular gray (PVG) and sensory thalamus (VPthal) which play key roles in endogenous opioid release and ascending pain transmission, respectively (Hosobuchi et al. (1975) Surg Neurol 4(1 ):91 -2, Coffey (2001) Pain Med Malden Mass 2(3):183-92). However, more recent neuroimaging studies of experimental and spontaneous pain reveal numerous other hubs that may engage descending pain modulation circuits for analgesia (Wager et al. (2013) N Engl J Med 368(15):1388-97, Davis et al. (2020) Nat Rev Neurol 16(7):381 -400). These include the anterior cingulate cortex (ACC), insula, ventromedial prefrontal cortex (vmPFC), basal ganglia and centromedian thalamus (CM).Although recent attempts at ACC (Boccard et al. (2017) World Neurosurg 106(Supplement C):625- 37) or CM (Hollingworth et al. (2017) Brain Sci 7(1 ):9). Although DBS has shown transient benefit in non-blinded, non-randomized case-series, identifying the best stimulation target in a particular individual is complicated by significant interindividual variability in pain representations (Kohoutova et al. (2022) Nat Neurosci 25(6):749-59).SUMMARY OF THE INVENTION

[0006] Devices, systems, software, and methods are provided for treating chronic pain in an individual using deep brain stimulation (DBS). An implantable brain stimulation and neural recording device is used to identify personalized pain biomarkers to automate brain stimulation using adaptive DBS algorithms in response to pain-associated brain states. Methods are also provided for identifying effective personalized brain stimulation sites for an individual through a systematic, multisite brain mapping trial and tailoring real-time adaptive DBS programs to an individual to provide significant pain relief.

[0007] In one aspect, a method for treating chronic pain in a subject is provided, the method comprising: positioning a first electrode at a first location of the brain of the subject to deliver electrical stimulation to a target stimulation site to relieve the chronic pain; positioning a second electrode at a second location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; detecting the brain electrical signal at the sensorimotor cortex region of the brain the subject using the second electrode; analyzing the brain electrical signal using a binary pain classification model that classifies the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state; and applying electrical stimulation to the target stimulation site in the brain of the subject using the first electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state.

[0008] In certain embodiments, the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.

[0009] In certain embodiments, the method further comprises calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.

[0010] In certain embodiments, the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region. In some embodiments, the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain. In some embodiments, the OFC region is a contralateral OFC region.In certain embodiments, the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.

[0011] In certain embodiments, the target stimulation site is in an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.

[0012] In certain embodiments, the method further comprises mapping the brain of the subject to identify an optimal location in the sensorimotor cortex region to detect the brain electrical signal associated with the chronic pain.

[0013] In certain embodiments, the method further comprises mapping the brain of the subject to identify an optimal target stimulation site to deliver the electrical stimulation to relieve the chronic pain. In some embodiments, mapping comprises identifying the optimal target stimulation site in each cerebral hemisphere.

[0014] In certain embodiments, the chronic pain is caused by a pain-associated disorder, wherein applying the electrical stimulation relieves the pain. In some embodiments, the pain-associated disorder is post-stroke pain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-related pain, peripheral vascular disease, pain after an amputation, visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, or chronic leg pain, neck pain, foot pain, or arm pain. In some embodiments, the nerve-related pain is caused by diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy.

[0015] In certain embodiments, the pain-associated disorder is said post-stroke pain, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a right ACC region and a left ACC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right ACC region, a right VPthal region, and a left ACC region.

[0016] In certain embodiments, the pain-associated disorder is said complex regional pain syndrome, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a left ACC region and a right VMPFC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right CM region and a left body of the caudate nucleus region.

[0017] In certain embodiments, the brain electrical signal is a local field potential.

[0018] In certain embodiments, the method further comprises using a control algorithm to automate said applying electrical stimulation when the brain electrical signal exceeds the specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state. In some embodiments, the control algorithm uses a machine learning algorithm for pain classification. In some embodiments, the machine learning algorithm is a supervised machine learning algorithm. In some embodiments, the machine learning algorithm is an unsupervised machine learning algorithm. In some embodiments, the machine learning algorithm is a semisupervised machine learning algorithm. In some embodiments, the control algorithm further modulates one or more programmed stimulation parameters to relieve the chronic pain. In some embodiments, the control algorithm further uses linear discriminant analysis (LDA) to adjust stimulation amplitude, or frequency of the electrical stimulation.

[0019] In certain embodiments, the electrical stimulation is applied unilaterally or bilaterally.

[0020] In certain embodiments, the brain electrical signal comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations. In some embodiments, the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz. In some embodiments, the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.

[0021] In certain embodiments, the control algorithm further modulates one or more programmed stimulation parameters based on a level of power of the brain electrical signal. In some embodiments, the level of power is a level of delta power from a contralateral OFC region. In some embodiments, changes in power features associated with the chronic pain are sustained for at least 4 seconds.

[0022] In certain embodiments, the first electrode is placed within or on a surface of the anterior cingulate cortex (ACC) region, the subgenual cingulate gyrus (SGC) region, the insula region, the motor / sensory cortex region, the ventral striatum (VS) region, the thalamus region, the thalamusventral posterior complex (VPthal) region, the centromedian thalamus (CM) region, the periaqueductal gray region, the periventricular gray (PVG) region, the caudate nucleus region, the ventromedial prefrontal cortex (VMPFC) region, or the frontal operculum region.

[0023] In certain embodiments, the second electrode is placed within or on a surface of the OFC region, the ACC region, the insula region, the vmPFC region, the BG region, or the CM region.

[0024] In certain embodiments, the first electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

[0025] In certain embodiments, the second electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

[0026] In certain embodiments, the second electrode is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.

[0027] In certain embodiments, the method further comprises assessing effectiveness of the treatment in the subject. In some embodiments, the assessing comprises using a visual analog scale (VAS), a pain intensity numerical rating score (NRS), or a verbal rating scale.

[0028] In another aspect, a computer implemented method for programming a deep brain stimulation (DBS) device to treat chronic pain in a subject is provided, the computer performing steps comprising: a) receiving recorded brain electrical signal data from a sensorimotor cortex region of the brain of the subject; b) analyzing the recorded brain electrical signal data using a binary pain classification model that classifies the brain electrical signal data as indicating that the subject is in either a high pain severity state or a low pain severity state; c) adjusting one or more programmed stimulation parameters based on the recorded brain electrical signal data according to an algorithm control law; and d) instructing the DBS device to apply an electrical stimulation to a target stimulation site in the brain of the subject using the first electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal data indicates the subject is transitioning to the high pain severity state or is in the high pain severity state.

[0029] In certain embodiments, the brain electrical signal data comprises field potential data.

[0030] In certain embodiments, the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.

[0031] In certain embodiments, the computer implemented method further comprises calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.

[0032] In certain embodiments, the computer implemented method further comprises splitting the recorded brain electrical signal data into consecutive time epochs. In some embodiments, the computer implemented method further comprises calculating average power values for one or more neurophysiological activity frequency bands for each time epoch, wherein the average power values are used to calculate pain scores for each time epoch. In some embodiments, the computer implemented method further comprises calculating pain scores using brain electrical signal data during a stimulation-free data epoch to distinguish the pain scores that are at or above the 70thpercentile corresponding to the high pain severity state and the pain scores that are at or below the 30thpercentile corresponding to the low pain severity state. In some embodiments, the stimulation- free data epoch comprises at least 5 minutes, 10-minutes, 30-minutes or 40 minutes of time of the recorded brain electrical signal data.

[0033] In certain embodiments, the computer implemented method further comprises: a) ranking predicted stimulation effectiveness for available settings of the DBS device based on classifier scores for stimulation effectiveness of each setting using a linear classification model; b) selecting stimulation settings predicted to have highest stimulation effectiveness based on the linear classification model; c) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness; d) analyzing the recorded brain electrical signal data to evaluate neural response of the subject to the electrical stimulation; e) updating the linear classification model based on the neural response of the subject to the electrical stimulation to generate an updated linear classification model; f) updating the ranking of predicted stimulation effectiveness for the available settings of the DBS device using the updated linear classification model; g) selecting stimulation settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; h) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying the electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; and i) repeating e) - h) to adjust the available settings of the DBS device to optimize stimulation effectiveness.

[0034] In certain embodiments, the linear classification model uses linear discriminant analysis (LDA) to adjust amplitude of current and frequency of the electrical stimulation.

[0035] In certain embodiments, the recorded brain electrical signal data comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations. In some embodiments, the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz. In some embodiments, the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.

[0036] In certain embodiments, the algorithm control law further modulates one or more programmed stimulation parameters based on a change in a level of power of a brain electrical signal. In some embodiments, the level of power is a level of delta power from a contralateral OFC region. In some embodiments, the change in the level of power is sustained for at least 4 seconds.

[0037] In certain embodiments, the computer implemented method further comprises storing a user profile for the subject comprising information regarding the recorded brain electrical signal data associated with the chronic pain.

[0038] In certain embodiments, the computer implemented method, further comprises storing a user profile for the subject comprising information regarding the programmed stimulation parameters used to apply electrical stimulation to the target stimulation site in the brain of the subject to treat the chronic pain in the subject based on the recorded brain electrical signal data.

[0039] In another aspect, a non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the computer implemented method, described herein, for programming a deep brain stimulation (DBS) device to treat chronic pain in a subject is provided.

[0040] In another aspect, a kit comprising the non-transitory computer-readable medium described herein and instructions for treating chronic pain in a subject with a deep brain stimulation device is provided.

[0041] In another aspect, a system for treating chronic pain in a subject is provided, the system comprising: a first electrode adapted for positioning at a location to deliver electrical stimulation to a target stimulation site to relieve the chronic pain in the subject; a second electrode adapted for positioning at a location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; and a processor programmed according to a computer implemented method, described herein, to adjust one or more stimulation parameters based on the recorded brain electrical signal data and instruct the first electrode to apply an electricalstimulation to the target stimulation site in the brain of the subject in a manner effective to treat the chronic pain in the subject.

[0042] In certain embodiments, the brain electrical signal data comprises field potential data.

[0043] In certain embodiments, the first electrode of the system is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

[0044] In certain embodiments, the second electrode of the system is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

[0045] In certain embodiments, the second electrode of the system is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.

[0046] In certain embodiments, the system further comprises a user interface comprising an input electronically coupled to the processor for instructing the first electrode to apply an electrical stimulation to the target stimulation site to relieve the chronic pain in the subject. In some embodiments, the user interface is password protected and is operable by a health care practitioner.BRIEF DESCRIPTION OF THE DRAWINGS

[0047] FIGS. 1A-1 E. Long-term ambulatory tracking of chronic pain metrics. FIG. 1A, Self-drawn body maps corresponding to the anatomical distribution of each of four participants’ spontaneous chronic pain location. Red and blue indicate areas of high and low pain, respectively. FIG. 1 B, Scatterplot of an example participant’s report of overall pain intensity VASs over 105 d (mean 3.2 reports / d), with overlying moving average (red line; window = 3 samples), demonstrating a range larger than MCID. Each black point represents one pain report simultaneous with a neural recording. FIG. 1 C, Histogram of each participant’s reported pain intensity NRS; most values were high (>6 / 10) but similar across participants. FIG. 1 D, Group data demonstrating high correlation between VAS and NRS for pain intensity and unpleasantness across participants who reported them (CP3-4) with associated Pearson’s correlation R. FIG. 1 E, Partial autocorrelation stem plots for each participant’s pain NRSs. Different pain score reporting frequencies for each participant resulted in different autocorrelation resolution. Bold stems indicate time lags achieving statistical significance P< 0.05) based on two-sided 95% confidence intervals (for CP1 -4, respectively: ±0.12, 0.14, 0.13 and 0.09) not corrected for multiple comparisons.

[0048] FIGS. 2A-2E. Ambulatory neural recordings from ACC and OFC predict chronic pain state. FIG. 2A, Example X-ray of a participant with bilateral implant of Activa PC + S DBS generators attached to depth leads in the ACC and paddle leads in the OFC (red highlights). FIG. 2B, Grouplocalization of all electrode contacts in coronal (top) and sagittal (bottom) view. Blue shaded area is the ACC; yellow shaded area is OFC. Below are example raw LFP recordings from ACC (top) and OFC (bottom). FIG. 2C, Summary of the chronic pain-state decoding scheme for pain VAS using example data from one participant. Normalized power spectra are computed for each recording (ordered by increasing pain VAS (overlaid white circles) for display) and power values at each frequency are displayed across all recordings for an example participant. Above the color bar scale, a horizontal histogram of pain VAS shows the distribution of pain scores, which are split by the median value to define a dichotomous response variable (high (1 ) versus low (0) pain states). Average power values in frequency bands of interest serve as predictive features in two complementary decoding schemes. Power features sub-selected from either brain region or hemisphere are used to train models that decode high versus low pain states using LDA. FIG. 2D, Bar plots of decoding performance with successful pain-state prediction based on NRS for all participants using LDA. CP1 - CP4 had n = 89, 137, 234 and 452 independent simultaneous recordings and pain score reports, respectively. One-sided empirical Rvalues were calculated using permutation tests (n = 1 ,000; black dots), without correction for multiple comparisons (Methods) as reported in FIG. 9 for all metrics and models. R values for NRS models in FIG. 2D from left to right are: CP1 : 0.092, 0.801 , 0.016; CP2: 0.001 , 0.001 , 0.001 , 0.001 , 0.001 , 0.002, 0.001 , 0.003, 0.777; CP3: 0.001 , 0.005, 0.041 , 0.001 , 0.002, 0.027, 0.001 , 0.001 , 0.143; CP4: 0.001 , 0.001 , 0.001 , 0.001 , 0.001 , 0.001 , 0.001 , 0.612, 0.001. FIG. 2E, Normalized mean feature weights (importance) for each participant from full models (bilateral OFC / ACC) in FIG. 2D. (Note CP1 only has a unilateral implant. Contra, brain hemisphere contralateral to participants’ body side with chronic pain; ipsi, ipsilateral. tR< 0.05, *P< 0.01 , fR< 0.001 ).

[0049] FIGS. 3A-3G. Acute pain-state prediction with ACC and OFC neural recordings is unreliable. Experimental scheme and decoding performance for acute, evoked pain state. FIG. 3A, Quantitative sensory testing thermode placement at most painful region on the side affected (aff-ACUTE) or unaffected (unaff-ACUTE) with chronic pain. FIG. 3B, Acute, thermal pain protocol (see text). FIG. 3C, Distribution of pain NRS during the task for all participants. FIG. 3D, Example temperature (blue) and NRS (red dot) data for one testing session. FIGS. 3E, 3F, LDA decoder performance for high / low acute pain NRS is shown when thermal pain was applied on the side of the body either affected (e; for CP1-4; n = 13, 20, 25 and 25 independent trials, respectively) or unaffected (FIG. 3F, for CP1 -4; n = 16, 16, 25 and 25 independent trials, respectively) by usual chronic pain. Gray points show chance level performance based on permutation tests (n = 1 ,000) used to calculate one-sided empirical R values without multiple-comparisons correction. Significant Rvalues from left to right in FIG. 3E: CP1 : 0.045; CP2: 0.012, 0.008, 0.037 and 0.013. See FIG. 15 for all R values for FIGS. 3E and 3F. FIG.3G, Normalized mean feature weights (importance) from full models for the two participants that showed significant acute-affected pain decoding according to the color scale. *P< 0.05 (FIG. 15).

[0050] FIGS. 4A-4B. Decoding of chronic and acute pain states are differentially supported by ACC and OFC features across participants. FIG. 4A, Relative importance of OFC versus ACC normalized power features to decoding of chronic pain NRS full models in each participant. Individual histograms show distributions of the magnitudes of OFC feature weights minus ACC feature weights across all recording sessions for real data (black, left facing) and shuffled surrogate data (gray, right facing; Methods). Values above 0 indicate greater OFC weights, and values below 0 indicate greater ACC weights. Contralateral OFC delta power was more important than ACC for discriminating high versus low pain across all participants (red highlight). FIG. 4B, Relative feature importance for acute-affected pain for the two participants that had significant decoding. For acute pain, there was a shift to greater ACC importance across frequencies compared with chronic pain. Two-sided Wilcoxon rank-sum test. R values corrected for multiple comparisons with the Benjamini-Hochberg method. In FIG. 4A, *P < 10-4; **P = 0.002 (CP1 ), **P = 0.043 (CP2), **P = 0.001 (CP3). In FIG. 4B, *P < 10-3; **P= 0.046 (CP1), **P= 0.042 (CP2).

[0051] FIGS. 5A-5D. Temporal dynamics of power features distinguish chronic from acute pain states. FIG. 5A, Example power time-series plot of a single feature (contralateral OFC delta) averaged across all recordings from low (blue line) versus high (red line) pain states, for one participant. Colored and shaded error bars show the s.e.m. The blue square in the upper-left corner represents that feature’s average weight as plotted in FIG. 2e (blue indicates negative weight ~ -1 .0). This negative feature weight corresponds to decreased mean power during high pain states; periods of decreased power are highlighted with background gray shading. FIG. 5B, Box plots with overlying raw data of the percentage of total recording clip time on which increases or decreases occurred for chronic pain features (black dots) and acute pain features (red dots). FIG. 5C, Mean duration of bouts of increases and decreases similarly. FIG. 5D, Number of increases or decreases per second for chronic and acute pain features; the symbol legend in FIG. 5D applies to FIGS. 5B-5D. In FIGS. 5B-5D for CP1-4 chronic pain features, respectively, n = 215, 455, 880 and 1 ,765 independent power time series. For CP1 and CP2 acute pain features, respectively, n = 35 and 60 independent power time series. Box plot bounds indicate the 25thand 75thpercentiles, the pink line shows median, and whiskers show the full extent of data from minima to maxima with points outside the whiskers considered outliers. Two- sided Wilcoxon rank-sum tests with correction for multiple comparisons. In c, ZCP1 = 7.7, *P= 10-14; ZCP2 = 11.6, *P = 10-30. In FIG. 5D, ZCP1 = -8.9, *P = 10-19; ZCP2 = -12.5, *P = 10-35. See FIG. 18 for additional details and the top five features per participant.

[0052] FIGS. 6A-6B. DTI based electrode targeting and signal stability. FIG. 6A shows a Diffusion Tensor Image tracing of the cingulum bundle bilaterally in one subject (green fibers represent those running in anterior-posterior direction) and example trajectory of bilateral depth electrodes targeting ACC (blue stems). D,A,L represent dorsal, anterior and lateral, respectively. FIG. 6B shows normalized power in multiple frequency bands from each brain region over time for each subject. There were no clear trends that indicated nonneural changes in recordings over time. Top legend applies to subject CP1 while legend in box on right applies to CP2-4. Gaps in data over time indicate missing recordings or those eliminated due to misalignment with pain score reporting.

[0053] FIGS. 7A-7C. Confusion Matrices of LDA Full Models. Panels A-C show LDA classification tables with raw counts of all true versus predicted samples for all full models of spontaneous, chronic pain (FIG. 7A), and acute pain on the affected side (FIG. 7B) and unaffected side (FIG. 7C). Box color intensity represents proportion of total samples in blue for accurate classifications and red for false positives or negatives.

[0054] FIG. 8. Chronic Pain Subregion Regression Model Performance (LASSO regression). Heatmap table shows coefficient of determinations (R2), root mean squared error (RMSE), and onesided p-values from permutation tests (n=1000, without multiple comparisons adjustment) in parenthesis for all calculated chronic pain models (across columns) and pain score report types (along rows) for regression. R2also represented in color based on color bar scale on right. The only R2values > 0.3 occurred for CP4 Unpleasantness NRS for models including OFC+ACC and for SF-MPQ, when OFC was included in models.

[0055] FIG. 9. Chronic Pain Subregion Classification Model Performance (LDA). Heatmap table shows AUG and one-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) below in parenthesis for all cross- validated chronic pain models (across columns) and pain score report types (across rows) underlying FIG. 2. AUC also represented in color based on color bar scale on right.

[0056] FIG. 10. Example LDA classification performance at individual recording session level in one subject. Example of classifier performance over recordings using LDA with leave one out cross validation for CP2 pain NRS full model. Filled red circles indicate successful classification, while empty circles indicate misclassified pain reports. Dashed horizontal line shows median NRS value for this example.

[0057] FIG. 11. Chronic pain subregion Model Stability over time (LDA) using 70 / 30 split training / testing datasets. Heatmap table shows AUC and one-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) below in parenthesis for all calculatedchronic pain models using the first 70% of data for training and the last 30% for testing (across columns) and pain score report types (across rows). For pain intensity NRS, all models were similarly significant to FIG. S4 except for CP1 contraOFC and CP2 IpsiACC / OFC. Note that pain intensity VAS for CP3 and CP4 were not significant (when using 70 / 30 split method) due to increased variance in the latter half of the data reflecting decreased consistency in VAS reporting during this period. Missing data for CP4 indicates occasions where the last 30% of data contained no variation in pain metric values so models could not be computed.

[0058] FIGS. 12A-12E. LSSM Schematic: Dynamic models can decode chronic pain state from neural activity. FIG. 12A) The decoder consists of a dynamic linear state space model (LSSM) that summarizes the information in neural features into a latent state, which is then provided to a classifier to estimate the probability of high chronic pain. Neural features consisted of log-powers computed in canonical frequency bands in non-overlapping 1 -second time windows (Methods). FIG. 12B) Example time series for the neural features (first row) and the extracted latent states (second row) are shown for several recording sessions in subject CP2. FIG. 12C) Predicted probability of the high pain class through the length of a recording session, averaged over sessions with high pain (red, N = 91 ), and over session with low pain (blue, N = 46). Shaded areas show the S.E.M. FIG. 12D) Extract of data shown in FIG. 12C, showing only the last time step of the recording session, which is used for final decoding. Predicted pain probability was significantly higher for the N = 91 sessions that had high pain compared to the N = 46 sessions that had low pain (***P = 3.7 x 1 O’8, one-sided Wilcoxon rank sum). The horizontal line on the box represents the median, box edges show the 25thand 75thpercentiles, whiskers represent the minimum and maximum values of the distribution. Throughout the roughly 30s recordings, the extracted latent states converge to being predictive of chronic pain. FIG. 12E) Bar plots of pain state decoding performance (AUG) for all pain metrics in all subjects using LSSM full models (all brain regions). One-sided empirical p-values based on permutation tests (n=1000, with multiple comparisons adjustment). *pFDRadj< 0.05.

[0059] FIG. 13. Chronic pain state decoding model performance for alternative pain metrics in two subjects (LDA). Bar plots of pain state decoding performance (AUG) for alternative pain metrics in two subjects using LDA full models (all brain regions). One-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment), from left to right, are CP3: 0.001 , 0.004, 0.003, 0.001 ; and CP4: 0.002, 0.001 , 0.001 , 0.001 . (see FIG. S4 and Table S2 for all values).

[0060] FIG. 14A. Chronic Subregion Models for Pain Differences (LDA, Sequential Pain Difference or “Fluctuation”) Heatmap table shows AUG and one-sided empirical p-values based on permutationtests (n=1000, without multiple comparisons adjustment) below in parenthesis for all calculated chronic pain sequential difference models (across columns) and pain score report types (across rows). AUG also represented in color based on color bar. Note this reflects classification of stable pain (no difference) vs increases or decreases (non-zero difference) in successive chronic pain scores. UNP = Unpleasantness. FIG. 14B. Chronic Subregion Models for Pain Differences (LASSO, Sequential Pain Difference or “Fluctuation”) Heatmap table shows coefficient of variation (R2) and one-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) below in parenthesis for all calculated chronic pain sequential difference models (across columns) and pain score report types (across rows). R2also represented in color based on color bar.

[0061] FIG. 15. Acute Subregion Model Performance (LDA). Heatmap table of AUG and one-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) below in parenthesis for all calculated models (across columns) and pain score report types (across rows) for acute affected (top panel) and unaffected sides (bottom panel). Red boxes outline significantly decoded pain. AUG represented in color as per color bar on right.

[0062] FIG. 16. Acute Temperature Classification - Full Model Performance (LDA). Heatmap table of AUG and one-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) below in parenthesis for full models classifying dichotomized thermal probe temperature during Acute Pain when stimulus was applied to affected side and unaffected side. Classification was significant only for CP4 on the unaffected side. AUG value corresponds to color bar on right.

[0063] FIG. 17. Cross-trained acute pain decoding models show generalization of chronic pain neural features to acute pain states in only one subject (LDA). Decoder performance for each subject when full models trained on acute thermal pain neural features and pain metrics were used to classify chronic pain data. Data for acute pain stimuli applied to the chronic pain affected (aff-ACUTE) side reached significance for one subject, while no significant cross-decoding was seen for unaffected (unaff-ACUTE) side. One-sided empirical p-values based on permutation tests (n=1000, without multiple comparisons adjustment) *p = 0.007.

[0064] FIGS. 18A-18B. Temporal Dynamics of Top 5 Power Features supporting Acute and Chronic Pain Decoding. Power timeseries plots of top 5 weighted features showing temporal dynamics of relative power increases or decreases during dichotomized high pain states for chronic (FIG. 18A) and acute-affected pain (FIG. 18B) conditions. Red curves show recording clip-averaged curves for high pain reports and blue curves show average for low pain reports (shaded area = s.e.m.). Colored square in top left corner shows feature importance of that feature taken from FIG. 2E for chronic (FIG.18A) and FIG. 3G for acute pain (FIG. 18B). Note that when the feature weight was negatively signed (blue square), average feature power was lower during high pain states and vice versa (red square = positive sign weight). Text at bottom quantifies bouts (number of times higher curve crossed above lower curve), mean bout duration (mean duration higher curve was above lower curve before passing under), and total duration above (total duration higher curve was higher than lower curve). See FIG. 5.

[0065] FIG. 19. Diurnal patterns of pain intensity NRS and neural power features are not correlated. Blue points indicate z-scored pain NRS values (top row) or feature power (indicated in top left text) organized by time-of-day of report for each subject (in columns). Red overlying line shows linear trend line of diurnal association, sampled at 3-hour resolution. See Table S4 for pain NRS trend vs feature trend correlations.

[0066] FIGS. 20A-20B. Schematic of personalized brain stimulation methods and closed loop DBS for pain. Using a temporary brain stimulation trial (FIG. 20A), patients have multiple implanted electrodes targeting pain relevant brain regions (Stage 0). If optimal stimulation and sensing targets are determined in Stage 0, patients are implanted with a sensing capable brain interface at those targets (Stage 1 ) and followed for open and closed loop algorithm optimization (Stages 2 and 3). A closed loop controller is shown using personalized biomarkers to drive a control signal to deliver therapeutic DBS for pain (FIG. 20B). The bottom panel shows pain visual analog score (VAS) reports across these stages over 3 years for an example patient with successful pain relief for refractory complex regional pain syndrome.

[0067] FIGS. 21A-21 D. Personalized pain, biomarker, decoding, pipeline, and results. (FIG. 21 A) Using various window durations of neural data prior to individual pain surveys, neural features are extracted from each window. In this example, spectral power is calculated for each contact and run through both LASSO regression (FIG. 21 B) and LDA (FIG. 21 C) machine learning outstanding accuracy. (FIG. 21 D) These models for each patient are interrogated to choose the optimal brain target at harbors, important pain signals for feature clothes loop application. P1 - 4 indicate participants 1 -4. AUG is area under the Receiver operating characteristic curve. ACC = anterior cingulate cortex, DLPFC = dorsolateral prefrontal cortex, Fop = frontal operculum., SFG = superior frontal gyrus , MFG= middle frontal gyrus, IFG = inferior frontal gyrus, S1 = primary somatosensory cortex

[0068] FIGS. 22A-22C. Temporary brain stimulation trial mapping reveals optimal, personalized brain targets for pain relief. Example pain with body diagrams at baseline and with optimal deep brain stimulation targets for three example patients: patient 1 (FIG. 22A), patient 2 (FIG. 22B), and patient3 (FIG. 22C). Individual pain numerical rating scores (NRS), visual analog scores (VAS) and pain relief VAS are shown at bottom with a list of brain targets tested for each patient (left to right shows best to worst). Black bar indicates mean, and grey bars show standard deviation.

[0069] FIG. 23. Group was data demonstrating clinically significant pain relief using closed loop and open loop brain stimulation compared to sham, (top) Active DBS significantly reduced pain shown as a percent difference from baseline NRS, compared to sham, (bottom). The absolute NRS differenes were clinically meaningful. Sham results were acquired from 2 out of 4 patients. A mixed effects regression model confirm good fit (R2>0.7) with significant effect of closed and open loop DBS (p< 0.05).

[0070] FIGS. 24A-24D. Ambulatory biomarker detection and implementation into a closed loop control DBS algorithm for chronic pain relief. (FIG. 24A) Example neural power spectra representing low pain (blue) and high pain states (red). (FIG. 24B) AUC result of linear discriminant analysis classification of high vs low pain, using a 5 Hz window of spectral power (black line). The same analysis is shown for a standard T test (red line). (FIG. 24C) The optimal spectral feature chosen from (FIG. 24B) shows good correlation with McGill Pain Questionnaire score (R>0.6) (FIG. 24D) Example functionality of closed loop deep brain stimulation operation in a patient over six days at home. The black line shows one hour moving average of percentage of time that stimulation therapy was delivered during that hour. Pain scores plotted on top correspond to the legend above. Individual decoded states on the embedded classifier are shown below. (State 0 = low pain, State 1 = high pain). State 1 entry triggers activation of the DBS therapy for a limited duration.

[0071] FIG. 25. Example, sagittal (top)and coronal (bottom) T1 MRI brain images showing personalized targeting at the Globus Pallidus (green) for pain relief.DETAILED DESCRIPTION OF THE INVENTION

[0072] Devices, systems, software, and methods are provided for treating chronic pain in an individual using deep brain stimulation (DBS). An implantable brain stimulation and neural recording device is used to identify personalized pain biomarkers to automate brain stimulation using adaptive DBS algorithms in response to pain-associated brain states. Methods are also provided for identifying effective personalized brain stimulation sites for an individual through a systematic, multisite brain mapping trial and tailoring real-time adaptive DBS programs to an individual to provide significant pain relief.

[0073] Before the present devices, systems, software, and methods are described, it is to be understood that this invention is not limited to the particular devices, systems, software, and methodsdescribed, as such may, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0074] Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit unless the context clearly dictates otherwise, between the upper and lower limits of that range is also specifically disclosed. Each smaller range between any stated value or intervening value in a stated range and any other stated or intervening value in that stated range is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range, and each range where either, neither or both limits are included in the smaller ranges is also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0075] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, some potential and preferred methods and materials are now described. All publications mentioned herein are incorporated herein by reference to disclose and describe the methods and / or materials in connection with which the publications are cited. It is understood that the present disclosure supersedes any disclosure of an incorporated publication to the extent there is a contradiction.

[0076] As will be apparent to those of skill in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has discrete components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the present invention. Any recited method can be carried out in the order of events recited or in any other order which is logically possible.

[0077] It must be noted that as used herein and in the appended claims, the singular forms “a”, “an”, and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “an electrode” or “the electrode” includes a plurality of such electrodes and reference to “an electrical signal” or “the electrical signal” includes reference to one or more electrical signals, and so forth.

[0078] It is further noted that the claims may be drafted to exclude any element which may be optional. As such, this statement is intended to serve as antecedent basis for use of such exclusiveterminology as “solely”, “only” and the like in connection with the recitation of claim elements, or the use of a “negative” limitation.

[0079] The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Nothing herein is to be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. Further, the dates of publication provided may be different from the actual publication dates which may need to be independently confirmed.Definitions

[0080] The term “about,” particularly in reference to a given quantity, is meant to encompass deviations of plus or minus five percent.

[0081] The terms “individual”, “subject”, “recipient”, and “patient” are used interchangeably herein and refer to any mammalian subject for whom diagnosis, treatment, or therapy is desired, particularly humans. "Mammal" for purposes of treatment refers to any animal classified as a mammal, including human and non-human mammals such as non-human primates, including chimpanzees and other apes and monkey species; laboratory animals such as mice, rats, rabbits, hamsters, guinea pigs, and chinchillas; domestic animals such as dogs and cats; and farm animals such as sheep, goats, pigs, horses and cows.

[0082] The term “user” as used herein refers to a person that interacts with a device and / or system disclosed herein for performing one or more steps of the presently disclosed methods. The user may be a patient being diagnosed or receiving treatment for chronic pain. The user may be a health care practitioner, such as the patient’s physician.

[0083] The term “pain-associated disorder” is used herein to refer to conditions that cause pain in a person suffering from the disorder. The pain may be chronic, idiopathic, nociceptive, inflammatory, myofascial, visceral, or neuropathic pain. Pain-associated disorders include, without limitation, poststroke pain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-related pain (e.g., such as caused by diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy), peripheral vascular disease, pain after an amputation (phantom limb pain), visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, or chronic leg pain (e.g., sciatica), neck pain, foot pain, or arm pain.

[0084] By “treatment” or “treating” is meant that at least an amelioration of pain associated with the condition afflicting the subject is achieved such that the patient has a desired or beneficial clinicalresult, where amelioration refers to at least a reduction in the magnitude of a parameter, e.g., pain associated with the condition being treated. As such, treatment includes a broad spectrum of situations ranging from lessening intensity, duration or extent of pain caused by a condition and / or correlated with a condition, up to and including completely eliminating the pain, along with any associated symptoms. Treatment therefore includes situations where the condition, or at least pain associated therewith, is completely inhibited, e.g., prevented from happening, or stopped, e.g., terminated, such that the subject no longer suffers from the condition, or at least the pain symptoms that characterize the condition. Treatment also includes situations where the progression of the condition, or at least the progression of a symptom associated therewith, is slowed, delayed, or halted. In such cases, a subject might still have residual symptoms associated with the pathological condition, but any increase in the severity or magnitude of the symptoms is slowed, delayed, or prevented.Methods

[0085] The present disclosure provides methods for ameliorating pain using deep brain stimulation. In particular, deep brain stimulation is performed with a neural recording device that records electrical signals from neural activity associated with chronic pain. Machine learning computational models are used to detect and classify patterns of neural activity associated with various pain states and pain relief. The neural signatures of “chronic pain” and “pain relief” are used to assist with deep brain stimulation programming to determine therapeutic stimulation parameters that achieve analgesia. The methods and systems can be used in performing open-loop therapy to provide clinical guidance to clinicians or technicians for adjusting deep brain stimulation programming. Methods and systems are also provided for performing closed-loop therapy with a neurostimulator that records brain electrical signals from neural activity associated with pain and automatically adjusts the neurostimulator settings and / or delivers electrical stimulation to the brain when pre-specified patterns of neural activity associated with pain are detected. Such closed-loop methods may also include automated provision of clinical guidance for stimulation programming or direct computer-to-computer implementation of programming.

[0086] The subject methods can be used to treat any condition causing pain, including chronic, idiopathic, nociceptive, inflammatory, myofascial, visceral, or neuropathic pain. Pain-associated disorders which may be treated with deep brain stimulation, as described herein, include, without limitation, post-stroke pain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-relatedpain (e.g., such as caused by diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy), peripheral vascular disease, pain after an amputation (phantom limb pain), visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, and chronic leg pain (e.g., sciatica), neck pain, foot pain, or arm pain

[0087] The method includes positioning a first electrode to deliver electrical stimulation to a target stimulation site in the brain (i.e., DBS electrode) and positioning a second electrode at a second location in a sensorimotor cortex region of the brain of the subject to detect brain electrical signals from neural activity associated with pain (i.e., detection electrode). In some embodiments, one or more DBS electrodes are positioned to deliver electrical stimulation to one or more target stimulation sites in the brain, and one or more detection electrodes are positioned at one or more locations in the sensorimotor cortex region of the brain of the subject to detect brain electrical signals from neural activity associated with pain. The DBS electrodes and the detection electrodes may be non-brain penetrating surface electrodes or brain-penetrating depth electrodes. In some cases, the DBS electrodes and the detection electrodes are extracranial electrodes, for example, subgaleal or skull mounted (in burrhole cap or cranially mounted neurostimulator) electrodes. The electrical stimulation may be applied to the brain using a DBS electrode in a manner effective for ameliorating pain in a subject having a pain-associated disorder when a brain electrical signal is detected from the sensorimotor cortex region of the brain using a detection electrode that has a signal amplitude exceeding a threshold level indicating that the subject has a level of pain severity in need of treatment.

[0088] In certain embodiments, one or more detection electrodes are used to record brain electrical signals for neural activity associated with pain or pain relief in one or more locations of the sensorimotor cortex. A detection electrode may be placed, for example, in an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region, or in other regions of the brain suitable for detection. In certain embodiments, a detection electrode is placed in a right OFC region, a left OFC region a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof. In some embodiments, the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain. In some embodiments, the detection electrode is positioned at a location to record brain electrical signals from a contralateral OFC region. In certain embodiments, the recorded brain electrical signal data comprises field potential data. The site chosen for detection may differ for different subjects and may depend on mapping of the brain of an individual subject to identify the optimal location for positioningan electrode for detecting brain electrical signals from neural activity associated with pain or pain relief. In some embodiments, mapping comprises identifying optimal detection sites in each cerebral hemisphere.

[0089] As used herein, the phrases “an electrode” or “the electrode” refer to a single electrode or multiple electrodes such as an electrode array. As used herein, the term “contact” as used in the context of an electrode in contact with a region of the brain refers to a physical association between the electrode and the region. In other words, a detection electrode that is in contact with a region of the brain is physically touching the region of the brain. A DBS electrode can conduct electricity to specific targets in the brain. Electrodes used in the methods disclosed herein may be monopolar (cathode or anode) or bipolar (e.g., having an anode and a cathode).

[0090] Positioning a detection electrode for recording neural activity at specified region(s) of the brain may be carried out using standard surgical procedures for placement of intra-cranial electrodes. In certain cases, placing the detection electrode may involve positioning the electrode on the surface of the specified region(s) of the brain. For example, electrodes may be placed on the surface of the brain at an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region, or any combination thereof. The electrode may contact at least a portion of the surface of the brain at the OFC, ACC, insula, vmPFC, BG, or CM region. In some embodiments, the electrode may contact substantially the entire surface area at OFC, ACC, insula, vmPFC, BG, or CM region. In some embodiments, the electrode may additionally contact area(s) adjacent to the OFC, ACC, insula, vmPFC, BG, or CM regions. In some embodiments, an electrode array, arranged on a planar support substrate, may be used for detecting brain electrical signals for neural activity from one or more of the brain regions specified herein. The surface area of the electrode array may be determined by the desired area of contact between the electrode array and the brain. An electrode for implanting on a brain surface, such as, a surface electrode or a surface electrode array may be obtained from a commercial supplier. A commercially obtained electrode / electrode array may be modified to achieve a desired contact area. In some cases, the non-brain penetrating electrode (also referred to as a surface electrode) that may be used in the methods disclosed herein may be a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode, or an electrocorticogram (ECoG) electrode. In some cases, an electrode is subgaleal mounted, burrhole mounted, or cranially mounted. In certain embodiments, a plurality of electrodes is positioned at one or more of the brain regions specified herein for detection of electroencephalographic signals by stereoelectroencephalography (sEEG).

[0091] In certain cases, placing the detection electrode at a target area or site (e.g., OFC, ACC, insula, vmPFC, BG, or CM) may involve positioning a brain penetrating electrode (also referred to as depth electrode) in the specified region(s) of the brain. For example, a brain penetrating detection electrode may be placed in an OFC, ACC, insula, vmPFC, BG, or CM region. In some embodiments, the detection electrode may additionally contact area(s) adjacent to the OFC, ACC, insula, vmPFC, BG, or CM regions. In some embodiments, a brain penetrating electrode array may be used for detecting neural activity from the OFC, ACC, insula, vmPFC, BG, or CM, or a combination thereof, as specified herein.

[0092] The depth to which a detection electrode is inserted into the brain may be determined by the desired level of contact between the electrode array and the brain. A brain-penetrating electrode array may be obtained from a commercial supplier. A commercially obtained electrode array may be modified to achieve a desired depth of insertion into the brain tissue.

[0093] Positioning an electrode for delivering electrical stimulation to a target stimulation site in the brain may be carried out using standard surgical procedures for placement of electrodes for deep brain stimulation. For example, a DBS electrode may be placed in an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region, or other intracranial region. Medical imaging using, for example, magnetic resonance imaging (MRI) or computerized tomography (CT) may be used to provide guidance for placement of DBS electrodes and verify correct placement of the DBS electrodes in the brain. In addition, a neurostimulator that generates electrical pulses may be placed under the skin of the chest, typically below the collarbone or in the abdomen. In some embodiments the neurostimulator is cranially mounted. The surgical procedure may involve placing DBS electrodes within the brain through small holes in the skull. An electrode lead is tunneled under the skin down the neck and under the skin of the chest to connect to a chest implanted neurostimulator.

[0094] Current is supplied by the neurostimulator to the DBS electrodes. Parameters such as pulse width, shape, frequency, amplitude, pattern, and temporal distribution can be adjusted in response to changes in neural activity in the sensorimotor cortex region of the brain to treat pain. In some embodiments, a closed loop system is used to adjust DBS settings automatically in response to changes in neural activity in a sensorimotor cortex region of the brain. In other embodiments, an open loop system is used in which DBS settings are adjusted by a user or medical practitioner basedon the neural activity in the sensorimotor cortex region of the brain. In some cases, infrequent intermittent stimulation may produce prolonged benefits outlasting electrical stimulation, which may reduce the need for closed loop therapy using adaptive algorithms.

[0095] The electrical stimulation may be applied using a single electrode, electrode pairs, or an electrode array. In some embodiments, the number of electrodes used to deliver electrical stimulation to the brain ranges from 8 to 32, including any number of electrodes in this range such as 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, or 32 electrodes. In some embodiments, the electrical stimulation is applied to more than one target stimulation site. The site to which the electrical stimulation is applied may be alternated or otherwise spatially or temporally patterned. Electrical stimulation may be applied to the sites simultaneously or sequentially. In certain embodiments, the region of the brain to which electrical stimulation is applied is an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region, or other regions of the brain suitable for stimulation. The site chosen for stimulation may differ for different subjects and will depend on mapping of the brain of an individual subject to identify the optimal location for positioning an electrode for delivery of electrical stimulation to treat pain. In some embodiments, mapping comprises identifying optimal target stimulation sites in each cerebral hemisphere.

[0096] In some embodiments, an electrode array arranged on a planar support substrate may be used for electrically stimulating the basal ganglia. The surface area of the electrode array may be determined by the desired area of contact between the electrode array and the basal ganglia. In some cases, cylindrical electrode arrays, paddle-style electrode arrays, or plate-style electrode arrays may be used in the methods disclosed herein for deep brain stimulation. Such DBS electrode arrays for implanting in the brain, may be obtained from a commercial supplier. A commercially obtained electrode / electrode array may be modified to achieve a desired contact area.

[0097] The precise number of DBS electrodes or detection electrodes contained in an electrode array (e.g., for electrical stimulation or detection of neural activity) may vary. In certain aspects, an electrode array may include two or more electrodes, such as 3 or more, including 4 or more, e.g., about 3 to 6 electrodes, about 6 to 12 electrodes, about 12 to 18 electrodes, about 18 to 24 electrodes, about 24 to 30 electrodes, about 30 to 48 electrodes, about 48 to 72 electrodes, about 72 to 96 electrodes, or about 96 or more electrodes. The electrodes may be arranged into a regularrepeating pattern (e.g., a grid, such as a grid with about 1 cm spacing between electrodes), or no pattern. An electrode that conforms to the target site for optimal delivery of electrical stimulation may be used. One such example, is a single multi contact electrode with eight contacts separated by 21 / a mm. Each contract would have a span of approximately 2 mm. Another example is an electrode with two 1 cm contacts with a 2 mm intervening gap. Yet further, another example of an electrode that can be used in the present methods is a 2 or 3 branched electrode to cover the target site. Each one of these three-pronged electrodes has four 1 -2 mm contacts with a center to center separation of 2 of 2.5 mm and a span of 1 .5 mm.

[0098] The size of each electrode may also vary depending upon such factors as the number of electrodes in the array, the location of the electrodes, the material, the age of the patient, and other factors. In certain aspects, an electrode array has a size (e.g., a diameter) of about 5 mm or less, such as about 4 mm or less, including 4 mm-0.25 mm, 3 mm-0.25 mm, 2 mm-0.25 mm, 1 mm-0.25 mm, or about 3 mm, about 2 mm, about 1 mm, about 0.5 mm, or about 0.25 mm.

[0099] In certain embodiments, the method further comprises mapping the brain of the subject to optimize positioning of an electrode for applying electrical stimulation. Positioning of a DBS electrode is optimized to maximize clinical responses to electrical stimulation to treat pain. In some embodiments, an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region, or other regions of the brain are mapped to determine optimal positioning of DBS electrodes.

[0100] Assessment of the effectiveness of electrical stimulation at a particular site for relieving pain may be performed using any standard method. In certain cases, an interviewer-administered pain assessment may be used alone or in conjunction with a self-reporting tool. Interviewer-administered pain assessments may use a visual analog scale (VAS), a pain intensity numerical rating score (NRS), or a verbal rating scale. In some cases, a pain unpleasantness VAS (unpVAS), a pain relief VAS (reliefVAS), or a short-form McGill Pain Questionnaire (SF-MPQ) is used to rate pain unpleasantness (NRS and VAS). Amelioration of pain may include a reduction in the level of severity of pain compared to the level of pain prior to deep brain stimulation. In some embodiments, amelioration of pain includes at least about a 50% to at least about a 90% reduction in the level of pain compared to the level of pain prior to the deep brain stimulation. In some embodiments, amelioration of pain includes a 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 100%reduction in the level of pain, or any percentage reduction in the level of pain within this range, compared to the level of pain prior to the deep brain stimulation.

[0101] In certain embodiments, the method further comprises mapping the brain of the subject to optimize positioning of a detection electrode. Positioning of the detection electrode is optimized to detect brain activity features that indicate that pain is severe enough to need treatment. For example, the levels of overall power, average power, or power in specific frequency ranges (e.g., alpha, delta, beta, gamma, and / or theta) may be correlated with pain severity and need for treatment with electrical stimulation. Thus, detection electrodes may be positioned to optimize detection of brain activity in specific frequency ranges that correlate with pain severity of a pain-associated disorder. Alternatively or additionally, coherence within certain spectral frequency bands or other features of network connectivity may be correlated with pain severity and need for treatment with DBS.

[0102] Detection of brain activity may be performed by any method known in the art. For example, functional brain imaging of neural activity may be carried out by electrical methods such as electroencephalography (EEG), stereoelectroencephalography (sEEG), magnetoencephalography (MEG), single photon emission computed tomography (SPECT), as well as metabolic and blood flow studies such as functional magnetic resonance imaging (fMRI), and positron emission tomography (PET). In some embodiments, the right frontal lobe, left frontal cortex, or other regions are mapped to determine optimal positioning for detection electrodes. One or more of these regions may be implanted with detection electrodes to measure electrical signals from neural activity associated with pain or relief of pain for a subject having a pain-associated disorder.

[0103] As set forth here, the subject methods involve applying electrical stimulation to the brain in a manner effective to treat pain caused by a pain-associated disorder in a subject when neural activity associated with pain is detected. Closed-loop therapy can be performed with DBS used in combination with a neural recording device that records brain electrical signals from neural activity associated with pain, wherein electrical stimulation is delivered to the brain when the level of a brain electrical signal that is detected exceeds a set threshold level indicating a level of pain severity in need of treatment. In certain embodiments, brain electrical signals are analyzed using a binary pain classification model that classifies brain electrical signals as indicating that the subject is in either a high pain severity state or a low pain severity state, wherein electrical stimulation is applied to the target stimulation site(s) in the brain of the subject in a manner effective to treat the chronic pain in the subject when a brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state. In some embodiments, the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classifybrain electrical signals as indicating that the subject is in either a high pain severity state or a low pain severity state, as described further below (see, e.g., Example 1 ).[OO1 O4] The parameters for applying the electrical stimulation to the brain may be determined empirically during treatment or may be pre-defined, such as, from a trial study with a subject. For example, brain electrical signal data from the sensorimotor cortex is recorded (e.g., from the OFC, ACC, insula, vmPFC, BG, or a CM, or a combination thereof) with varying stimulation settings, including baseline (stimulation off), optimal therapeutic stimulation, modified and ineffective stimulation, and maximum tolerated stimulation to identify personal neural signatures of “chronic pain” and “pain relief” for a patient, which are used to assist with deep brain stimulation programming to determine optimal therapeutic stimulation parameters for treatment of chronic pain. The parameters of the electrical stimulation may include one or more of frequency, pulse width / duration, duty cycle, intensity / amplitude, pulse pattern, program duration, program frequency, and the like.

[0105] Frequency refers to the pulses produced per second during stimulation and is stated in units of Hertz (Hz, e.g., 60 Hz = 60 pulses per second). The frequencies of electrical stimulation used in the present methods may vary widely depending on numerous factors and may be determined empirically during treatment of the subject or may be pre-defined. In certain embodiments, the method may involve applying electrical stimulation to the brain at a frequency of 2 Hz - 250 Hz, such as, 10 Hz - 250 Hz, 10 Hz - 100 Hz, 20 Hz - 100 Hz, 50 Hz - 100 Hz, 25 Hz - 200 Hz, 50 Hz - 250 Hz, 50 Hz -185 Hz, 50 Hz -150 Hz, 75 Hz - 200 Hz, 100 Hz - 200 Hz, 100 Hz - 180 Hz, 100 Hz - 160 Hz, or 130 Hz - 150 Hz. In some embodiments, the electrical stimulation to the brain is applied at a frequency of about 10 Hz to about 100 Hz, including any pulse frequency within this range such as 10 Hz, 15 Hz, 20 Hz, 25 Hz, 30 Hz, 35 Hz, 40 Hz, 45 Hz, 50 Hz, 55 Hz, 60 Hz, 65 Hz, 70 Hz, 75 Hz, 80 Hz, 85 Hz, 90 Hz, 95 Hz, or 100 Hz. In some embodiments, non-integer pulse frequencies are used (e.g., 50.2 Hz, 50.4 Hz, etc.).

[0106] The electrical stimulation may be applied in pulses such as a uniphasic or a biphasic pulse. The time span of a single pulse is referred to as the pulse width or pulse duration. The pulse width used in the present methods may vary widely depending on numerous factors (e.g., severity of the disease, status of the patient, and the like) and may be determined empirically or may be pre-defined. In certain embodiments, the method may involve applying an electrical stimulation at a pulse width of about 10 psec - 500 psec, for example, 50 psec - 450 psec, 100 psec - 300 psec, 100 psec - 250 psec, 100 psec - 200 psec, or 100 psec - 150 psec. In some embodiments, the electrical stimulation to the brain is applied at a pulse width of about 100 psec to about 300 psec, including any pulse width within this range such as 100 psec, 1 10 psec, 120 psec, 130 psec, 140 psec,150 | sec, 160 p.sec, 170 |isec, 180 p.sec, 190 p.sec, 200 .sec, 210 p.sec, 220 jisec, 230 p.sec, 240 | sec, 250 |j.sec, 260 psec, 270 psec, 280 psec, 290 psec, or 300 psec.

[0107] The electrical stimulation may be applied for a stimulation period of 0.1 sec-1 month, with periods of rest (i.e., no electrical stimulation) possible in between. In certain cases, the period of electrical stimulation may be 0.1 sec-1 week, 1 sec-1 day, 10 sec-12 hours, 1 min-6 hours, 10 min- 1 hour, and so forth. In certain cases, the period of electrical stimulation may be 1 sec-1 min, 1 sec- 30 sec, 1 sec-15 sec, 1 sec-10 sec, 1 sec-6 sec, 1 sec-3 sec, 1 sec-2 sec, or 6 sec-10 sec. The period of rest in between each stimulation period may be 60 sec or less, 30 sec or less, 20 sec or less, or 10 sec. In some embodiments, electrical stimulation may be applied for a year or more, 2 years or more, 3 years or more, 5 years or more, or 10 years or more. In some embodiments, electrical stimulation may be continued indefinitely as part of a long-term DBS therapy regimen.

[0108] The electrical stimulation may be applied with an amplitude of current of 0.1 mA - 30 mA, such as, 0.1 mA - 25 mA, such as, 0.1 mA - 20 mA, 0.1 mA - 15 mA, 0.1 mA - 10 mA, 0.1 mA - 4 mA, 0.1 mA - 3 mA, 0.1 mA - 2 mA, 0.1 mA - 1 mA, or 0.1 mA - 0.5 mA. In some embodiments, the amplitude of current is 0.1 mA - 4 mA, or any amplitude of current in this range such as 0.1 mA, 0.2 mA, 0.3 mA, 0.4 mA, 0.5 mA, 0.6 mA, 0.7 mA, 0.8 mA, 0.9 mA, 1 .0 mA, 1.1 mA, 1 .2 mA, 1 .3 mA, 1 .4 mA, 1 .5 mA, 1 .6 mA, 1 .7 mA, 1 .8 mA. 1 .9 mA, 2.0 mA, 2.1 mA, 2.2 mA, 2.3 mA, 2.4 mA, 2.5 mA, 2.6 mA, 2.7 mA, 2.8 mA, 2.9 mA, 3.0 mA, 3.1 mA, 3.2 mA, 3.3 mA, 3.4 mA, 3.5 mA, 3.6 mA, 3.7 mA, 3.8 mA, 3.9 mA, or 4.0 mA.

[0109] The electrical stimulation may be applied with an amplitude of voltage of 0.1 V-15 V, such as, 0.1 V-10 V, 0.1 V-5 V, 1 V-10 V, 1 V-5, V, or 1 V-3.5 V. In some embodiments, the amplitude of voltage is 1 V-3.5 V, or any amplitude of voltage in this range such as 1 V, 1 .1 V, 1 .2 V, 1 .3 V, 1 .4 V, 1 .5 V, 1 .6 V, 1 .7 V, 1 .8 V, 1 .9 V, 2.0 V, 2.1 V, 2.2 V, 2.3 V, 2.4 V, 2.5 V, 2.6 V, 2.7 V, 2.8 V, 2.9 V, 3.0 V, 3.1 V, 3.2 V, 3.3 V, 3.4 V, or 3.5 V.

[0110] The electrical stimulation having the parameters as set forth above may be applied over a program duration of around 1 day or less, such as, 18 hours, 6 hours, 3 hours, 2 hours, 1 hour, 45 minutes, 30 minutes, 20 minutes, 10 minutes, or 5 minutes, or less, e.g., 30 seconds - 300 seconds, 1 minute - 5 minutes, 2 minutes - 10 minutes, 2 minutes - 20 minutes, 2 minutes - 30 minutes, 5 minutes - 10 minutes, 5 minutes - 30 minutes, or 5 minutes - 15 minutes, 10 minutes - 400 minutes, 25 minutes - 300 minutes, 50 minutes - 200 minutes, or 75 minutes - 150 minutes, which period would include the application of pulses and the intervening rest period. The program may be repeated at a desired program frequency to relieve chronic pain in the subject. As such, a treatment regimen may include a program for electrical stimulation at a desired program frequency andprogram duration. In some embodiments, the treatment regimen is controlled by a control unit in communication with a pulse generator connected to the one or more DBS electrodes in a closed- loop treatment regimen.

[0111] The electrical stimulation having the parameters as set forth above may be applied over a program duration of around 1 day or less, such as, 18 hours, 6 hours, 3 hours, 1 hour, 45 minutes, 30 minutes, 20 minutes, 10 minutes, or 5 minutes, or less, e.g., 1 minute- 5 minutes, 2 minutes - 10 minutes, 2 minutes - 20 minutes, 2 minutes - 30 minutes, 5 minutes - 10 minutes, 5 minutes - 30 minutes, or 5 minutes - 15 minutes which period would include the application of pulses and the intervening rest period. The program may be repeated at a desired program frequency to relieve pain in the subject. As such, a treatment regimen may include a program for electrical stimulation at a desired program frequency and program duration. In some embodiments, the treatment regimen is controlled by a control unit in communication with a pulse generator connected to the one or more DBS electrodes in a closed-loop treatment regimen.

[0112] In some embodiments, a cap on the maximum number of electrical stimulations per day can be set. For example, the maximum number of electrical stimulations per day may range from 50 therapies per day to 500 therapies per day, including any number of therapies per day in this range such as 50, 75, 100, 125, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, or 500 therapies per day. Alternatively or addionally a cap can be set on the total amount of time of electrical stimulation per day. For example, the total amount of time of electrical stimulation per day may range from 10 minutes to 100 minutes of total stimulation per day, including any amount of time within this range such as 10, 20, 30, 40, 50, 60, 70, 80, 90, or 100 minutes of total stimulation time per day. Such caps on maximum number of electrical stimulations and / or total amount of time of electrical stimulation per day can be employed so as not to disturb the sleep of a patient from evening therapy. In addition, a rest period without electrical stimulation when a patient wishes to sleep may also be employed to avoid interfering with sleep.

[0113] As noted above, the treatment may ameliorate pain of a pain-associated disorder suffered by the subject. Assessment of effectiveness of the treatment may be performed using any known method for evaluating pain. In certain cases, an interviewer-administered pain assessment may be used alone or in conjunction with a self-reporting tool. Interviewer-administered pain assessments may use a visual analog scale (VAS), a pain intensity numerical rating score (NRS), or a verbal rating scale. Amelioration of pain may include a reduction in the level of severity of pain compared to the level of pain prior to deep brain stimulation.

[0114] In certain cases, effectiveness of treatment may be assessed by detecting brain activity (e.g., electrical signals) associated with pain or pain relief, which may be within an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC), a basal ganglia (BG) region, or a centromedian thalamus (CM) region, or another area. For example, the brain region(s) may be one or more regions selected from a right OFC region, a left OFC region a right ACC region, a left ACC region, a right VMPFC region, and a left VMPFC region. Detection of brain activity may be performed by functional brain imaging. Functional brain imaging may be carried out by electrical methods such as electroencephalography (EEG), magnetoencephalography (MEG), single photon emission computed tomography (SPECT), as well as metabolic and blood flow studies such as functional magnetic resonance imaging (fMRI), and positron emission tomography (PET). In some embodiments, electrical methods for assessing effectiveness of treatment may involve use of a detection electrode as described herein or placement of an additional electrode for measuring electrical signals at a secondary region of the brain. One or more regions of the brain may be implanted with an electrode and electrical signals measured for assessment of effectiveness of the treatment. Any suitable electrodes may be used for measurements and may include one or more surface electrodes (non-brain penetrating electrode(s)) or one or more depth electrodes (brain penetrating electrode(s)) as described herein.

[0115] Assessment of effectiveness of treatment and assessment of amelioration of pain of a pain- associated disorder may be performed at any suitable time point after commencement of the treatment procedure, for example, during open-loop or closed-loop therapy or after a treatment regimen is complete. Embodiments of the subject methods include assessing effectiveness of treatment or amelioration of pain from the pain-associated disorder within seconds, minutes, hours, or days after the initial treatment regimen has been completed. In some instances, assessment may be performed at multiple time points. In some cases, more than one type of assessment may be performed at the different time points. In some embodiments, a subject’s brain activity (e.g., at the OFC, ACC, insula, vmPFC, BG, or CM) may be measured prior to the application of electrical stimulation, and assessing may include comparing the subject’s brain activity after the treatment to that before the treatment and a change in the post-treatment brain activity may indicate successful treatment.

[0116] In some embodiments, the subject is monitored for a period of time during open-loop or closed-loop therapy using a detection electrode, wherein increasing numbers of the brain electrical signals exceeding the threshold level detected by the detection electrode within a later recording period compared to an earlier recording period indicate increasing severity of pain from the pain-associated disorder; and decreasing numbers of brain electrical signals exceeding the threshold level detected by the detection electrode within a later recording period compared to an earlier recording period indicate decreasing severity of the pain of the pain-associated disorder.

[0117] The methods and systems provided herein may be used to ameliorate pain in a patient. A person skilled in the art will appreciate that amelioration of pain may provide relief to a patient suffering from two separate pain-associated disorders.

[0118] Upon completion of a treatment regimen, the patient may be assessed for effectiveness of the treatment and the treatment regimen may be repeated, if needed. In certain cases, the treatment regimen may be altered before repeating. For example, one or more of the frequency, pulse width, current amplitude, period of electrical stimulation, program duration, program frequency, and / or placement of DBS or detection electrodes may be altered before starting a second treatment regimen.

[0119] Application of the method may include a prior step of selecting a patient for treatment based on need as determined by clinical assessment, which may include assessment of severity of chronic pain (e.g., pain lasting at least 3 months), physical condition, medication regime, pain history, cognitive assessment, anatomical assessment, behavioral assessment and / or neurophysiological assessment. In certain cases, a subject may be further assessed to determine if deep brain stimulation will completely or partially (e.g., at least 50%) relieve the chronic pain. Such a patient may undergo DBS on a temporary trial basis to determine if DBS decreases the severity of pain experienced by the patient. Such a patient may also be implanted with detection electrodes to identify personalized pain and pain relief neural signatures of “chronic pain” and “pain relief” to assist with deep brain stimulation programming to determine therapeutic stimulation parameters for the patient and / or evaluate whether DBS therapy will be effective for ameliorating pain for the patient.

[0120] In certain aspects, the methods and systems of the present disclosure may include measurement of brain activity in a sensorimotor cortex region, for example, electrical activity in an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region, where the level of alpha frequency (8 Hz to 12 Hz), beta frequency (12 Hz to 30 Hz), gamma frequency (low gamma at 30 Hz to 70 Hz or high gamma at 70 Hz to 100 Hz), delta frequency (1 Hz to 4 Hz), or theta frequency (4 Hz to 8 Hz) power may be measured. In certain cases, electrical activity from a plurality of locations in the OFC, ACC, insula, vmPFC, BG, or CM may be measured and averaged. In some embodiments, electrical activity in the delta frequency range (such as 1 Hz to 4 Hz) may be measured and averaged. In some cases, electrical activity inone or more locations in the brain may be measured during a period extending from prior to stimulation to the period during which stimulation to the brain is applied, or to a period after stimulation to the brain has been applied, and monitored using a binary classification model to determine if the subject is in a high pain severity state in need of further electrical stimulation or a low pain severity state that does not require electrical stimulation. In some cases, when the subject is determined to be in a low pain severity state using the binary classification model, the methods and systems do not apply a further stimulation to the brain. Alternatively, when analysis of the brain electrical signals from the sensorimotor cortex using the binary classification model indicates the subject is transitioning to a high pain severity state or in a high pain severity state, the methods and systems may apply a further stimulation to the brain.

[0121] A closed-loop method allows determination of parameters of electrical stimulation based upon real-time feedback signals from the brain of the subject. Closed-loop methods and systems allow for automation of treatment of the subject including real-time need-based modulation of the treatment regimen. Exemplary closed-loop methods and associated systems for treatment of chronic pain are further discussed in the Examples section and are depicted in FIGS. 2 and 6. Closed-loop methods and systems for automated delivery of electrical stimulation are further described below.Closed-Loop Method for Automated Delivery of Electrical Stimulation

[0122] In certain embodiments, a control algorithm is used to automate the delivery of electrical stimulation to the brain in response to detection of neural activity associated with pain. According to certain embodiments, the method may include receiving an electrical signal from a sensorimotor cortex region (e.g., an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region) of the brain of the subject via a detection electrode; applying electrical signal metrics to a control algorithm that is tuned to a clinically relevant target (e.g., a range of signal indicative of effective treatment and / or a range of signal indicative of severity of pain and the need for treatment); automatically delivering electrical stimulation to the brain via a DBS electrode in a manner effective to treat the pain if the electrical signal metrics indicate that the patient is in need of treatment. For example, brain electrical activity from a sensorimotor cortex region may be measured with a detection electrode, wherein the control algorithm receives the electrical activity data from the detection electrode, analyzes the brain electrical activity using a binary pain classification model that classifies the brain electrical activity as indicating that the subject is in either a high pain severity state or a low pain severity state (see, e.g., Example 1), and automates deliveryof electrical stimulation via a DBS electrode to the brain when a brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state. In some embodiments, one or more programmed stimulation parameters are modulated according to the algorithm’s control law based on the recorded electrical activity data; and modulated electrical stimulation is delivered to the brain via the DBS electrode in a manner effective to ameliorate the pain.

[0123] As described in the foregoing sections, effectiveness of treatment or amelioration of pain from a pain-associated disorder may be assessed by detecting brain electrical activity associated with pain using a detection electrode. In an open-loop system, stimulation is delivered in a preprogrammed way or manually by a user but is not automatically controlled by real-time neural feedback from the patient’s brain. The electrical activity may be analyzed by a computing means which may output recommendations based on comparing the electrical activity to a predetermined range (e.g., range of brain electrical activity corresponding to high pain severity or low pain severity). A user may then carry out the recommendations, such as, changing a parameter of the electrical stimulation program prior to starting another treatment regimen. In a closed-loop system, by contrast, a computing means can automatically update stimulation parameters based upon analysis of the recorded electrical signal and / or automatically deliver stimulation to the brain according to the electrical stimulation program. In some embodiments, either an open-loop or a closed-loop system may be integrated with a mechanism for user intervention, for example by allowing user-override of open-loop or closed-loop stimulation programs to enact or prevent stimulation that would ordinarily occur, or to manually change parameters of such stimulation.

[0124] In some embodiments, the computing means for directing closed-loop stimulation may be a combination of hardware / software which may be connected wirelessly or by wire to the measurement electrodes. The computing means may communicate with a control unit (also referred to as a control module) that controls a pulse generator connected to the DBS electrodes. In certain embodiments, the computing means may be connected to a recorder (e.g., a neurophysiological recorder or neural recording device) that records brain activity measured by the detection electrodes. The computing means may include a control algorithm that determines modification of stimulation parameters based on real-time outputs of the neurophysiological recorder. The algorithm may operate by simple on / off control of stimulation at set parameters, modifying only the on / off parameter with each evaluation cycle, or may determine sophisticated modification of a range of stimulation parameters with each cycle. In some cases, the algorithm may be based on information related to the pain-associated disorder, such as, a range of electrical activity that is indicative of the presence of pain. The algorithmmay also include additional information such as a brain activity profile of a normal subject (not suffering from pain). Regardless of the particular control algorithm structure, the computing means may be tuned to a clinically relevant target (e.g., a range of signal indicative of effective treatment and / or a range of signal indicative of severity of pain and the need for treatment) that directs modulation of one or more programmed stimulation parameters according to the algorithm’s control law, applying the modulated electrical stimulation to the brain via the DBS electrode.

[0125] In some cases, the computing means, via a control algorithm, may determine whether the received electrical signals are within or outside a predetermined range of neural signals indicative of the presence of pain of high severity or low severity, as classified according to a binary classification model. When the received electrical signals are outside this predetermined range, then the computing means determines that the pain has been treated or ameliorated. The computing means may then communicate with the control unit to direct stimulation shut-off by the pulse generator. When the received electrical signals are within the predetermined range of neural signals indicative of the presence of the pain, then the computing means determines that the pain is in need of treatment and / or severe enough to require treatment. The control algorithm within the computing means may then determine whether the initial step of applying electrical stimulation to the brain should be repeated and / or whether a parameter of the electrical stimulation should be modified prior to the step of applying electrical stimulation. The computing means, via the control unit, may then communicate with the control unit to provide the appropriate instructions to the pulse generator.

[0126] In some embodiments, the computing means may determine whether the received electrical signals are within or outside a second predetermined range, where the second predetermined range is indicative of treatment and / or amelioration of pain from the pains-associated disorder. When the received electrical signals are within the second predetermined range, then the computing means determines that the pain has been treated and / or ameliorated. The computing means may then communicate with the control unit to direct stimulation switch-off by the pulse generator. When the received electrical signals are outside the second predetermined range, then the computing means determines that the pain still needs to be treated and / or ameliorated. The control algorithm within the computing means may then determine whether the initial step of applying electrical stimulation should be repeated and / or whether a parameter of the electrical stimulation modified prior to the step of applying electrical stimulation. The processor may then communicate with the control unit to provide the appropriate instructions to the pulse generator.

[0127] Thus, in certain aspects, the subject methods operate as a closed-loop control system which may automatically adjust one or more parameters in response to electrical activity from a region ofthe brain of a subject and / or automatically deliver stimulation to the brain according to the electrical stimulation program. In some embodiments, the closed-loop control system automatically delivers stimulation according to set parameters when the received electrical signals are within a predetermined range indicative of the need for treatment. Exemplary closed-loop methods and associated systems are described in the Examples section of the application and are illustrated in FIGS. 2 and 6.

[0128] In certain embodiments, the level of alpha frequency (8 Hz to 12 Hz), beta frequency (12 Hz to 30 Hz), gamma frequency (low gamma at 30 Hz to 70 Hz or high gamma at 70 Hz to 100 Hz), delta frequency (1 Hz to 4 Hz), or theta frequency (4 Hz to 8 Hz) power may be measured. In certain cases, electrical activity from a plurality of locations in the orbitofrontal cortex (OFC), anterior cingulate cortex (ACC), insula region, ventromedial prefrontal cortex (vmPFC), basal ganglia (BG), or centromedian thalamus (CM) may be measured and averaged. In some embodiments, electrical activity in the delta frequency range (such as 1 Hz to 4 Hz) may be measured and averaged. In some cases, electrical activity in one or more locations in the brain may be measured during a period extending from prior to stimulation to the period during which stimulation to the brain is applied, or to a period after stimulation to the brain has been applied, and monitored using a binary classification model to determine if the subject is in a high pain severity state in need of further electrical stimulation or a low pain severity state that does not require electrical stimulation. In some cases, when the subject is determined to be in a low pain severity state using the binary classification model, the methods and systems do not apply a further stimulation to the brain. Alternatively, when analysis of the brain electrical signals from the sensorimotor cortex using the binary classification model indicates the subject is in a high pain severity state, the methods and systems may apply a further stimulation to the brain.

[0129] In some aspects, the closed loop system may be used to sense a subject’s need for treatment using the methods disclosed herein. For example, the closed loop system may be programmed to monitor brain activity from one or more regions of the brain and compare the brain activity to a range indicative of pain severity (e.g., low pain severity or high pain severity using binary classification model). Upon detection of electrical activity indicative of pain, the closed loop system may automatically commence a treatment protocol of applying electrical stimulation to the brain.

[0130] In additional aspects, the closed loop system may be used as a system for monitoring brain activity and correlating the brain activity to the subject’s level of pain. For example, since the closed loop system is configured for recording electrical signals from a subject’s brain, the subject’s level of pain may be monitored in real-time and correlated to the measured electrical signals to provide abiomarker that is related to the subject’s pain. For example, electrical activity measured when a subject is experiencing pain can be used to develop a biomarker, e.g., as range of electrical activity indicative of pain, and so on. As such, closed loop systems are useful for detecting pain caused by pain-associated disorders.

[0131] It is understood that electrical signals that are indicative of pain or pain relief for a subject may be recorded from a subject’s brain and may be used in aspects outside of a closed loop system. For example, electrical signals indicative of pain or pain relief for a subject may be recorded from the OFC, ACC, insula, vmPFC, BG, or CM, or other brain region using electrodes or another device operably coupled to the patient’s brain, which electrodes or device may or may not be part of a closed loop system. The patient may be treated as disclosed herein (e.g., by applying electrical stimulation to the brain), and electrical signals recorded from the OFC, ACC, insula, vmPFC, BG, or CM, or other region in real time as the treatment is administered or after the treatment is administered. The electric signals recorded after the administration of electrical stimulation is commenced may then be compared to the electric signals recorded prior to the treatment to determine features in the recorded electric signals that change post-treatment. These features provide a feedback signal to indicate whether the treatment is having an effect on the patient’s level of pain. These features can also serve as feedback signals to a closed loop system. These features may include the overall power, average power, or power in specific frequency ranges (e.g., alpha, delta, beta, gamma, and / or theta). In some cases, these features may be patient specific or specific to a particular pain state or both. For example, some of the features may be features found in a plurality of patients having a particular level of pain or a particular pain-associated disorder; some of the features may be features in a particular patient which may not be found in a significant number of other patients having the same level of pain. In some embodiments, a combination of patient-specific features and pain-specific features may be monitored to assess efficacy of treatment.

[0132] In a particular aspect, the closed loop system and methods provided herein may involve a recording of electrical signals from one or more regions (e.g., a right OFC region, a left OFC region a right ACC region, a left ACC region, a right VMPFC region, a left VMPFC region, or a combination thereof) of a patient’s brain, where the patient has a pain-associated disorder. The patient may then be treated by application of electrical stimulation to the brain, and electrical signals may be recorded from the same regions of the brain and compared to the pre-treatment recording. Features in the recorded signals that changed after the treatment would correspond to biomarkers that indicate whether the treatment is having an effect. The change in recorded signals can also optionally be correlated to the level of pain reported by the patient after the treatment. The change can be usedfor modulating the treatment in a closed loop system. For example, when the change in the recorded signal correlates with amelioration of pain, those features would indicate to a computing means of a closed loop system that further treatment need not be performed.

[0133] In some embodiments, one or more pattern recognition methods can be used in analyzing recorded brain electrical activity data to automate detection of brain activity features that distinguish pain states that are severe enough to need treatment from symptomless states (baseline) or minor pain states that do not need treatment. The models and / or algorithms can be provided in machine readable format and may be used to correlate the levels of overall power, average power, or power in specific frequency ranges (e.g., alpha, delta, beta, gamma, and / or theta) with pain severity and need for treatment with electrical stimulation to the brain. In some embodiments, the level of delta frequency power (such as 1 Hz to 4 Hz) in an OFC region contralateral to the to the side of body where the subject feels pain is correlated with pain severity to determine if a patient is in need of treatment with electrical stimulation. Alternatively or additionally, coherence within certain spectral frequency bands or other features of network connectivity may be correlated with pain severity and need for treatment with electrical stimulation to the brain.

[0134] In some embodiments, frequency decompositions are used to calculate the average power in each neurophysiological activity band (e.g., delta, theta, alpha, beta, gamma etc.). The slope of 1 / f power values across frequency bands is determined for a given epoch by fitting a curve to the band power data for that epoch (e.g., slope is calculated on multiple samples for each EEG contact and each time epoch, over the axes ‘frequency band’ vs ‘power’); this value is referred to as the spectral tilt. The spectral tilt values are fit to a pain classification model (e.g., binary pain classification model that distinguishes pain of high severity from pain of low severity) to determine how to adjust the programmed stimulation parameters.

[0135] In some embodiments, a computer implemented method is used for programming a DBS device to treat chronic pain in a subject, the computer performing steps comprising: a) receiving recorded brain electrical signal data from a sensorimotor cortex region of the brain of the subject; b) analyzing the recorded brain electrical signal data using a binary pain classification model that classifies the brain electrical signal data as indicating that the subject is in either a high pain severity state or a low pain severity state; c) adjusting one or more programmed stimulation parameters based on the recorded brain electrical signal data according to an algorithm control law; and d) instructing the DBS device to apply an electrical stimulation to a target stimulation site in the brain of the subject using an electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal data indicates the subject is transitioning to the high pain severity state or is ina high pain severity state. In some embodiments, the computer implemented method further comprises calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.

[0136] In certain embodiments, the computer implemented method further comprises splitting the recorded brain electrical signal data into consecutive time epochs. In some embodiments, each time epoch comprises 0.5 second to 10 minutes of time of the recorded brain electrical signal data, including any amount of time within this range such as 0.5 seconds, 0.75 seconds, 1 seconds, 2 seconds, 3 seconds, 4 seconds, 5 seconds, 6 seconds, 7 seconds, 8 seconds, 9 seconds, 10 seconds, 15 seconds, 20 seconds, 25 seconds, 30 seconds, 35 seconds, 40 seconds, 45 seconds, 50 seconds, 55 seconds, 1 minute, 2 minutes, 3 minutes, 4 minutes, 5 minutes, 6 minutes, 7 minutes,8 minutes, 9 minutes, or 10 minutes of time. In certain embodiments, the computer implemented method further comprises assigning a label to each time epoch indicating whether the binary classification model classifies the electrical signal data as indicating a high pain severity state or low pain severity state.

[0137] In some embodiments, the computer implemented method further comprises calculating average power values for one or more neurophysiological activity frequency bands for each time epoch, wherein the average power values are used to calculate pain scores for each time epoch. In some embodiments, the computer implemented method further comprises calculating pain scores using brain electrical signal data during a period when the subject is not undergoing electrical stimulation (i.e., stimulation-free data epoch) to distinguish the pain scores that are at or above the 70thpercentile corresponding to the high pain severity state and the pain scores that are at or below the 30thpercentile corresponding to the low pain severity state. In some embodiments, the stimulation-free data epoch comprises at least 5 minutes, 10 minutes, 30 minutes or 40 minutes of time of the recorded brain electrical signal data, including 5 minutes, 6 minutes, 7 minutes, 8 minutes,9 minutes, 10 minutes, 11 minutes, 12 minutes, 13 minutes, 14 minutes, 15 minutes, 16 minutes, 17 minutes, 18 minutes, 19 minutes, 20 minutes, 21 minutes, 22 minutes, 23 minutes, 24 minutes, 25 minutes, 26 minutes, 27 minutes, 28 minutes, 29 minutes, 30 minutes, 31 minutes, 32 minutes, 33 minutes, 34 minutes, 35 minutes, 36 minutes, 37 minutes, 38 minutes, 39 minutes, or 40 minutes. In some embodiments, the stimulation-free data epoch comprises 5 minutes to 1 hour, 5 minutes to 40 minutes, 10 minutes to 30 minutes, or 10 minutes to 20 minutes of stimulation-free data.

[0138] In some embodiments, a computer implemented method is used for selecting stimulation settings, the computer performing steps comprising a) ranking predicted stimulation effectivenessfor available settings of the DBS device based on classifier scores for stimulation effectiveness of each setting using a linear classification model; b) selecting stimulation settings predicted to have highest stimulation effectiveness based on the linear classification model; c) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness; d) analyzing the recorded brain electrical signal data to evaluate neural response of the subject to the electrical stimulation; e) updating the linear classification model based on the neural response of the subject to the electrical stimulation to generate an updated linear classification model; f) updating the ranking of predicted stimulation effectiveness for the available settings of the DBS device using the updated linear classification model; g) selecting stimulation settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; h) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying the electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; and i) repeating e) - h) to adjust the available settings of the DBS device to optimize stimulation effectiveness.

[0139] In some embodiments, a computer implemented method is used to predict whether a subject who has chronic pain will respond to deep brain stimulation based on calculating a classifier score to predict the likelihood of DBS therapy success.

[0140] Analyzing the recorded brain electrical activity may comprise the use of an algorithm or classifier. In some embodiments, a machine learning algorithm is used to classify brain activity as indicating whether or not pain is severe enough that electrical stimulation should be delivered to the brain of the patient. The machine learning algorithm may comprise a supervised learning algorithm. Examples of supervised learning algorithms may include Average One-Dependence Estimators (AODE), Artificial neural network (e.g., Backpropagation), Bayesian statistics (e.g., Naive Bayes classifier, Bayesian network, Bayesian knowledge base), Case-based reasoning, Decision trees, Inductive logic programming, Gaussian process regression, Group method of data handling (GMDH), Learning Automata, Learning Vector Quantization, Minimum message length (decision trees, decision graphs, etc.), Lazy learning, Instance-based learning Nearest Neighbor Algorithm, Analogical modeling, Probably approximately correct learning (PAG) learning, Ripple down rules, a knowledge acquisition methodology, Symbolic machine learning algorithms, Subsymbolic machine learning algorithms, Support vector machines (SVM), Random Forests, Ensembles of classifiers,Bootstrap aggregating (bagging), and Boosting. Supervised learning may comprise ordinal classification such as regression analysis and Information fuzzy networks (IFN). Alternatively, supervised learning methods may comprise statistical classification, such as AODE, Linear classifiers (e.g., Fisher's linear discriminant, Logistic regression, Naive Bayes classifier, Perceptron, and Support vector machine), quadratic classifiers, k-nearest neighbor, Boosting, Decision trees (e.g., C4.5, Random forests), Bayesian networks, and Hidden Markov models.

[0141] The machine learning algorithms may also comprise an unsupervised learning algorithm. Examples of unsupervised learning algorithms may include artificial neural network (recurrent or convoluted), Data clustering, Expectation-maximization algorithm, Self-organizing map, Radial basis function network, Vector Quantization, Generative topographic map, Information bottleneck method, and IBSEAD. Unsupervised learning may also comprise association rule learning algorithms such as Apriori algorithm, Eclat algorithm and FP-growth algorithm. Hierarchical clustering, such as Single-linkage clustering and Conceptual clustering, may also be used. Alternatively, unsupervised learning may comprise partitional clustering such as K-means algorithm and Fuzzy clustering.

[0142] In some instances, the machine learning algorithms comprise a reinforcement learning algorithm. Examples of reinforcement learning algorithms include, but are not limited to, temporal difference learning, Q-learning and Learning Automata. Alternatively, the machine learning algorithm may comprise Data Pre-processing. The methods described herein can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware.

[0143] In some embodiments, the machine learning algorithm uses artificial neural networks. In some embodiments, the machine learning algorithm uses a deep learning algorithm, which may include the use of convolutional neural networks, deep neural networks, recurrent neural networks, deep residual neural networks, long short-term memory networks, deep belief networks, multilayer perceptrons, or deep reinforcement learning, and the like. See, e.g., Pedrycz et al. Deep Learning: Algorithms and Applications (Studies in Computational Intelligence Book 865, Springer, 2019), Goodfellow et al. Deep Learning (Adaptive Computation and Machine Learning series, The MIT Press, 2016), and Various Deep Learning Algorithms in Computational Intelligence (edited by Oscar Humberto Montiel Ross, Mdpi AG, 2023); herein incorporated by reference in their entireties.

[0144] In certain embodiments, the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.

[0145] The methods can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware. The disclosed and other embodiments can be implemented as one or morecomputer program products, i.e., one or more modules of computer program instructions encoded on a computer readable medium for execution by, or to control the operation of, a data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine- readable storage substrate, a memory device, a composition of matter effecting a machine-readable propagated signal, or any combination thereof.

[0146] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0147] In a further aspect, the system for performing the computer implemented method, as described, may include a processor, a storage component (i.e., memory), a display component, and other components typically present in general purpose computers. In some embodiments, the processor is provided by a computer or handheld device (e.g., a cell phone or tablet). The storage component stores information accessible by the processor, including instructions that may be executed by the processor and data that may be retrieved, manipulated or stored by the processor.

[0148] The storage component includes instructions. For example, the storage component includes instructions for determining symptom severity of a subject having a movement disorder and / or diagnosing a subject with a movement disorder according to the methods described herein. The computer processor is coupled to the storage component and configured to execute the instructions stored in the storage component in order to receive a video of the subject’s entire body while the subject is walking and / or a video of a hand of the subject while the subject is finger-tapping and analyze the video images according to one or more algorithms, as described herein.

[0149] The processor and / or memory may be operably connected to a display device, for example, via a wired, such as a Universal Serial Bus (USB) connection, or wireless connection, such as a Bluetooth connection. Any convenient display device, such as a liquid crystal display (LCD), lightemitting diode (LED) display, plasma (PDP) display, quantum dot (QLED) display or cathode ray tube display device may be used. The display component displays information regarding thesymptom severity and / or diagnosis of the subject. In some embodiments, the display further displays the body and / or hand videos or video segments thereof. In some embodiments, the display further displays body movement features comprising relative power in 0.5 Hz to 1 Hz of ankle separation, standard deviation (STD) of ankle-hip distance, relative power in 2 Hz to 4 Hz of neck angle, relative power in 4 Hz to 6 Hz of arm-body angular velocity, and mean of neck angle. In some embodiments, the display further displays hand movement features comprising half-power frequency in pinky finger movement speed, relative power in 2 Hz to 4 Hz of finger tapping speed, and median absolute deviation (MAD) of finger tapping amplitude. In some embodiments, the display further displays information comparing movement metrics for single joint or multi-joint movements of the subject to reference value ranges for a normal age-matched control subject (i.e., healthy subject, subject not having a movement disorder). In some embodiments, the movement metrics are for finger-tapping or gait speed.

[0150] The storage component may be of any type capable of storing information accessible by the processor, such as a hard-drive, memory card, ROM, RAM, DVD, CD-ROM, USB Flash drive, write- capable, and read-only memories. The processor may be a general purpose processor, a graphics processor unit, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor can be a microprocessor, but in the alternative, the processor can be a controller, microcontroller, or state machine, combinations of the same, or the like. A processor can also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Although described herein primarily with respect to digital technology, a processor can also include primarily analog components. A computing environment can include any type of computer system, including, but not limited to, a computer system based on a microprocessor, a graphics processor unit, a mainframe computer, a digital signal processor, a portable computing device, a personal organizer, a device controller, and a computational engine within an appliance, to name a few.

[0151] The steps of a method, process, or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module, engine, and associated databases can reside in memory resources such as in RAM memory, FRAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or anyother form of non-transitory computer-readable storage medium, media, or physical computer storage known in the art. An exemplary storage medium can be coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. The ASIC can reside in a user terminal. In the alternative, the processor and the storage medium can reside as discrete components in a user terminal.

[0152] The instructions may be any set of instructions to be executed directly (such as machine code) or indirectly (such as scripts) by the processor. In that regard, the terms "instructions," "steps" and "programs" may be used interchangeably herein. The instructions may be stored in object code form for direct processing by the processor, or in any other computer language including scripts or collections of independent source code modules that are interpreted on demand or compiled in advance.

[0153] Data may be retrieved, stored or modified by the processor in accordance with the instructions. For instance, although the system is not limited by any particular data structure, the data may be stored in computer registers, in a relational database as a table having a plurality of different fields and records, XML documents, or flat files. The data may also be formatted in any computer-readable format such as, but not limited to, binary values, ASCII or Unicode. Moreover, the data may comprise any information sufficient to identify the relevant information, such as numbers, descriptive text, proprietary codes, pointers, references to data stored in other memories (including other network locations) or information which is used by a function to calculate the relevant data.

[0154] In certain embodiments, the processor and storage component may comprise multiple processors and storage components that may or may not be stored within the same physical housing. For example, some of the instructions and data may be stored on removable CD-ROM and others within a read-only computer chip. Some or all of the instructions and data may be stored in a location physically remote from, yet still accessible by, the processor. Similarly, the processor may comprise a collection of processors which may or may not operate in parallel.

[0155] In some embodiments, the method can be performed using a cloud computing system. In these embodiments, recorded brain electrical signal data from a sensorimotor cortex region of the brain of the subject, and the programming can be exported to a cloud computer, which runs the program, and returns an output to the user.

[0156] Components of systems for carrying out the presently disclosed methods are further described in the examples below.Administration of a Pharmacological Agent

[0157] Embodiments of the methods and systems provided in this disclosure may also include administration of an effective amount of at least one pharmacological agent. By “effective amount” is meant a dosage sufficient to prevent or treat a pain-associated disorder in a subject as desired. The effective amount will vary somewhat from subject to subject, and may depend upon factors such as the age and physical condition of the subject, severity of the pain-associated disorder being treated, the duration of the treatment, the nature of any concurrent treatment, the form of the agent, the pharmaceutically acceptable carrier used if any, the route and method of delivery, and analogous factors within the knowledge and expertise of those skilled in the art. Appropriate dosages may be determined in accordance with routine pharmacological procedures known to those skilled in the art, as described in greater detail below.

[0158] If a pharmacological approach is employed in the treatment of a pain-associated disorder, the specific nature and dosing schedule of the agent will vary depending on the particular nature of the disorder to be treated. Representative pharmacological agents that may find use in certain embodiments of the subject invention include, but are not limited to, nonsteroidal anti-inflammatory drugs (NSAIDs) such as aspirin, ibuprofen and naproxen; COX2 inhibitors such as rofecoxib, celecoxib, and etoricoxib; opioids such as morphine, codeine, oxycodone, hydrocodone, dihydromorphine, and pethidine; selective serotonin reuptake inhibitors (SSRIs) such as fluoxetine (Prozac), paroxetine (Paxil, Pexeva), sertraline (Zoloft), citalopram (Celexa) and escitalopram (Lexapro); and serotonin-norepinephrine reuptake inhibitors (SNRIs) such as duloxetine (Cymbalta), venlafaxine (Effexor XR), desvenlafaxine (Pristiq, Khedezla) and levomilnacipran (Fetzima), and the like.

[0159] In certain aspects, the administration of a pharmacological agent involves using a pharmacological delivery device such as, but not limited to, pumps (implantable or external devices), epidural injectors, syringes or other injection apparatus, catheter and / or reservoir operatively associated with a catheter, etc. For example, in certain embodiments a delivery device employed to deliver at least one pharmacological agent to a subject may be a pump, syringe, catheter or reservoir operably associated with a connecting device such as a catheter, tubing, or the like. Containers suitable for delivery of at least one pharmacological agent to a pharmacological agent administration device include instruments of containment that may be used to deliver, place, attach, and / or insert the at least one pharmacological agent into the delivery device for administration of the pharmacological agent to a subject and include, but are not limited to, vials, ampules, tubes,capsules, bottles, syringes and bags. Administration of a pharmacological agent may be performed by a user or by a closed loop system.Systems

[0160] The present disclosure also provides systems which find use, e.g., in practicing the subject methods. The system may be an open-loop or closed-loop system configured for performing the methods provided herein. In some embodiments, the system may include a DBS electrode adapted for positioning at a location to deliver electrical stimulation to a target stimulation site (e.g., in an ACC SGC, insula, motor / sensory cortex, VS, thalamus, VPthal, CM, periaqueductal gray, periventricular gray, caudate nucleus, VMPFC, or frontal operculum region, or other region) of the brain of the subject and a detection electrode adapted for positioning at a sensorimotor cortex region for detection of neural activity associated with pain or pain relief (e.g., OFC, ACC, insula, vmPFC, BG, or CM region, or other region) and for recording an electrical signal from the brain region before, during, or after an electrical stimulation is applied to the brain. In a closed-loop system, the system may also include a computing means and control unit programmed to instruct the DBS electrode to apply an electrical stimulation to the brain in a manner effective to treat the pain in the subject and / or ameliorate the pan caused by the pain-associated disorder; receive the electrical signals associated with pain or pain relief from the brain of the subject via the detection electrode; apply electrical signal metrics to a control algorithm that is tuned to a clinically relevant target (e.g., a range of signal indicative of severity of pain and the need for treatment); and automatically delivering electrical stimulation to the brain via the control unit, pulse generator and DBS electrode in a manner effective to treat the pain or ameliorate the pain of the pain-associated disorder if the electrical signal metrics indicate that the patient is in need of treatment. For example, electrical activity in the theta frequency range (such as 4 Hz to 8 Hz) and / or beta frequency range (such as 12 Hz to 30 Hz) from the right amygdala, left amygdala, right orbitofrontal cortex, left subgenual cingulate, or right hippocampus, or other region may be measured with a detection electrode using this system, wherein the control algorithm receives the electrical activity data from the detection electrode and automates delivery of electrical stimulation via the control unit, a pulse generator and the DBS electrode to the brain when the level of theta frequency (such as 4 Hz to 8 Hz) and / or beta frequency (such as 12 Hz to 30 Hz) power is in a certain range indicating that a patient has pain severe enough that the patient is in need of treatment. In some embodiments, one or more programmed stimulation parameters are modulated according to the algorithm’s control law based on the recorded electrical activity data, and modulated electrical stimulation is delivered to the brain via the control unit, pulse generator and DBS electrodein a manner effective to ameliorate pain caused by the pain-associated disorder. The closed loop system may include an on-body pulse generator that is connected to the implanted DBS electrodes and hence can apply electrical stimulation to the brain automatically upon receiving a communication from the control unit.

[0161] The processor of the closed-loop system may run programming for assessing the effectiveness of treatment and / or effectiveness of amelioration of pain and modulate a parameter of the treatment as needed without user intervention. Thus, the closed-loop system may not necessarily include a user interface for a user to instruct the DBS electrode to apply an electrical stimulation to the brain to treat the pain in the subject. However, in some embodiments, a user interface may be included in the closed-loop system which may be used to confirm the recommendation of the closed loop system or to override it or to change the recommendation.

[0162] In certain aspects, a control algorithm for the methods and systems of the present disclosure may include steps of comparing an electrical signal from a region of the brain of a subject to a normal or reference electrical signal (e.g., for substantially pain-free state), wherein when the electrical signal is significantly different from the normal or reference electrical signal, the control algorithm includes steps of directing a device to apply electrical stimulation to the brain of the subject, followed by measurement of electrical signals from the region of the brain and comparing it to a normal or reference electrical signal, where when the measured signal is significantly different from normal or reference electrical signal, the algorithm includes the step of applying another electrical stimulation to the brain.

[0163] In some embodiments, the control algorithm utilizes a machine learning algorithm to analyze inputted brain electrical activity data to automate detection of brain activity features that distinguish pain states that are severe enough to need treatment from symptomless states (baseline) or minor pain states that do not need treatment. The control algorithm then directs a device to apply electrical stimulation to the brain of the subject if the brain activity features indicate a pain state severe enough that the subject should be treated with electrical stimulation. For example, brain electrical activity from a sensorimotor cortex region may be measured with a detection electrode, wherein the control algorithm receives the electrical activity data from the detection electrode, analyzes the brain electrical activity using a binary pain classification model that classifies the brain electrical activity as indicating that the subject is in either a high pain severity state or a low pain severity state (see, e.g., Example 1 ), and automates delivery of electrical stimulation via a DBS electrode to the brain when a brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state. In some embodiments, one or moreprogrammed stimulation parameters are modulated according to the algorithm’s control law based on the recorded electrical activity data; and modulated electrical stimulation is delivered to the brain via the DBS electrode in a manner effective to ameliorate the pain. In certain embodiments the algorithm provides updated optimal stimulation setting recommendations to the clinician for guiding programing and decision making.

[0164] Components of systems for carrying out the presently disclosed methods are further described in the examples below.Kits

[0165] Also provided are kits comprising software for carrying out the computer implemented methods, described herein, for programming a DBS device to treat chronic pain in a subject. In some embodiments, the kit comprises a non-transitory computer-readable medium comprising the software and instructions for treating chronic pain in a subject with DBS according to the methods described herein (e.g., using neural features for decoding pain to adjust DBS settings to treat chronic pain). In some embodiments, the kit comprises a system for treating chronic pain comprising a processor programmed according to a computer implemented method described herein; and a display component for displaying information regarding the pain severity of the subject (e.g., whether the subject is in a high pain severity state or a low pain severity state according to the binary classification model).

[0166] In addition to the above components, the subject kits may further include (in certain embodiments) instructions for practicing the subject methods. These instructions may be present in the subject kits in a variety of forms, one or more of which may be present in the kit. One form in which these instructions may be present is as printed information on a suitable medium or substrate, e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert, and the like. Yet another form of these instructions is a computer readable medium, e.g., diskette, compact disk (CD), flash drive, and the like, on which the information has been recorded. Yet another form of these instructions that may be present is a website address which may be used via the internet to access the information at a removed site.Utility

[0167] The methods and systems of the present disclosure find use in a variety of different applications, including the treatment of pain such as chronic, idiopathic, nociceptive or neuropathic pain. The pain may be caused by a pain associated disorder such as, but not limited to, post-strokepain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-related pain (e.g., diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy), peripheral vascular disease, pain after an amputation, visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, or chronic leg pain, neck pain, foot pain, or arm pain.

[0168] In certain aspects, the methods and systems of the present disclosure may be used to treat chronic pain as well other symptoms associated with chronic pain, e.g., constant attention to the experience of pain, emotional toll of the experience of constant pain, fatigue, depression and / or anxiety. Efficacy of the treatment of patients suffering from chronic pain may be measured in an art accepted manner such as, by using a visual analog scale or a verbal rating scale.

[0169] Closed-loop stimulation can be finely targeted and tuned in a personalized manner to achieve more reliable and / or more effective pain relief compared to conventional DBS techniques. The use of DBS for treatment of chronic pain has been limited by poor efficacy in most patients. In a feasibility clinical trial for refractory chronic pain, we demonstrated a precision-medicine approach to identify unique, optimal brain targets and real-time adaptive DBS programs that provided significant pain relief (see, e.g., Example 2). The most effective personalized stimulation sites are identified through a systematic, multi-site brain mapping trial period. Personalized pain biomarkers (i.e., neural activity patterns associated with chronic pain) are identified and used to adaptively adjust DBS using algorithms that control and adjust neurostimulator settings in response to pain-associated brain states.Examples of Non-Limiting Aspects of the Disclosure

[0170] Aspects, including embodiments, of the present subject matter described above may be beneficial alone or in combination, with one or more other aspects or embodiments. Without limiting the foregoing description, certain non-limiting aspects of the disclosure numbered 1-73 are provided below. As will be apparent to those of skill in the art upon reading this disclosure, each of the individually numbered aspects may be used or combined with any of the preceding or following individually numbered aspects. This is intended to provide support for all such combinations of aspects and is not limited to combinations of aspects explicitly provided below:1 . A method for treating chronic pain in a subject, the method comprising: positioning a first electrode at a first location of the brain of the subject to deliver electrical stimulation to a target stimulation site to relieve the chronic pain;positioning a second electrode at a second location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; detecting the brain electrical signal at the sensorimotor cortex region of the brain the subject using the second electrode; analyzing the brain electrical signal using a binary pain classification model that classifies the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state; and applying electrical stimulation to the target stimulation site in the brain of the subject using the first electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state.2. The method of aspect 1 , wherein the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.3. The method of aspect 1 or 2, further comprising calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.4. The method of any one of aspects 1 -3, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.5. The method of aspect 4, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.6. The method of aspect 5, wherein the OFC region is a contralateral OFC region.7. The method of any one of aspects 4-6, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.8. The method of any one of aspects 1 -7, wherein the target stimulation site is in an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.9. The method of any one of aspects 1 -8, further comprising mapping the brain of the subject to identify an optimal location in the sensorimotor cortex region to detect the brain electrical signal associated with the chronic pain.10. The method of any one of aspects 1 -9, further comprising mapping the brain of the subject to identify an optimal target stimulation site to deliver the electrical stimulation to relieve the chronic pain.11. The method of aspect 10, wherein said mapping comprises identifying the optimal target stimulation site in each cerebral hemisphere.12. The method of any one of aspects 1 -11 , wherein the chronic pain is caused by a pain- associated disorder, wherein applying the electrical stimulation relieves the pain.13. The method of aspect 1 , wherein the pain-associated disorder is post-stroke pain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-related pain, peripheral vascular disease, pain after an amputation, visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, or chronic leg pain, neck pain, foot pain, or arm pain.14. The method of aspect 13, wherein the nerve-related pain is caused by diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy.15. The method of aspect 13, wherein the pain-associated disorder is said post-stroke pain, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a right ACC region and a left ACC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right ACC region, a right VPthal region, and a left ACC region.16. The method of aspect 13, wherein the pain-associated disorder is said complex regional pain syndrome, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a left ACC region and a right VMPFC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right CM region and a left body of the caudate nucleus region.17. The method of any one of aspects 1 -16, wherein the brain electrical signal is a local field potential.18. The method of any one of aspects 1 -17, further comprising using a control algorithm to automate said applying electrical stimulation when the brain electrical signal exceeds the specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state.19. The method of aspect 18, wherein the control algorithm uses a machine learning algorithm for pain classification.20. The method of aspect 19, wherein the machine learning algorithm is a supervised machine learning algorithm, an unsupervised machine learning algorithm, or a semi-supervised machine learning algorithm.21. The method of any one of aspects 18-20, wherein the control algorithm further modulates one or more programmed stimulation parameters to relieve the chronic pain.22. The method of any one of aspects 18-21 , wherein the control algorithm further uses linear discriminant analysis (LDA) to adjust stimulation amplitude, or frequency of the electrical stimulation.23. The method of any one of aspects 1 -22, wherein the electrical stimulation is applied unilaterally or bilaterally.24. The method of any one of aspects 1-23, wherein the brain electrical signal comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations.25. The method of aspect 24, wherein the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz.26. The method of aspect 25, wherein the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.27. The method of any one of aspects 1 -26, wherein the control algorithm further modulates one or more programmed stimulation parameters based on a level of power of the brain electrical signal.28. The method of aspect 27, wherein the level of power is a level of delta power from a contralateral OFC region.29. The method of aspect 27 or 28, wherein changes in power features associated with the chronic pain are sustained for at least 4 seconds.30. The method of any one of aspects 1-29, wherein the first electrode is placed within or on a surface of the ACC region, the subgenual cingulate gyrus (SGC) region, the insula region, the motor / sensory cortex region, the ventral striatum (VS) region, the thalamus region, the thalamus ventral posterior complex (VPthal) region, the centromedian thalamus (CM) region, the periaqueductal gray region, the periventricular gray (PVG) region, the caudate nucleus region, the ventromedial prefrontal cortex (VMPFC) region, or the frontal operculum region.31 . The method of any one of aspects 1 -30, wherein the second electrode is placed within or on a surface of the OFC region, the ACC region, the insula region, the vmPFC region, the BG region, or the CM region.32. The method of any one of aspects 1-31 , wherein the first electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.33. The method of any one of aspects 1-32, wherein the second electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.34. The method of any one of aspects 1 -33, wherein the second electrode is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.35. The method of any one of aspects 1-34, wherein the method further comprises assessing effectiveness of the treatment in the subject.36. The method of aspect 35, wherein said assessing comprises using a visual analog scale (VAS), a pain intensity numerical rating score (NRS), or a verbal rating scale.37. A computer implemented method for programming a deep brain stimulation (DBS) device to treat chronic pain in a subject, the computer performing steps comprising: a) receiving recorded brain electrical signal data from a sensorimotor cortex region of the brain of the subject; b) analyzing the recorded brain electrical signal data using a binary pain classification model that classifies the brain electrical signal data as indicating that the subject is in either a high pain severity state or a low pain severity state; c) adjusting one or more programmed stimulation parameters based on the recorded brain electrical signal data according to an algorithm control law; and d) instructing the DBS device to apply an electrical stimulation to a target stimulation site in the brain of the subject using an electrode in a manner effective to treat the chronic pain in thesubject when the brain electrical signal data indicates the subject is transitioning to the high pain severity state or is in the high pain severity state.38. The computer implemented method of aspect 37, wherein the brain electrical signal data comprises field potential data.39. The computer implemented method of aspect 37 or 38, wherein the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.40. The computer implemented method of any one of aspects 37-39, further comprising calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.41 . The computer implemented method of any one of aspects 37-40, further comprising splitting the recorded brain electrical signal data into consecutive time epochs.42. The computer implemented method of aspect 41 , further comprising calculating average power values for one or more neurophysiological activity frequency bands for each time epoch, wherein the average power values are used to calculate pain scores for each time epoch.43. The computer implemented method of aspect 41 or 42, further comprising calculating pain scores using brain electrical signal data during a stimulation-free data epoch to distinguish the pain scores that are at or above the 70thpercentile corresponding to the high pain severity state and the pain scores that are at or below the 30thpercentile corresponding to the low pain severity state.44. The computer implemented method of aspect 43, wherein the stimulation-free data epoch comprises at least 5 minutes, 10-minutes, 30-minutes or 40 minutes of time of the recorded brain electrical signal data.45. The computer implemented method of any one of aspects 37-44, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC)region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.46. The computer implemented method of aspect 45, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.47. The computer implemented method of aspect 46, wherein the OFC region is a contralateral OFC region.48. The computer implemented method of any one of aspects 45-47, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.49. The computer implemented method of any one of aspects 37-48, wherein the target stimulation site is in an ACC region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.50. The computer implemented method of any one of aspects 37-49, further comprising: a) ranking predicted stimulation effectiveness for available settings of the DBS device based on classifier scores for stimulation effectiveness of each setting using a linear classification model; b) selecting stimulation settings predicted to have highest stimulation effectiveness based on the linear classification model; c) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness; d) analyzing the recorded brain electrical signal data to evaluate neural response of the subject to the electrical stimulation;e) updating the linear classification model based on the neural response of the subject to the electrical stimulation to generate an updated linear classification model; f) updating the ranking of predicted stimulation effectiveness for the available settings of the DBS device using the updated linear classification model; g) selecting stimulation settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; h) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying the electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; and i) repeating e) - h) to adjust the available settings of the DBS device to optimize stimulation effectiveness.51 . The computer implemented method of aspect 50, wherein the linear classification model uses linear discriminant analysis (LDA) to adjust amplitude of current and frequency of the electrical stimulation.52. The computer implemented method of any one of aspects 37-51 , wherein the recorded brain electrical signal data comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations.53. The computer implemented method of aspect 52, wherein the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz.54. The computer implemented method of aspect 53, wherein the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.55. The computer implemented method of any one of aspects 37-54, wherein the algorithm control law further modulates one or more programmed stimulation parameters based on a change in a level of power of a brain electrical signal.56. The computer implemented method of aspect 55, wherein the level of power is a level of delta power from a contralateral OFC region.57. The computer implemented method of aspect 55 or 56, wherein the change in the level of power is sustained for at least 4 seconds.58. The computer implemented method of any one of aspects 37-57, further comprising storing a user profile for the subject comprising information regarding the recorded brain electrical signal data associated with the chronic pain.59. The computer implemented method of any one of aspects 37-58, further comprising storing a user profile for the subject comprising information regarding the programmed stimulation parameters used to apply electrical stimulation to the target stimulation site in the brain of the subject to treat the chronic pain in the subject based on the recorded brain electrical signal data.60. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the method of any one of aspects 37-59.61. A kit comprising the non-transitory computer-readable medium of aspect 60 and instructions for treating chronic pain in a subject with a deep brain stimulation device.62. A system for treating chronic pain in a subject, the system comprising: a first electrode adapted for positioning at a location to deliver electrical stimulation to a target stimulation site to relieve the chronic pain in the subject; a second electrode adapted for positioning at a location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; and a processor programmed according to the computer implemented method of any one of aspects 37-59 to adjust one or more stimulation parameters based on the recorded brain electrical signal data and instruct the first electrode to apply an electrical stimulation to the target stimulation site in the brain of the subject in a manner effective to treat the chronic pain in the subject.63. The system of aspect 62, wherein the brain electrical signal data comprises field potential data.64. The system of aspect 62 or 63, wherein the first electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.65. The system of any one of aspects 62-64, wherein the second electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.66. The system of any one of aspects 62-65, wherein the second electrode is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.67. The system of any one of aspects 62-66, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.68. The system of aspect 67, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.69. The system of aspect 68, wherein the OFC region is a contralateral OFC region.70. The system of any one of aspects 67-69, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.71 . The system of any one of aspects 62-70, wherein the target stimulation site is in an ACC region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, aperiventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.72. The system of any one of aspects 62-71 , wherein the system further comprises a user interface comprising an input electronically coupled to the processor for instructing the first electrode to apply an electrical stimulation to the target stimulation site to relieve the chronic pain in the subject.73. The system of aspect 72, wherein the user interface is password protected and is operable by a health care practitioner.

[0171] It will be apparent to one of ordinary skill in the art that various changes and modifications can be made without departing from the spirit or scope of the invention.EXPERIMENTAL

[0172] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used (e.g. amounts, temperature, etc.) but some experimental errors and deviations should be accounted for. Unless indicated otherwise, parts are parts by weight, molecular weight is weight average molecular weight, temperature is in degrees Centigrade, and pressure is at or near atmospheric.

[0173] All publications and patent applications cited in this specification are herein incorporated by reference as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference.

[0174] The present invention has been described in terms of particular embodiments found or proposed by the present inventor to comprise preferred modes for the practice of the invention. It will be appreciated by those of skill in the art that, in light of the present disclosure, numerous modifications and changes can be made in the particular embodiments exemplified without departing from the intended scope of the invention. All such modifications are intended to be included within the scope of the appended claims.Example 1First-In-Human Prediction of Chronic Pain State Using Intracranial Neural BiomarkersIntroduction

[0175] Chronic pain syndromes pose a major healthcare problem and are leading contributors to disability worldwide1. Neuropathic pain syndromes such as post-stroke and phantom limb pain are particularly refractory to treatment and impose substantial suffering. A hurdle to a mechanistic understanding of chronic pain, and the development of effective diagnostics and therapeutics, is the lack of objective measures of pain severity or underlying neurophysiology. Of necessity, chronic pain is commonly measured by the individual’s subjective report, an interrogative approach limited by difficulties in quantitation, reliability and interindividual comparability. Objective biomarkers for chronic pain would greatly facilitate diagnosis and classification of pain pathophysiology, assist with disease prognostication or prediction of therapy response, and catalyze therapeutic development23.

[0176] Most previous attempts to identify pain biomarkers have focused on healthy participants and experimental thermal pain, which ignores natural, spontaneous fluctuations in individuals’ chronic pain experience4-7. Even studies of spontaneous chronic pain severity are limited by characterization over short timescales (minutes) due to reliance on ex vivo technologies such as electroencephalography7and blood-oxygen-level-dependent functional magnetic resonance imaging89that are not amenable to frequent, long-term measurement. Further, interpretation of blood-oxygen-level-dependent functional magnetic resonance imaging responses associated with chronic back pain is complicated due to inclusion of participants with heterogeneous sources of pain (for example, inflammatory, myofascial and neuropathic). It is unclear whether lessons learned from healthy human participants or mixed pain syndromes translate to individuals with chronic neuropathic pain over clinically relevant time periods. A growing body of evidence from humans and animals suggests that chronic pain processing engages the medial frontal cortex in a manner distinct from acute, thermal pain10-13.

[0177] Frontal brain regions harbor important signals that integrate the somatosensory, affective and cognitive dimensions of pain. The rostral anterior cingulate cortex (ACC) has been extensively implicated in affective / emotional processing related to both acute and chronic pain10 14and proposed as a therapeutic stimulation target to treat chronic pain15. The orbitofrontal cortex (OFC) has rarely been studied in pain, but its reciprocal connections with the ACC and many other regions in thefunctional pain network such as the amygdala, insula and ventral striatum16 17make it well positioned to influence pain perception. Functional imaging has revealed increased OFC activity in response to multimodal sensory stimuli during tasks requiring the integration of bodily signals to guide behavior (for example, interoception)18 19, reversal learning and pain expectation related to monetary reward20. Direct recordings from the OFC have confirmed key signals that track internal deliberation of reward choices21and OFC stimulation can improve mood state in depressed patients22. Other studies have identified the ventromedial prefrontal cortex, a region overlapping with definitions of OFC, as an important region associated with interindividual variability in acute pain perception23and negative affective responses to multimodal stimuli24. Specific coactivation of the dorsal ACC and OFC is seen both during opioid consumption and in response to placebo analgesia25, implicating these regions in myriad functions ranging from sensory perception and expectation to cognitive flexibility and mood regulation. While long-distance connections between the ACC / OFC and subcortical structures may support plasticity underlying chronic pain, signals from either brain region alone may be sufficient to track pain state. It remains an open question whether previous studies of chronic back pain due to heterogeneous pain generators generalize to true neuropathic pain states or is reflected in direct brain activity in vivo. Therefore, we targeted two relatively less-studied, non-somatosensory brain regions as part of a larger deep brain stimulation (DBS) clinical trial. We hypothesized that neural activity from the ACC or OFC may provide an integrated biomarker of the subjective experience underlying chronic pain severity. To our knowledge, long-term direct brain measurement of chronic pain-related neural activity has never been done.

[0178] By studying brain biomarkers with high resolution across both short (sub-second) and long (months) timescales in humans, we aimed to develop a sensitive test of high pain states and study the neurophysiologic basis of spontaneous chronic pain. Such insights potentially bring new clarity to pain biology and help to inform personalized treatments such as tailored closed-loop DBS therapies for chronic pain2627. We collected pain reports and intracranial recordings in the ambulatory setting using a novel bidirectional brain implant (Medtronic Activa PC + S) in four human participants with long-standing refractory, chronic neuropathic pain over 3-6 months. These are first-in-human chronic in vivo neural recordings of key pain-related cortical regions8 14, the ACC and OFC. We analyzed local field potentials (LFPs) with machine learning to develop personalized neural signatures for chronic pain severity and characterize the relative feature importance from each brain region. By comparing brain-based signatures for spontaneous, chronic pain with signatures derived for acute, evoked thermal pain, we identify distinct neural substrates for each. Brain biomarkers of chronic pain reflected sustained changes in neural power features that tended to rely more on theOFC, which only generalized to the acute pain condition within one participant. Acute pain decoding tended to reflect greater contributions from the ACC with more frequent, transient changes in power. Biomarker decoding was superior for current versus recent changes in pain state and largely stable over time. These data demonstrate that spontaneous, chronic pain states can be predicted from direct brain activity over ecological timescales in the ambulatory setting.ResultsLongitudinal pain-state tracking

[0179] Each participant provided pain score reports multiple times daily (range 2-8 / d) over a range of 78-184 d (mean 320 reports / participant; Supplementary Table 1). All participants reported 11 -point pain intensity using a numerical rating score (NRS; 0-10), while two participants (CP3 and CP4) also provided pain intensity via a visual analog score (VAS), the short-form McGill Pain Questionnaire (SF-MPQ) and pain unpleasantness (NRS and VAS). Pain NRS values were high for all participants (mean ± s.d., 7.8 ± 0.9), and all ratings exhibited fluctuation exceeding established minimal clinically important differences (MCID for SF-MPQ > 5 (ref. 28), NRS > 2 (ref. 29), VAS > 1 .6; FIGS. 1 B, 1 C). To assess the internal validity of pain score reporting, we computed Pearson’s correlations between VAS and NRS for pain intensity and unpleasantness, which were highly correlated within participants (FIG. 1 D; R2range 0.84-0.96). We evaluated periodicity in chronic pain fluctuation by computing an autocorrelation for each participant’s pain NRS (FIG. 1 E). While all participants showed diurnal cycles in their pain states, we also observed clinically significant multidien cycles of pain fluctuation, with two participants exhibiting pain fluctuation nearly every 3 d (72 h, CP1 and CP3).Cross-validated decoding of chronic pain state

[0180] To identify biomarkers of chronic pain state, we used ambulatory, intracranial LFP recordings from the ACC and OFC (FIGS. 2A, 2B) to build models decoding participants’ pain severity scores. Candidate biomarkers consisted of spectral power in frequency bands of interest, which showed no clear trend over time to suggest non-neural changes in recordings (FIG. 6). We first tried predicting exact reported pain metrics using cross-validated multivariate least absolute shrinkage and selection operator (LASSO) regression on neural band-power data (FIG. 8). First, we created subregion decoding models using band-power values sub-selected from individual brain regions and / or hemispheres and full models using all available data (that is, bilateral ACC / OFC; FIGS. 2C, 2D). For participants with bihemispheric electrode implants (CP2-4), there were nine possible subregionmodels through possible combinations of contralateral, ipsilateral or both hemispheres (relative to side of pain) from the ACC, OFC or both regions. Overall regression-based prediction showed mixed results (FIG. 8), often poor but at other times comparable to classification (below). Regression of NRS pain intensity was poor in all participants with multivariate coefficients of determination (R2) between observed versus predicted scores ranging from -0.2 to 0.1 across all subregion and full models. LASSO regression performance on the alternative metrics of pain intensity VAS and unpleasantness NRS in two participants showed R2values as high as 0.3 (CP4). Even an R2value of 0.18 (CP3 VAS) is equivalent to a Pearson’s r= 0.42, a moderate effect size which corresponds a binary classification area under the curve (AUG) = 0.74 assuming normally distributed, equal samples30. Notably, regression performed well when predicting MPQ in two participants, as evidenced by many values of R2> 0.7 (CP4). In summary, regression models performed moderately at best, when predicting pain intensity (NRS or VAS), but fared better for other pain metrics. Specifically for tracking pain intensity, these results may further support previous observations that a forced choice between low versus high pain states may be more pragmatic for clinical application6.

[0181] To investigate the role of different cortical circuits in a forced-choice framework, we turned our attention to predicting dichotomized pain scores (high versus low divided by median). Using cross-validated linear discriminant analysis (LDA; Methods) with at least one subregion model, dichotomized chronic pain NRSs could be classified in all participants (FIGS. 2D, 2E, highest LDA AUCs (R value), and FIG. 9): 0.673 (0.016), 0.851 (0.001 ), 0.721 (0.001), 0.802 (0.001 ) for CP1 - CP4, respectively. Ranges for chronic pain NRS decoding statistics were as follows: positive predictive value 0.83-0.93, sensitivity 0.62-0.84 and specificity 0.56-0.82 (Supplementary Table 2). In three participants, the best decoding performance resulted from either full models or combining ACC and OFC activity in either hemisphere. Overall, chronic pain state could be significantly decoded across all participants using contralateral OFC alone. Misclassified pain states were distributed across timepoints spanning months and the range of pain scores (that is, not only near the dichotomized boundary), confirming robustness of prediction (FIG. 10).

[0182] Decoding and biomarkers appeared stable over time. Compared to the leave-one-out cross- validation (LOOCV) method (which leaves out power values and pain score class associated with one recording clip at a time), we observed better performance when we trained LDA pain NRS models on the first 70% of data and tested on the remaining 30% with AUCs near 0.9 in all but in one participant (CP1 ), with the fewest number of recordings (FIG. 11 ). This suggests stability in the biomarkers tracking pain intensity NRSs over the timescale of months. To test performance of an independent method (linear state space model, LSSM) that may be used to flexibly guide brainstimulation, we used full models (FIG. 12) to predict pain intensity NRSs. LSSM had similar performance to LDA in all participants except CP1 . Moreover, two participants reported additional pain metrics including pain intensity VAS, pain unpleasantness NRS / VAS and SF-MPQ. Full models could significantly decode these pain metrics using both LDA and LSSM (Supplementary FIGS. 4, 5A, 7, and 8) except for CP3 unpleasantness VAS using LSSM. Using LDA, SF-MPQ decoding showed equally high performance using either the full model or contralateral OFC / ACC subregion model (AUC (R value) CP3 = 0.623 (0.004); CP4 = 0.995 (0.001 )), with positive predictive values ranging from 0.77 to 0.99 (Supplementary Table 2). Further, using single region LDA models for classification of chronic pain unpleasantness NRS, a measure of the affective dimension of pain, showed the highest AUCs for data from contralateral ACC (AUC 0.69 (CP3) and 0.84 (CP4)) as expected.

[0183] By visualizing the normalized, mean LDA feature weights supporting successful LDA decoding (FIG. 2E), we observed that the most important power features varied across participants. Despite this variability, a single brain region was sufficient to track chronic pain states in each participant, with contralateral OFC being common across all participants.

[0184] It is possible that neural features may track the relative fluctuation of pain, rather than the current pain state. To test for this possibility, we constructed regression and LDA models of successive pain score differences (that is, prior pain score subtracted from current pain score; FIGS. 14A, 14B). Dichotomizing these difference values by those equal to zero (no change) versus those that were non-zero (increasing or decreasing) may identify neural activity distinguishing stable versus fluctuating pain states. Classification of stable versus fluctuating pain NRSs was broadly significant only in one participant (CP4). Consistent with previous results8, we observed that stable versus changing pain scores in two participants (CP2 and CP3) could be distinguished only when models included neural signals from ACC. All LDA model AUC values were inferior to contemporaneous pain-state prediction (FIG. 14A). Further, regression of pain NRS difference was poor across participants (R2range -0.2 to 0.1 ) and alternative pain metric regression (for example, VAS) was not successful except for unpleasantness NRS and VAS each in one participant (FIG. 14B). The failure of regression to robustly predict increases or decreases in sequential pain scores over many hours may imply that ACC and OFC harbor weak signals for prediction errors over these timescales or that subjective pain experience is integrated over shorter historical durations than many hours. Therefore, while LFPs from the ACC and OFC may contain information relevant to recent changes in pain, the current chronic pain experience appears to be more reliably represented across participants and pain-related metrics compared to pain fluctuations.Cross-validated decoding of acute pain state

[0185] To compare with neural mechanisms supporting chronic pain, we next sought to identify biomarkers of acute, experimental thermal pain state. Brain recordings and pain ratings were collected while participants participated in quantitative sensory testing (FIG. 3A). Background chronic pain ratings just before the acute task were equivalent to the mode of participants’ reported scores (participants CP1 -4 and NRS 9, 8, 7 and 8 / 10, respectively; compare to Supplementary Table 1 ). Heat stimuli at five different temperatures, calibrated for each participant, were applied with a thermal probe to the most painful region on the side of the body affected by chronic pain (aff-ACUTE) and the same body part on the unaffected side (unaff-ACUTE) for five trials per target temperature; pain intensity NRS was reported 3 s after target temperature was reached (FIGS. 3B-3D). Participants were extensively trained on the task to avoid potential reporting confounds with respect to usual chronic pain (Methods). Subregion LDA models demonstrated significant prediction of high versus low acute pain in the aff-ACUTE condition in two participants, only when trained on data including ACC (CP1 AUCLDA = 0.738, P= 0.045; CP2, AUCLDA = 0.74, P= 0.037; FIG. 3E and FIG. 15). Given unsuccessful decoding from any models trained on OFC alone, these data may suggest preferential involvement of ACC in circuits harboring acute pain signals. Tested models could not significantly predict acute pain in the unaff-ACUTE condition (FIG. 3F). In contrast to acute pain metrics on the affected side, we found no significant decoding of the actual delivered temperature on the affected side in any participant (FIG. 16). One participant (CP4) showed significant decoding of temperature when applied to the unaffected side.

[0186] Again, key neural features supporting acute pain-state prediction varied between the two participants in whom decoding was possible, with the highest magnitude feature weight coming from the ACC for each (CP1 contra ACC alpha, CP2 contra ACC beta; FIG. 3G). Notably, features supporting acute pain decoding visually differed from those supporting chronic pain (compare to FIG. 2E). If neural features that drive decoding of chronic pain state are similarly important for supporting acute pain decoding, full LDA models trained on the former should perform well when tested on data from the latter. We only observed such generalization from chronic to acute pain in one participant (CP2) who did not have a preexisting brain lesion (FIG. 17). The lack of such generalization between acute and chronic pain representation in the remaining three participants suggests distinct neural codes in the ACC and OFC in these participants.ACC and OFC distinguish acute and chronic pain states

[0187] The extent to which neural representations of chronic pain and acute pain states resemble one another is a major gap in knowledge. One way to gain additional insight into associated neurophysiologic mechanisms underlying chronic versus acute pain is to compare the importance of neural features driving prediction of each pain type.

[0188] We assessed the relative importance of OFC versus ACC by subtracting each normalized ACC feature weight from the corresponding normalized OFC feature weight to obtain a distribution of feature importance differences; this difference distribution was then compared to a permuted, null distribution to assess if activity from the OFC or ACC was more important to pain decoding than by chance (see ‘LDA feature importance’ and FIGS. 4A, 4B). Values greater than 0 indicate greater OFC importance. We observed that the OFC was more important than the ACC to chronic pain NRS decoding in a greater proportion of all features for three participants (OFC proportion of significant features: CP1 = 100%, CP3 = 75%, CP4 = 58%). Common for all participants, contralateral OFC delta power was consistently more important than contralateral ACC delta power to decoding chronic pain NRS (Wilcoxon rank-sum tes, z= 13-36.3, P= 10-6). In contrast, for acute pain decoding in two participants, the ACC appeared more important than the OFC for a greater proportion of all significant features when compared to chronic pain (FIG. 4B; acute versus chronic pain ACC proportion: CP1 = 100% versus 0%, CP2 = 64% versus 58%). Because feature importance for acute pain was only assessed in two participants showing successful acute pain decoding, caution is advised in interpretation due to low participant numbers.Neural time dynamics distinguish acute and chronic pain

[0189] We next sought to characterize the temporal dynamics of these power bands by analyzing patterns of power changes at short-term (seconds, within a recording) and long-term (diurnal, across recordings) scales. Chronic and acute pain decoding model features consisted of band-pass filtered power, averaged by time over the duration of the recording clip (30 s for chronic, 3 s for acute), which variably showed increases or decreases associated with predicting a ‘high pain’ state (FIGS. 2E and 3G).

[0190] It is possible that if a feature was important to ‘high pain’ decoding, power changes may be associated with either transient or sustained changes within specific frequency bands over time or some combination of the two. To distinguish these possibilities, we calculated the power time series of each feature over the duration of the recording clip and averaged among clips in the high versus low pain groups (FIG. 5A and FIG. 18). This analysis provides an impression of how average power associated with high or low pain fluctuates in real time. As expected, we observed that if a featureweight was positive or negative (FIGS. 2E and 3G), then that feature’s average power was respectively increased or decreased over time during high pain states, compared to low pain states (FIG. 5A and FIG. 18). For example, a negative-signed feature weight for contralateral OFC delta indicates that the contralateral OFC delta power was concordantly decreased during high pain states. Across each 30-s recording clip for chronic pain decoding, the total power amplitude maintained this concordant relationship 76% of the time (group mean ± s.d. = 22.8 ± 6.2 s; FIG. 5B) across all participants. Although we observed fluctuations in the power values over the 30-s duration, the top five power features were associated with sustained changes lasting a mean of 4.4 s (±8.8 s.d.) above or below the low pain-state average. We rarely observed transient bouts of fluctuation among chronic pain features, suggesting that temporal dynamics in decoding features reflected sustained oscillations relevant to ongoing perceived pain.

[0191] Despite a shorter 3-s recording clip duration for acute pain trials, we observed that power increases and decreases occupied a similar total proportion of recording clip time for both acute and chronic pain features (76.4% of total time acute versus 76% chronic; acute group mean ± s.d. = 2.1 s ± 0.3 s, rank sum P> 0.08; FIG. 5B). In contrast to chronic pain; however, the top five acute pain features showed more temporal variability with more transient bouts of fluctuation lasting an average of 0.56 s (s.d. = 0.18 s, rank sum P< 10-14; FIG. 5C). Notably, we rarely observed such bouts of increases or decreases approaching 3 s, which was the theoretical upper limit given acute recording duration of 3 s. Because these bouts of power increases or decreases occurred for a similar proportion of recording clip duration for chronic and acute pain data, we expected that the more transient bouts would occur more frequently for acute pain features. As expected, acute pain features showed many more frequent bouts of increases or decreases compared to chronic pain features (acute = 1 .3 ± 0.29 bouts per second versus chronic = 0.23 ± 0.18, rank sum P< 10-19; FIG. 5D). Therefore, compared to chronic pain, the top five acute ‘high pain’ biomarkers exhibited shorter, more frequent bursts of power changes, resulting in a similar total percentage of time with increases or decreases between high and low pain states.

[0192] To characterize longer timescale dynamics of chronic pain neural features, we performed a diurnal analysis by organizing both pain scores and neural features by time of day of report and resampling this diurnal pattern at a 3-h time resolution to model diurnal fluctuations (Methods, FIG. 19 and Supplementary Table 3). We did not observe significant correlations between the diurnal trends of neural features and of pain NRS for any feature, in any participant. Overall, chronic pain decoding was associated with sustained power changes over time, distinct from more transient bursts of power changes supporting acute pain decoding. Diurnal time dynamics of neural featuresdid not appear grossly related to diurnal fluctuations in pain metrics, suggesting a possible importance for short timescales in encoding subjective pain states.Discussion

[0193] Objective biomarkers of spontaneous chronic pain severity are requisite for understanding basic mechanisms of clinically relevant pain, diagnosing pain syndromes, prognosticating disease course and devising new therapies. Using first-in-human long-term, ambulatory, intracranial recordings in four participants with chronic, neuropathic pain, we used LFP signals from the ACC and OFC to predict various measures of chronic pain severity, with high sensitivity. While we could predict dichotomized pain state using two different methods, regardless of pain etiology, duration of symptoms or nervous system lesion, prediction of continuous metrics using regression was not as successful. Chronic pain biomarkers appeared stable over months in three of four participants and were more predictive of current pain state rather than recent transitions in pain report. In two participants, we could also predict acute, evoked thermal pain during an experimental heat pain task. We observed that chronic pain decoding tended to rely more on activity in the OFC (for example, contralateral OFC delta) as evidenced by OFC subregion models and greater OFC feature weights across participants. Acute pain decoding was supported more by ACC activity. Finally, the time course of neural activity supporting chronic pain decoding reflected sustained increases or decreases in power in the order of seconds, while acute pain decoding was associated with more frequent and transient changes in power. These results provide a proof of principle that signals from key neural hubs can be used to track clinically relevant chronic pain states in humans and have important implications for understanding circuit mechanisms underlying the chronification of pain.

[0194] While a complex neural network underlies the maintenance of chronic pain states1731, we found that activity from either ACC or OFC alone was sufficient to track pathological network activity underlying chronic pain fluctuation. Previous neuroimaging studies of chronic back pain established the importance of the ACC and medial prefrontal cortex3233, although notably failed to identify a role for the OFC, possibly due to sparse time resolution of up to 4 visits over 1 year. Consistent with prior studies of spontaneous, chronic back pain measured over seconds to minutes in a single visit, we observed that ACC activity in three participants could discriminate when spontaneous pain was stable versus fluctuating8 12. The present findings suggest that signals in the OFC can track current chronic pain severity for neuropathic pain syndromes such as central post-stroke pain (CPSP) or phantom limb pain. In the context of previous studies supporting a broad role of the OFC in reward, punishment and the placebo effect1825, OFC circuits may integrate pain expectation and context-dependent predictions that influence subjective pain evaluation, which may include rumination or engagement of coping mechanisms. Areas we indicate as OFC in this study also overlap with prior definitions of ventromedial prefrontal cortex, which has been associated with interindividual variability in pain perception and aversiveness23'24. In two participants with CPSP, decoding of pain unpleasantness was driven more by ACC activity, also consistent with prior roles of ACC circuits in affective14’34dimensions of pain processing. However, the preferential importance of ACC in acute pain decoding (which is described as having a smaller affective component than chronic pain states) challenges prior concepts of a medial pain pathway35from the medial thalamus to ACC as selective for affective processing11.

[0195] In contrast to the spontaneous and enduring phenomenology of chronic pain, acute thermal pain administered to patients reflects a transient, externally evoked pain experience. In two participants, acute, evoked pain was associated with distinct spatiotemporal activation patterns consisting of transient power fluctuations biased toward ACC. This observation is consistent with numerous imaging studies showing the ACC as a key node activated with experimental pain stimuli across individuals4'6’1223’36. Transient bursts of power among cingulate neurons may reflect facilitation of nociceptive hypersensitivity by recruitment of descending serotonergic projections to the spinal cord11or ascending information flow from the medial thalamus37. Although most participants exhibited thermal allodynia, the transient power change patterns underlying acute pain decoding in the current study were found in two participants (both women): one with severe allodynia and hyperpathia (CP1) and one with a normal neurological sensory exam (CP2), which may suggest preserved information processing mechanisms across disparate phenotypes. However, within- participant generalization of chronic to acute pain biomarkers was only found in one participant with phantom pain (CP2), without an ischemic brain lesion, suggesting that more robust models that directly incorporate acute versus chronic pain classes may potentially aid in discriminating pain syndrome subtypes or reflect cortical reorganization after a lesion33. As acute, evoked pain was only decodable when the painful stimulus was applied to the same body side as ongoing chronic pain, the absence of evidence of successful decoding should not be interpreted as evidence of absence, possibly due to sample size of participants, trial number or environmental context (for example, at- home versus in-clinic testing).

[0196] Global cerebral pain networks in all participants likely underwent rewiring over many years living with chronic pain32. Still, ongoing ‘background’ chronic pain may have influenced acute pain perception even in the unaffected body side (see Methods for testing precautions used). Rather, the inability to decode acute pain in the unaffected side may suggest that the ACC and OFC are eithersensitive to pain on one side of the body or perhaps more sensitive to representing pain in body regions where chronic pain is ongoing. It is possible that an acute pain signal may be detectable in the OFC if a task explicitly requires decision-making or pain anticipation components as seen in prior studies38. If the contributions of ascending versus descending pathways to acute versus chronic pain processing can be elucidated in future studies, this would help to explain the common clinical observation that the chronic pain experience is not simply a more enduring version of acute pain.

[0197] Classical models of the role of neural oscillations propose that frequency-specific activity serves to flexibly route information flow through networks to support selective attention39and working memory40. Oscillations within a particular brain region may represent modulation of postsynaptic excitability, which itself impacts input gain39. The present observation that chronic pain intensity was associated with sustained increases or decreases in OFC power may therefore reflect maladaptive plasticity or changes in excitation / i n h ibitio n41between the OFC and afferent primary sensory cortices, magnocellular cells of the medial thalamus or reciprocal connections with the ACC18to perpetuate pathological sensory integration. In contrast, transient burst-like oscillatory activity associated with acute pain intensity may reflect stochastic packets reflecting fast computations42between ascending medial thalamic inputs or descending rostral brainstem outputs supporting allodynia or evoked pain. Similar transient ‘high current spikes’ in rodent ACC have been observed to shift to more transient states in response to acute noxious stimuli and more sustained spiking states in a rodent model of post-stroke pain37. We further speculate that transient burst-like power changes may support aberrant sensory, affective and cognitive integration in chronic pain in a manner analogous to how pathological beta bursting influences motor processing in Parkinson’s disease43. Of note, because our decoder inputs consisted of time-averaged power values, it is unlikely that these disparate temporal dynamics were responsible for the lack of generalization between acute and chronic pain we observed here.

[0198] Although we could reliably predict multiple metrics of chronic pain severity when scores were dichotomized into ‘high’ and ‘low’ categories, prediction of exact pain scores with linear regression faired relatively poorer for metrics tracking pain intensity. One reason for this may be that subjective pain perception is psychophysically non-linear44, in a manner that binarization may better capture45. Binary classification yielded high sensitivities across participants, which is crucial for practical diagnostic or therapeutic biomarkers. Consistent with prior suggestions3, such a binary biomarker could be more useful for diagnosis or disease classification, while continuous measures or grading scales may be more useful for assessing treatment response. Future analysis of neural signatures that predict treatment response would lend further insights to basic pain mechanisms.

[0199] Three of four participants suffered from CPSP, due to ischemic brain injuries acquired greater than 2 years before data collection. Despite potential concern that a previous stroke may have induced brain plasticity specifically related to stroke symptoms, all participants had stable symptoms and physical exams over the study period. While OFC activity was sufficient to decode chronic pain in these participants despite varying locations of ischemic infarcts or stable use of various pain medications, in many cases activity from the ACC also performed as well. Given the small sample size of the present study, and idiosyncratic decoding observations in acute pain from 1 -2 participants, caution must be used to avoid overinterpretation. Further studies are required to establish greater confidence in the specificity of the OFC for chronic pain prediction across larger groups.

[0200] A further limitation of the current study is that rather than directly predicting pain state per se, it is possible that machine learning models may be predicting other variables strongly correlated with reported pain metrics such as arousal or attention. However, as the adaptive value of pain may be to alert the organism to impending tissue damage, arousal effects may be a fundamental component of the complex subjective perception of pain itself. Further, in the diurnal analyses, we found many nonsignificant correlation values with large effect size (r> 0.5). Because we did not explicitly quantify Bayes’ factors related to the null hypothesis, the absence of evidence should not be interpreted as evidence of absence for a particular effect, particularly in cases of nonsignificant model results where the effect size may be high.

[0201] The development of personalized pain biomarkers will be central to accurate diagnosis, tracking prognosis and for future therapeutic drug and device development. Individualized biomarkers are also critical to the growing field of adaptive neurostimulation, where a participant’s ongoing neural activity may be used to control therapeutic brain stimulation46. Personalized biomarkers of chronic pain state may be used as input signals in real time, to control the amplitude or frequency of therapeutic electrical brain stimulation for treating refractory pain as they have for movement disorders. By adaptively delivering intermittent bouts of stimulation or adjusting parameters based on pain-state biomarkers, it may be possible to mitigate known effects of adaptation and loss of efficacy demonstrated in nearly all open-loop continuous DBS paradigms. It remains to be seen if similar signals may be non-invasively recorded using electroencephalography to track disease state or inform drug development.MethodsGeneral trial methods

[0202] This current study was undertaken as part of an ongoing multiyear clinical trial aimed at developing closed-loop control algorithms for DBS to treat chronic neuropathic pain. The clinical trial protocol (NCT03029884) was approved by UCSF Human Research Protection Program (IRB) and the Food and Drug Administration (FDA).Inclusion and exclusion criteria were:

[0203] Inclusion criteria.• Age > 21 years• Clinical diagnosis of post-stroke pain (thalamic pain), spinal cord injury or phantom limb pain with allodynia or dysesthesia with pinprick anesthesia or hypoesthesia on the affected hemi-body or limb (anesthesia dolorosa).• For post-stroke pain: stroke of ischemic etiology only. MRI done within 1 year of surgery showing a lesion that involves the contralateral brainstem, thalamus or cortex. The lesion involves cortical- subcortical areas in topography consistent with sensory thalamocortical connections. This will include participants with infarcts in the territory of the middle cerebral artery. A more recent MRI may be required if the participant’s condition changed within the previous year.• For phantom limb pain: MRI done within 1 year does not show any contraindication to surgery such as mass, lesion, hemorrhage or other abnormality near target.• For spinal cord injury pain: MRI done within 1 year does not show contraindication to surgery such as mass, lesion, hemorrhage or other abnormality near target.• One year or more of medically refractory severe pain (see below).• Average daily pain for the past 30 d reported as >5 on a 0-10 NRS.• Failure to respond adequately to at least one antidepressant, one anti-seizure medication and one oral narcotic with current stable doses of medications.• Ability to speak / read English.• Capable of understanding and providing informed consent.• Stable doses of pain medications (for example, anticonvulsant drug, antidepressants and opioids) for at least 30 d.• Women of childbearing age must regularly use an accepted contraceptive method(s).

[0204] Exclusion criteria.• Pregnancy or breastfeeding.• Inability to speak and / or read English.• Inability to give informed consent.• Significant cognitive impairment or dementia (MoCA < 25).• Aphasia severe enough to limit the consent process or communication between the investigators and the participant. Participants with mild or recovering aphasia may be considered candidates at the discretion of the Principal Investigator (PI).• Recommendation of exclusion by evaluating psychiatrist based on comprehensive neuropsychological evaluation, which may include: active depression (Beck Depression Inventory > 20) or other untreated or uncontrolled psychiatric illness (active general anxiety disorder, schizophrenia, bipolar disorder, obsessive-compulsive disorder or personality disorders (for example, multiple personality disorder and borderline personality disorder).• Suicide attempt within the previous 12 months or imminent suicide risk.• History of substance abuse in the previous 3 years.• Major medical comorbidities increasing the risk of surgery including uncontrolled hypertension, severe diabetes, major organ system failure, history of hemorrhagic stroke, need for chronic anticoagulation other than aspirin, active infection and immunocompromised state or malignancy with <5 years life expectancy.• Inability to stop Coumadin (warfarin) or platelet anti-aggregation therapy for surgery and after surgery. Participants taking these medications will need to discuss the need / risk of continuing these medications with their physicians, and the PI or study personnel may contact the treating physician(s) as well to discuss the risks of anticoagulation / anti-aggregation therapy discontinuation.• Coagulopathy. Participants will be excluded unless assessed and cleared by hematology.• MRI (done within 1 year of surgery) showing significant abnormalities other than those associated with the neurological disorder causing chronic pain.• Implantable hardware not compatible with MRI or with the study.• Inability to comply with study follow-up visits.• Previous ablative intracranial surgery for the management of thalamic pain syndrome.• Previously implanted with DBS system or any previously implanted device treatment involving brain stimulation.• Major neurological disorder other than the one that led to the chronic pain including epilepsy, neurodegenerative condition or any history of seizure.• Requires diathermy, electroconvulsive therapy or transcranial magnetic stimulation to treat a chronic condition.• Has an implanted electronic device such as a neurostimulator, cardiac pacemaker or medication pump.• Allergies or known hypersensitivity to materials in the Activa PC + S system (that is, titanium, polyurethane, silicone, polyetherimide and stainless steel).• Pregnancy or lack of regular use of contraceptives. Participants who become pregnant after enrollment may be excluded from the study. Participants who become pregnant before the surgical implantation of the DBS systems will be excluded from the study.• Participants may be excluded from enrollment due to a condition that, in the judgment of the PI, significantly increases risk or reduces significantly the likelihood of benefit from brain stimulation.Participants

[0205] Four right-handed participants (CP1-4) with refractory neuropathic pain limited to one side of the body for >1 year were enrolled in an ongoing study to devise adaptive brain stimulation for pain. Three participants had post-stroke pain and one had phantom limb pain (FIG. 1 A and Supplementary Table 1 ). Half were women and the mean (±s.d.) age was 58.5 ± 3.1 years; all were free from untreated depression, recent substance use disorder and cognitive impairment. All participants maintained constant doses of medications for 3 months. Each participant signed written informed consent and the study was approved by both FDA under an Investigational Device Exemption and IICSF Committee on Human Research. Participants were reimbursed for travel and accommodation but not otherwise compensated.

[0206] Brief participant descriptions. See Supplementary Table 3 for task-related temperature calibration quantitatively reflecting allodynia.

[0207] CP1 . Woman aged 58 years old with a history of CPSP syndrome, after right middle cerebral artery infarct in 2013. Approximately 12 weeks after her stroke, she began having sharp, burning, stabbing pain in the left hemi-body from her head to foot, most severely concentrated over the left hand / arm and leg. A pain-focused neurological exam was notable for hemiparesis in the left arm and leg, with reduced sensation to light touch and temperature along the whole left arm, ankle and foot.She experienced dynamic mechanical, cold and heat allodynia with symptoms of hyperpathia involving the spatial spread of lingering pain in response to rapidly repeated stimuli (interval <3 s).

[0208] CP2. Woman aged 63 years old with elective above-the-knee amputation of the right leg in 2016 (for pain) and refractory phantom limb pain with right leg, calf and foot pain involving burning and pressure sensation. Her leg pain began 4.5 years before enrollment resulting from compartment syndrome as a complication from total knee arthroplasty. A pain-focused neurological exam was normal, with no stump pain nor evoked sensations on palpation of the stump (negative Tinel’s sign). She did not have sensory loss, weakness or allodynia.

[0209] CP3. Man aged 56 years old with history of hypertension and left thalamic infarct in 2014, with persistent CPSP involving the right-sided face, arm and leg. Beginning 5 d after stroke, he developed severe burning pain associated with dysesthesias of the right bicep, forearm and leg. A pain-focused neurological exam showed intact motor and sensory function with notable hyperalgesia and allodynia in the right arm and leg. Allodynia was confined to cold and heat allodynia, with minimal mechanical allodynia.

[0210] CP4. Man aged 59 years old with history of prostate cancer (no current disease), a solitary episode of atrial fibrillation and CPSP after left posterior cerebral artery infarct in April 2017. Approximately 6 weeks after his stroke, he began having burning and stabbing pain in the right arm as well as the right ankle which progressed to involve the entire right hemi-body after 1 year. A pain- focused neurological exam was notable for mild hemiparesis in the right arm, with intact sensation to light touch and temperature all over. There was dynamic mechanical, cold and heat allodynia in the right arm (worst distal to the elbow circumferentially) and leg (worst below mid-calf on the right including whole ankle and foot).Device implant surgery

[0211] All participants underwent chronic intracranial electrode and neurostimulator implant targeting the ACC and OFC using Medtronic electrodes connected to the Activa PC + S device26 (bilateral in three participants, unilateral in CP1 contralateral to side of pain). Multielectrode leads targeted the dorsal ACC (models 3389 / 3387) and subdural surface of the OFC (paddle nos. 3587A / 0930A) using frame-based stereotaxy under general anesthesia. Individual trajectories weredetermined using neurosurgical mapping software from BrainLab and Medtronic Stealth 8.0 based on preoperative 3T MRI scans (FIG. 6).

[0212] For the dorsal ACC, target locations for each electrode tip were determined by computing diffusion tractography images (55 directions) and identifying the cingulum bundle in a region 20- 25 mm posterior to the anterior border of the lateral ventricle (FIG. 6 and Supplementary Methods) as in a previous study34. For the OFC, a burr hole was made through the temporal bone to allow passage of the paddle lead to the subdural surface. We targeted both medial and lateral OFC, areas that overlap with definitions of the ventrolateral medial prefrontal cortex in other studies. Electrodes were advanced through frontal burr holes, with correct placement confirmed by intraoperative computed tomography (CT) and reconfirmed by CT at the 2-month visit. Stim-loc devices were used for electrode lead anchoring, with a strain relief loop anterior to the site of anchoring to prevent movement or torque on the lead over time. The free ends of the leads were then placed in a subgaleal pocket, created by blunt dissection in the parietal area. A lead extender was tunneled subcutaneously between the parietal incision and a 5-cm incision over the pectoralis muscle, where a pocket was bluntly dissected for the Activa implantable neurostimulator system. A small amount of Medtronic medical glue was applied to the junction of the lead and each barrel of the implantable neurostimulator to mitigate electrical cardiac artifacts. Participants had at least 10 d of recovery before neural data collection began.Electrode localization and brain imaging

[0213] Electrode (depth and paddle electrocorticography (ECoG)) contact locations were reconstructed using a CT scan taken 2-3 months after implantation and fused with the preoperative T1 -weighted volumetric MRI. The T1 MRI was used to construct cortical surface models in FreeSurfer (version 7.1 .1 )4748. We projected ECoG contacts onto the cortical surface mesh with the imgpipe toolbox49using a surface vector projection method50. Once we identified locations for each ECoG and depth electrode, we projected all participants’ electrodes onto Montreal Neurological Institute 152 space for plotting and comparison.

[0214] In addition to the stereotactic preoperative MRI scans, high angular resolution diffusion imaging was acquired on the same 3 Tesla MR scanner (General Electric), using a spin-echo echo- planar imaging pulse sequence (TE = 71 ms, TR = 7,765 ms, flip angle alpha = 90°), field of view 28 x 28 cm, at least 70 axial slices, 2 mm3isotropic voxels, b value = 2,000 s mm-2in 55 non-collinear gradient directions and a signal-to-noise ratio >60. A single non-diffusion-weighted bO image was also obtained. The diffusion-weighted tractography was processed using BrainLab software.Preoperative MRI, postoperative CT and high angular resolution diffusion imaging scans were automatically merged and corrected for distortion. The contacts used for recordings were segmented on the CT scan based on visible contacts and known spacing measurements. The regions of interest were created using manual tracing of bilateral dorsal ACC, using a fractional anisotropy threshold between 0.2 and 0.3 and a minimum length of at least 8cm.Ambulatory data collection

[0215] We collected neural recordings coincident with pain reports multiple times daily (mean 3.5 per day) from morning to evening. The Activa PC + S system is an investigational, implantable ‘bidirectional’ neural interface, that can sense and store patient-triggered neural recordings (LFP time series) that can be analyzed offline or used for adaptive DBS26. Participants initiated a 30-s neural recording using their remote control while sitting quietly, immediately following a pain report. Each 30-s neural recording was self-triggered by participants by pressing a button on their remote control. Participants were instructed to space out recordings at least three times daily, once each in the morning, afternoon and evening. In rare instances (<10% of all recordings), devices were programmed to automatically capture a 30-s recording at prespecified times throughout the day, to increase recording frequency. In such cases, we attempt to vary specific times of recordings every 2 weeks to more equally sample morning, afternoon and evening periods. Also, in these cases participants were notified of the imminent start of a recording via text message and prompted to submit a pain report via text message or by clicking a hyperlink to a REDCap survey page. Neural recordings were included in analysis only if they occurred within 15 min of reported scores.

[0216] Pain metrics were reported using mobile text message or the online REDCap platform version 10.0.33 (REDCap, Vanderbilt). Reported metrics included pain intensity NRS (0-10) and VAS (where 0 is no pain and 10 is the worst pain they have ever experienced); pain unpleasantness NRS and VAS (where 0 is not bothersome at all and 10 is most bothersome ever felt) and the SF-MPQ (version 2)28. Although VAS and NRS may provide redundant information, both were collected when possible to assess reporting consistency within participants (FIG. 1 C). While data from all metrics were not available from all participants, pain intensity NRS was reported by all and so prioritized in the study.

[0217] Participants provided pain reports at least 2 weeks before surgery, and reported score means did not change after surgery. To standardize reporting, participants were trained to report pain intensity NRS values based on a narrative description of each number (0-10) from a comparative pain severity scale table51 . VAS sliders contained only the numbers 0 or 10 at each end and were presented horizontally with the slider bar starting at a value of 5 / 10. The SF-MPQ2 is amultidimensional survey tool that requires participants to rate the extent (none, mild, moderate, severe) to which each of 15 pain words matches their symptoms. We reported the SF-MPQ2 score as the sum of all verbal descriptor ratings (maximum = 45). The magnitude by which pain scores differed between the high and low group within each participant was confirmed to be clinically significant and surpassed a meaningful threshold for each participant. All participants confirmed that individual successive numbers in reported ranges carried discernable clinical meaning, and so an NRS difference of 1 of 10 was therefore noted separately by each participant to have different clinical meaning with reliability over repeated reports (FIG. 1 D). Brain recordings commenced immediately after pain reports were submitted.Activa PC + S neural recordings and preprocessing

[0218] Each recording sampled two bipolar referenced time-series channels capturing LFPs at a 422-Hz sampling rate. The contacts used for sampling were chosen based on those showing the largest root-mean-square voltage in the raw tracings as determined by investigators (for example, for the ACC we chose 0-3+ as one possible bipolar channel, where 0 was the deepest contact). Data download of collected recordings was performed every 2-3 weeks by either participants or investigators using the Medtronic sensing programmer (version 1.5.0.0) via a wand resting on the skin over the Activa device.

[0219] Data analysis used MATLAB 2022b (MathWorks) and Python 3.7.4 with the Scikit-learn library (version 1.0.2). To identify pain-state biomarkers, LFPs were high-pass filtered above 1 Hz and low-pass filtered below 100 Hz, converted to frequency domain using multitaper spectral estimation52, log 10 transformed and z-scored within frequency band. Standardized power values for each 30-s recording were averaged within canonical frequency bands: 5 (1 -4 Hz), 0 (4-8 Hz), a (8- 12 Hz), [3 (12-30 Hz), low y (30-70 Hz) and high y (70-100 Hz). Bandwidths containing idiosyncratic noise sources (known from Activa PC + S by the manufacturer) were visually identified for each participant (by an investigator blinded to results) and excluded from power averages. Outlier recordings were excluded from analysis if power within three or more frequency bands (of six in total) contained power averages greater than three scaled median absolute deviations from the median, in a manner blinded to the results.Acute / experimental thermal pain protocol

[0220] We applied transient heat stimuli of varying intensity to each participant’s most painful body part (and contralateral side) using the Medoc TSA 2001 while recording brain LFP activity. Heattemperatures were calibrated separately for each body side and participant, such that reported NRS values ranged from 0 to 9 out of 10. Background chronic pain NRSs were collected before starting acute pain trials. Participants underwent five trials at each of five intensities, where temperature first increased at a rate of 1 °C s-1 from a baseline of 32 °C, was held at target intensity for 3 s, and then returned to baseline. Around 10 to 15 s elapsed between trials, and participants were cued to trial start, although 3 s of neural data was only included after target temperatures were reached. Target temperatures ranged from 34 to 48 °C. Participants verbally reported pain NRSs 3 s after reaching target temperature. Trials with movement or uncertain pain report were discarded.

[0221] To mitigate the possibility that participants may ‘merge’ or conflate ratings of chronic and acute pain (even subconsciously), we took three precautionary steps: (1 ) Participants were exposed to at least four different blocks of the acute pain paradigm before formal testing, so that they became accustomed to the pacing and attention to acute, thermal pain phenomenology. (2) Thermal pain was applied adjacent to, instead of directly over, anatomical regions where background chronic pain was experienced, tailored to each participant. This was done to avoid confounding spontaneous underlying pain in the same body part that was in physical contact with the thermode. (3) Participants were explicitly trained to provide only the pain score corresponding in time (by verbal cueing at the 3-s plateau time point) and in space (in only the anatomical area contacted by the thermode). To specifically train participants to understand this difference, they were given visual feedback of the temperature as it ramped to target temperature during training blocks, so that they could understand the relationship between stimulus delivery and perceived sensation. However, to avoid bias, they were not shown the visual temperature during actual testing blocks.Pain decoding models and biomarker analysis

[0222] Separate models were constructed for each participant and each reported pain score, using as input features the standardized mean spectral power in each frequency band, per brain region, per hemisphere (that is, CP1 model used 12 features (6 each from unilateral implants in the ACC and OFC); CP2-4 models contained 24 features from bihemispheric recordings). We computed multivariate regularized regression (LASSO) on continuous pain metrics and two types of classification on dichotomized pain metrics (LDA and LSSM). For classification, binary response variables were pain scores dichotomized into ‘high pain state’ and ‘low pain state’ categories for each participant, based on dividing scores above and below that participant’s median value (test data point was not included in the median calculation for full independence; see ‘Cross-validation’). To ensure robust decoding of pain state, we used independent machine learning approaches: LDA, LSSM (FIG.7) and LASSO regression. While LDA produces directly interpretable feature weights that can be programmed into currently available adaptive DBS devices, LSSM is a dynamical systems approach that can more flexibly represent complex brain activity53.

[0223] The approaches below were used to predict multiple pain metric responses that were available for each participant (Supplementary Table 1 ): All participants: chronic pain intensity NRS, acute pain verbal NRS, acute pain temperature; CP3 and CP4: pain intensity VAS, pain unpleasantness NRS / VAS, SF-MPQ2.

[0224] LASSO regression. In addition to predicting binarized chronic and acute pain states, we also trained multivariate LASSO regression models to predict continuous pain scores. LASSO regression models implicitly perform variable selection through regularization whereby the weighting factors of multicollinear or extraneous input variables are shrunk, thereby having less influence on the outcome prediction. Similarly to LDA models, we stratified our regression analyses by constructing models trained on data separately from each available subregion. All models were fit with a regularization parameter of 0.001 and validated with our modified LOOCV. Model fit was evaluated by calculating the coefficient of determination (R2) values of the models, which can range from below 0 to 1 , and root mean square error.

[0225] LDA decoding. LDA is a common supervised classification method used to categorize labeled samples into two or more classes by determining an optimal decision boundary that best separates samples from each class54. LDA makes predictions by estimating the probability that each input belongs to a particular class based on variance in the input features.

[0226] We used LDA to classify high versus low dichotomized pain states for two main reasons. First, LDA produces an interpretable set of feature weights for each trained model, permitting inference of the importance of specific neural features (that is, brain regions, hemisphere, power bands)54. Second, the Activa PC + S, Summit RC + S and Percept sensing-enabled brain stimulation devices (Medtronic) exclusively use device-embedded LDA algorithms for ambulatory closed-loop control55,56. Results from LDA-based decoding can therefore be directly implemented for closed- loop control in current clinical devices and for future stimulation studies.

[0227] We constructed multiple LDA classification models using data sub-selected from different brain regions and / or hemispheres, which we called subregion models (for example, right hemisphere ACC and OFC, left ACC and left OFC). One model consisting of all available data (for example, both hemispheres and brain regions) was termed the full model (that is, contralateral ACC / OFC for CP1and bilateral ACC / OFC for CP2-4 represent the full model). Separate full and subregion models were constructed to classify each available pain metric for each participant (that is, pain intensity NRS and VAS, pain unpleasantness NRS and VAS and SF-MPQ2)28. However, pain intensity NRS was reported by all participants and so is emphasized in the main study. For the acute pain data, we also sought to predict the delivered temperature as a response variable. For all participants, we optimized the shrinkage parameter used for discriminant analysis, which was set to the default value for all participants except for CP1 for whom it was set to a value of 1 , indicating maximum shrinkage. Shrinkage is a form of regularization used to improve model estimation in scenarios when the number of training samples is low relative to input features57. Setting shrinkage to a value of 1 effectively maximizes regularization. All other hyperparameters were set to default values (for example, prior probabilities = 0.5 for each class) in Python 3.7.4 with the Scikit-learn library (version 1 .0.2)57.

[0228] AUG was calculated using the predictions from all test points compared to the ground truth pain states for all models. We also calculated sensitivity, specificity, positive predictive value and accuracy of binary classification.

[0229] To determine whether neural features correlated with chronic pain may generalize to acute (experimental) pain, we used full models trained on the chronic pain dataset to classify acute pain scores from the experimental thermal pain dataset, and vice versa (FIG. 4A). Specifically, for this acute / chronic generalization analysis, a particular participant’s neural features and pain reports from the chronic pain dataset were used as the training set, while that same participant’s neural features and pain reports from the acute, experimental thermal pain dataset were used as the test set. The same analysis was done by reversing the test and training sets for completeness. Neural input features and pain score response variables were standardized by subtracting the mean and dividing by the standard deviation for each model.

[0230] LSSM decoding. Neural decoding of pain may benefit from dynamic models that can predict how neural dynamics lead to pain score variations over time. Such an approach was previously used to decode interindividual mood state in patients with epilepsy and can reduce the dimensionality of neural features to fewer ‘latent states’, which can then be used to guide stimulation in a potential future control scheme (FIG. 7). We thus independently assessed the feasibility of decoding chronic pain states using dynamic LSSM53.

[0231] In this case, in nonoverlapping 1-s windows, we computed the log of signal power within the six canonical frequency bands (Methods) resulting in power time series with 30 time-steps during each 30-s recording. We modeled the dynamics of the power time series as the outputs of an LSSMwith a latent state and used the subspace identification algorithm and Akaike’s information criterion to learn the LSSM parameters and determine its latent state dimension53. Given the learned LSSM, we then applied the associated Kalman filter to the power time series in each recording to extract the latent state time series and decoded the pain state from the latent state at the end of each recording. Decoding was evaluated as above with a modified LOOCV and performance was quantified by computing the AUC. As LSSM methods may predict neural dynamics over time, they could hold promise for informing the closed-loop adaptive DBS system58, although present technology relies on LDA approaches2646.Cross-validation

[0232] To mitigate model overfitting, we used LOOCV in which one test sample was iteratively omitted from the training set (see Supplementary Table 1 for the number of recordings for each participant). Specifically, for all models, we predicted each test data point individually using a model trained on all other points. For all models, the median value used for dichotomizing pain scores was computed using only the training dataset (excluding the test data sample) to maintain strict independence between training and testing sets. Decoding AUCs were averaged across all runs to compute the model average AUC reported in the text and figures.

[0233] It is possible that autocorrelations in the pain score metric time series could result in potentially overfitted models and inflate prediction accuracy. Therefore, we performed a separate ‘modified cross-validation’ procedure, which omitted ±3 samples around the test point, from the training data. We chose a window of three samples based on the average number of pain scores reported per day across participants (3.2 reports per day). For participants CP2-4, the ‘modified cross-validation’ method resulted in significant decoding of pain intensity NRS for all subregion models containing the ACC or OFC from the contralateral hemisphere (6 of 9 models per participant). For the one participant with the smallest number of data samples (CP1 ), this method failed to produce significant classification of pain intensity NRS, due to the omission of substantially more training data than the standard LOOCV method. Consistent with shown results for the two participants with pain unpleasantness NRS reporting, the ‘modified cross-validation’ method resulted in significant decoding when the ACC was included in subregion models (CP3: 7 of 9 models; CP4: 6 of 9 models). For SF-MPQ2, all subregion models were significant for CP3, and 7 of 9 models were significant for CP4 (excluding those containing ipsilateral OFC).Sequential pain difference / pain fluctuation analysis

[0234] Predicting changes in successive pain scores versus state (FIGS. 14A, 14B): We were interested in understanding whether the neural features tracked fluctuations or differences between sequential pain ratings (instead of the actual reported individual pain scores / states). Pain score differences were calculated by subtracting the immediately preceding pain score from the current one to generate a metric of whether pain state recently increased or decreased. Therefore, pain scores that were previously higher than the current one would result in negative difference scores, while those that were previously lower would result in positive difference scores. To assess whether neural features discriminated changes in pain, we generated binary LDA classification models trained to distinguish stable pain scores (0 change in pain NRS) versus fluctuating pain scores (nonzero changes in pain NRS). We also built regression models trained to predict the change from the prior pain score to the current score. We performed the same subregion model stratification and validation methods as described for other models.

[0235] Decoding stability (70 / 30 split). To test whether the important neural features were stable across the ambulatory data collection period, we additionally performed a 70 / 30 train / test split validation protocol using an LDA classifier. For each participant, we trained an LDA model on the first 70% of their collected data and tested the model on the remaining 30%. We then calculated the AUG on the test set. If AUG from this analysis is comparable to the AUG computed using modified LOOCV, this suggests that neural features important for decoding are stable over time.

[0236] LDA feature importance. Model feature importance (FIGS. 2E and 3G) was calculated by taking the mean value of each feature coefficient weight across all cross-validation runs (using LDA) and linearly rescaling the resulting values between -1 and 1 for each feature and participant. The feature importance values were visualized by plotting colored heat maps ranging from blue to red to compare across participants.

[0237] OFC versus ACC feature comparison. To further assess the relative contribution of OFC versus ACC features on chronic and acute pain decoding, we computed the difference between the absolute magnitudes of each power band feature from the OFC and ACC in the same brain hemisphere (OFC minus ACC) for all recording clips. Values greater than 0 indicate greater OFC than ACC feature importance. To compare the distributions of these feature importance differences to a null hypothesis distribution, we generated null permuted distributions by randomly shuffling theOFC / ACC labels and recalculating feature differences 1 ,000 times. Real versus shuffled feature difference distributions were compared using two-tailed Wilcoxon rank-sum tests. Multiplecomparisons correction was performed across all features within each participant using false discovery rate correction as below.Temporal feature analysis

[0238] (1 ) To assess the time-varying characteristics of power features supporting pain-state decoding, we first computed multitaper spectrograms of neural recordings from each brain region52. Spectrograms were computed on 500-ms windows, with a 50-ms window step, TW = 3 (1.5-Hz resolution) and five tapers, with signals zero padded to the next power of 2 number of samples. We then averaged power values within canonical frequency bands as above (that is, delta, theta, alpha, and so on) to obtain power time-series plots for each feature from each brain region. To understand the relative changes in power that underlie decoding of high versus low pain states, we next averaged the feature power time-series clips belonging to each high versus low pain-state group (FIG. 5A and FIG. 18A). The same method was carried out for the acute-affected pain features for the two participants that had successful acute pain decoding (FIG. 18B). Note that because the 30- s neural recording clips were collected at various times of day within 5 min preceding a pain report, there is no well-defined task onset / offset for each 30-s recording as there is for each 3-s recording clip in the acute pain task. Therefore, one may expect to see reliable increases / decreases in power if there were sustained power changes across this time. However, one might not see any difference between low versus high pain power curves if power only fluctuated in transient bursts because the timing of such bursts may cancel out when averaged across sessions.

[0239] To quantify increases or decreases in power change patterns for chronic and acute pain features, we computed three metrics (FIGS. 5B-5D): (1) the number of bouts per second. A bout was defined as an event where the higher power curve crossed above the low power curve before crossing below if at all (for example, if a feature had a negative feature weight sign, then low pain state was associated with higher power than high pain state for that feature). Instances where the higher power curve did not intersect with the lower power curve were counted as one bout. We divided the number of bouts by clip duration to obtain bouts per second. (2) Mean bout duration was computed by taking an average duration of all detected bouts for that feature. (3) The total proportion of recording clip time where the higher curve was sustained above the lower curve (that is, equivalent to summing all bout durations and dividing by clip duration).

[0240] (2) To investigate diurnal patterns in the time series of each chronic pain neural feature, we separately plotted pain NRS and each feature’s clip averaged power as a function of time of day from midnight to midnight (FIG. 19). Because the time of day was non-uniformly sampled by each participant in the ambulatory setting, we resampled these diurnal plots with shape-preserving piecewise cubic interpolation (‘pchip’) using a uniform time resolution of 3 h to obtain a summary diurnal trend line for each feature. We chose a resolution of 3 h because this was the shortest time interval across each participant’s reports, and because doing so did not introduce visible artifactual trends. The resampled trend lines for each feature were then correlated with the pain NRS trend line using Pearson’s correlation, within participants, to assess for correlated patterns between neural features and pain reports (Supplementary Table 3).General statistical methods

[0241] False discovery rate correction for multiple comparisons used the Benjamini-Hochberg- Yekutieli procedure59.

[0242] Autocorrelation. Sample partial autocorrelations were calculated on reported pain intensity NRS using default parameters in MATLAB60. Because pain NRS was reported on an irregular timescale for each participant, we uniformly resampled the original pain NRS time series using shape-preserving piecewise cubic interpolation, rounded to the nearest 6 min (that is, 1 / 10 of 1 h).

[0243] Permutation tests: empirical P values. All machine learning model results (ALICs for LDA and LSSM and coefficient of determination R for regression) were compared against 5,000 (n) permuted results from null models (AUCs or RF) obtained by randomly shuffling the true response class labels or values (for example, randomly shuffling high or low pain class label) to obtain a distribution of chance level performance. We conservatively calculated empirical one-sided Rvalues by comparing the number of randomly permuted AUCs or fi^k) greater than the actual model AUC or FF using the following correction to avoid underestimation as previously demonstrated61: Pcorr= k -i- 1 / n + 1 . Note that because FF values can be <0, it is possible for very low FF values to sometimes be significant, if the FF results from shuffled / permuted, null models were largely negative.References

[0244] 1. Wide-Ranging Online Data for Epidemiologic Research (WONDER) (Centers forDisease Control and Prevention, 2016); wonder.cdc.gov /

[0245] 2. FDA-NIH Biomarker Working Group. BEST (Biomarkers, Endpoints, and other Tools)Resource (FDA, 2016).

[0246] 3. Davis, K. D. et al. Discovery and validation of biomarkers to aid the development of safe and effective pain therapeutics: challenges and opportunities. Nat. Rev. Neurol. 16, 381 -400 (2020).

[0247] 4. Coghill, R. C., McHaffie, J. G. & Yen, Y.-F. Neural correlates of interindividual differences in the subjective experience of pain. Proc. Natl Acad. Sci. USA 100, 8538-8542 (2003).

[0248] 5. Hutchison, W. D., Davis, K. D., Lozano, A. M., Tasker, R. R. & Dostrovsky, J. O. Pain- related neurons in the human cingulate cortex. Nat. Neurosci. 2, 403-405 (1999).

[0249] 6. Wager, T. D. et al. An fMRI-based neurologic signature of physical pain. N. Engl. J.Med. 3Q8, 1388-1397 (2013).

[0250] 7. Ploner, M., Sorg, C. & Gross, J. Brain rhythms of pain. Trends Cogn. Sci. 21 , 100-110(2017).

[0251] 8. Baliki, M. N. et al. Chronic pain and the emotional brain: specific brain activity associated with spontaneous fluctuations of intensity of chronic back pain. J. Neurosci. 26, 12165-12173 (2006).

[0252] 9. Baliki, M. N. et al. Corticostriatal functional connectivity predicts transition to chronic back pain. Nat. Neurosci. 15, 11 17-1119 (2012).

[0253] 10. Corder, G. et al. An amygdalar neural ensemble that encodes the unpleasantness of pain. Science 363, 276-281 (2019).

[0254] 11. Tan, L. L. et al. A pathway from midcingulate cortex to posterior insula gates nociceptive hypersensitivity. Nat. Neurosci. 20, 1591 -1601 (2017).

[0255] 12. Lee, J. -J. et al. A neuroimaging biomarker for sustained experimental and clinical pain.Nat. Med. 27, 174-182 (2021 ).

[0256] 13. Lee, M. et al. Activation of corticostriatal circuitry relieves chronic neuropathic pain. J.Neurosci. 35, 5247-5259 (2015).

[0257] 14. Rainville, P., Duncan, G. H., Price, D. D., Carrier, B. & Bushnell, M. C. Pain affect encoded in human anterior cingulate but not somatosensory cortex. Science 277, 968-971 (1997).

[0258] 15. Boccard, S. G. J. et al. Long-term results of deep brain stimulation of the anterior cingulate cortex for neuropathic pain. World Neurosurg. 106, 625-637 (2017).

[0259] 16. Kringelbach, M. L. & Rolls, E. T. The functional neuroanatomy of the human orbitofrontal cortex: evidence from neuroimaging and neuropsychology. Prog. Neurobiol. 72, 341 -372 (2004).

[0260] 17. Kucyi, A. & Davis, K. D. The dynamic pain connectome. Trends Neurosci. 38, 86-95(2015).

[0261] 18. Kringelbach, M. L. The human orbitofrontal cortex: linking reward to hedonic experience.Nat. Rev. Neurosci. 6, 691 -702 (2005).

[0262] 19. Damasio, A. R. The somatic marker hypothesis and the possible functions of the prefrontal cortex. Phils. Trans. R. Soc. 8351 , 1413-1420 (1996).

[0263] 20. Becker, S., Gandhi, W., Pomares, F., Wager, T. D. & Schweinhardt, P. Orbitofrontal cortex mediates pain inhibition by monetary reward. Soc. Cogn. Affect Neurosci. 12, 651 -661 (2017).

[0264] 21 . Rich, E. L. & Wallis, J. D. Decoding subjective decisions from orbitofrontal cortex. Nat.Neurosci. 19, 973-980 (2016).

[0265] 22. Rao, V. R. et al. Direct electrical stimulation of lateral orbitofrontal cortex acutely improves mood in individuals with symptoms of depression. Curr. Biol. 28, 3893-3902. e4 (2018).

[0266] 23. Kohoutova, L. et al. Individual variability in brain representations of pain. Nat. Neurosci.25, 749-759 (2022).

[0267] 24. Ceko, M., Kragel, P. A., Woo, C.-W., Lopez-Sola, M. & Wager, T. D. Common and stimulus-type-specific brain representations of negative affect. Nat. Neurosci. 25, 760-770 (2022).

[0268] 25. Petrovic, P., Kalso, E., Petersson, K. M. & Ingvar, M. Placebo and opioid analgesiaimaging a shared neuronal network. Science 295, 1737-1740 (2002).

[0269] 26. Swann, N. C. et al. Chronic multisite brain recordings from a totally implantable bidirectional neural interface: experience in 5 patients with Parkinson’s disease. J. Neurosurg. 128, 605-616 (2017).

[0270] 27. Giordano, J. J. et al. Proceedings of the fourth annual deep brain stimulation think tank: a review of emerging issues and technologies. Front Integr. Neurosci. 10, 38 (2016).

[0271] 28. Strand, L. I., Ljunggren, A. E., Bogen, B., Ask, T. & Johnsen, T. B. The short-form McGill pain questionnaire as an outcome measure: test-retest reliability and responsiveness to change. Eur. J. Pain. 12, 917-925 (2008).

[0272] 29. Hawker, G. A., Mian, S., Kendzerska, T. & French, M. Measures of adult pain: visual analog scale for pain (VAS pain), numeric rating scale for pain (NRS pain), McGill pain questionnaire (MPQ), short-form McGill pain questionnaire (SF-MPQ), chronic pain grade scale (CPGS), short form-36 bodily pain scale (SF-36 BPS), and measure of intermittent and constant osteoarthritis pain (ICOAP). Arthritis Care Res. 63, S240-S252 (2011 ).

[0273] 30. Salgado, J. F. Transforming the area under the normal curve (AUC) into Cohen’s d,Pearson’s r pb, odds ratio, and natural log odds ratio: two conversion tables. Psy. Intervention 10, 35-47 (2018).

[0274] 31 . T racey, I. & Mantyh, P. W. The cerebral signature for pain perception and its modulation.Neuron 55, 377-391 (2007).

[0275] 32. Hashmi, J. A. et al. Shape shifting pain: chronification of back pain shifts brain representation from nociceptive to emotional circuits. Brain 136, 2751 -2768 (2013).

[0276] 33. Reckziegel, D. et al. Deconstructing biomarkers for chronic pain: context- and hypothesis-dependent biomarker types in relation to chronic pain. Pain 160, S37-S48 (2019).

[0277] 34. Boccard, S. G. J. et al. Targeting the affective component of chronic pain: a case series of deep brain stimulation of the anterior cingulate cortex. Neurosurgery 74, 628-637 (2014).

[0278] 35. Kulkarni, B. et al. Attention to pain localization and unpleasantness discriminates the functions of the medial and lateral pain systems. Eur. J. Neurosci. 21 , 3133-3142 (2005).

[0279] 36. Derbyshire, S. W. G. et al. Cerebral responses to noxious thermal stimulation in chronic low back pain patients and normal controls. Neuroimage 16, 158-168 (2002).

[0280] 37. Shih, H.-C., Yang, J.-W., Lee, C.-M. & Shyu, B.-C. Spontaneous cingulate high-current spikes signal normal and pathological pain states. J. Neurosci. 39, 5128-5142 (2019).

[0281] 38. Brodersen, K. H. et al. Decoding the perception of pain from fMRI using multivariate pattern analysis. Neuroimage 63, 1162-1170 (2012).

[0282] 39. Fries, P. Rhythms for cognition: communication through coherence. Neuron 88, 220-235(2015).

[0283] 40. Colgin, L. L. et al. Frequency of gamma oscillations routes flow of information in the hippocampus. Nature 462, 353-357 (2009).

[0284] 41. Wang, X.-J. Neurophysiological and computational principles of cortical rhythms in cognition. Physiol. Rev. 90, 1195-1268 (2010).

[0285] 42. van Ede, F., Quinn, A. J., Woolrich, M. W. & Nobre, A. C. Neural oscillations: sustained rhythms or transient burst-events? Trends Neurosci. 41 , 415-417 (2018).

[0286] 43. Feingold, J., Gibson, D. J., DePasquale, B. & Graybiel, A. M. Bursts of beta oscillation differentiate postperformance activity in the striatum and motor cortex of monkeys performing movement tasks. Proc. Natl Acad. Sci. USA 112, 13687-13692 (2015).

[0287] 44. Foss, J. M., Apkarian, A. V. & Chialvo, D. R. Dynamics of pain: fractal dimension of temporal variability of spontaneous pain differentiates between pain states. J. Neurophysiol. 95, 730- 736 (2006).

[0288] 45. Rothaug, J., Weiss, T. & Meissner, W. How simple can it get? Measuring pain with NRS items or binary items. Clin. J. Pain. 29, 224-232 (2013).

[0289] 46. Gilron, R. et al. Long-term wireless streaming of neural recordings for circuit discovery and adaptive stimulation in individuals with Parkinson’s disease. Nat. Biotechnol. 39, 1078-1085 (2021).

[0290] 47. Dale, A. M., Fischl, B. & Sereno, M. I. Cortical surface-based analysis: I. segmentation and surface reconstruction. Neuroimage 9, 179-194 (1999).

[0291] 48. Fischl, B. et al. Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron 33, 341 -355 (2002).

[0292] 49. Hamilton, L. S., Chang, D. L., Lee, M. B. & Chang, E. F.

[0293] Semi-automated anatomical labeling and inter-subject warping of high-density intracranial recording electrodes in electrocorticography. Front. Neuroinform. 11 , 62 (2017).

[0294] 50. Kubanek, J. & Schalk, G. NeuralAct: a tool to visualize electrocortical (ECoG) activity on a three-dimensional model of the cortex. Neuroinform 13, 167-174 (2015).

[0295] 51. Rich, A. Comparative pain scale. The Cluster Headache Support Group clusterheadachewarriors.org / wp-content / uploads / 2017 / 03 / 0-10_Pain_Scale.pdf (2014).

[0296] 52. Bokil, H., Andrews, P., Kulkarni, J. E., Mehta, S. & Mitra, P. Chronux: a platform for analyzing neural signals. J. Neurosci. Methods 192, 146-151 (2010).

[0297] 53. Sani, O. G. et al. Mood variations decoded from multi-site intracranial human brain activity. Nat. Biotechnol. 36, 954-961 (2018).

[0298] 54. Bishop, C. Pattern Recognition and Machine Learning (Springer, 2006).

[0299] 55. Stanslaski, S. et al. Design and validation of a fully implantable, chronic, closed-loop neuromodulation device with concurrent sensing and stimulation. IEEE Trans. Neural Syst. Rehabil. Eng. 20, 410-421 (2012).

[0300] 56. Afshar, P. et al. A translational platform for prototyping closed-loop neuromodulation systems. Front. Neural Circuits 6, 117 (2013).

[0301] 57. Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12,2825-2830 (201 1 ).

[0302] 58. Shanechi, M. M. Brain-machine interfaces from motor to mood. Nat. Neurosci. 22, 1554-1564 (2019).

[0303] 59. Yekutieli, D. & Benjamini, Y. Resampling-based false discovery rate controlling multiple test procedures for correlated test statistics. J. Stat. Plan. Inference 82, 171 -196 (1999).

[0304] 60. Box, G. E. P., Jenkins, G. M. & Reinsei., G. C. Time Series Analysis: Forecasting andControl (Prentice Hall, 1994).

[0305] 61. North, B. V., Curtis, D. & Sham, P. C. A note on the calculation of empirical P Values from Monte Carlo procedures. Am. J. Hum. Genet 71 , 439-441 (2002).Supplementary MethodsJustification for Dichotomizing Pain Scores

[0306] Pain score outcome data were dichotomized because one main research goal was to assess the performance of such dichotomized data in the real-world, when implemented in a closed-loop brain stimulation device that requires defining binary symptom states. The device used for recording neural activity in this study, Medtronic Activa PC+S, is also capable of adaptively adjusting brain stimulation when the output of a linear discriminant crosses a specified threshold from one binary state (e.g. low pain) to another (e.g. high pain). A newly released commercial DBS system, Medtronic Percept, follows this same framework. The linear discriminant control signal is computed as a function of selected power band features; therefore, one main goal of the study was to determine which power band features would be most predictive of dichotomized pain state to inform a future closed-loop control signal in each subject. Therefore, when designing the study, we sought to undertake analysis that closely matched the demands of available clinical technology for ease of interpretation and feasibility of future closed-loop stimulation control.

[0307] DeCoster et al.,1 ,2 have provided a useful framework to guide the artificial dichotomization of data. They argue that dichotomization is specifically acceptable when three criteria are met 1 ) when the underlying variable is naturally categorical (such as NRS), 2) the observed measure has high reliability and 3) the relative group sizes of the dichotomized indicator match those of the underlying variable. The dichotomized variable (pain NRS) in the present study meets all three criteria. Specifically, the underlying variable (NRS) is itself naturally categorical and has high reliability (as evidenced high correlation with the separately reported variable of pain VAS in FIG. 1 D). Further, the group sizes of the dichotomized indicator are roughly equal in most cases (based on median split), and visually match the relative distribution of low vs high pain scores. Most importantly, the pain score dichotomization matches the intuitive grouping of the reported NRS scores as ‘high’ or ‘low’ by each subject, when they were queried about the median split used in our analyses.

[0308] Farrington and Loeber3 further highlight that dichotomization is more appropriate when distribution are very skewed, as is the case in all subjects in this manuscript that exhibited much more frequent high pain scores than medium or low scores both in the acute and chronic settings.Further, dichotomization may be beneficial when a variable is not linearly related to the measured outcome; in our case only a handful of band power variables exhibited a linear relationship with pain score, with high pain scores being associated with both high and lower power in most bands.Table S1 : Patient characteristics, recording / reporting details.Table S2: Decoding Characteristics for all full binary LDA models.Subject / Accuracy AUCPain Metric Threshold Sensitivity Specificity PPV (95% Cl) (p-value)Full models include all available neural data. Threshold indicates optimal LDA decision boundary.PPV = positive predictive value. AUG = area under the curve. Cl = Confidence Interval. One-sided P-values were determined based on permutation tests (n=1000, without multiple comparisons adjustment)Table S3. Acute Pain Calibrated Temperatures and Associated Pain ScoresTable shows temperatures calibrated for each patient for Acute pain thermal task showing mean reported pain NRS score and standard deviation (S.D.). Compare to ‘Brief Patient Descriptions’ in methods for description of allodynia / hypersensitivity.Subject: CP1AFFECTED SIDEUNAFFECTED SIDESubject: CP2AFFECTED SIDEUNAFFECTED SIDESubject: CP3AFFECTED SIDEUNAFFECTED SIDESubject: CP4AFFECTED SIDEUNAFFECTED SIDETable S4: Lack of correlation between diurnal NRS trend and diurnal feature trendsPearson’s correlation values (in bold) and corresponding p-values (two-sided t-statistic, with correction for multiple comparisons) are shown comparing the diurnal trend of each feature with diurnal trend of pain intensity NRS (see FIG. 19). There are no significant correlation values.Feature CP1 CP2 CP3 CP41 .00 0.99 0.24 0.89Supplementary References

[0309] 1. DeCoster, J., Iselin, A.-M. R. & Gallucci, M. A conceptual and empirical examination of justifications for dichotomization. Psychological Methods 14, 349 (20091207).

[0310] 2. DeCoster, J., Gallucci, M. & Iselin, A.-M. R. Best Practices for Using Median Splits, ArtificialCategorization, and their Continuous Alternatives. Journal of Experimental Psychopathology 2, 197- 209 (2011 ).

[0311] 3. Farrington, D. P. & Loeber, R. Some benefits of dichotomization in psychiatric and criminological research. Criminal Behaviour and Mental Health 10, 100-122 (2000).Example 2Personalized, Adaptive Brain Stimulation for Chronic PainIntroduction

[0312] Invasively mapping individual pain representations in the brain using selective stimulation with intracranial EEG (iEEG) electrodes can help identify optimal targets for stimulation in each individual. iEEG is commonly used in drug-resistant epilepsy to identify seizure foci and map eloquent cortex15. iEEG also provides the rare opportunity to examine brain regions that may harbor individualized biomarkers of pain symptoms1. By integrating these biomarkers into a personalized, adaptive brain stimulation algorithm, we propose a precision medicine therapy to individually deliver neurostimulation in response to elevated pain biomarkers. Such adaptive brain stimulation may potentially avert side effects or the development of habituation to electrical stimulation in patients with Parkinson’s disease16.

[0313] Here we describe the successful treatment of refractory chronic neuropathic pain syndromes in two participants by using a blinded and controlled, temporary iEEG discovery study to inform the neurosurgical implant of permanent electrodes for long-term ambulatory brain stimulation therapy. In one patient, we demonstrate rapid pain relief with the targeting of a novel brain region, the left caudate nucleus. Using key biomarkers of chronic pain intensity and relief, we successfully deployed adaptive deep brain stimulation therapy in response to real-time, personalized neural biomarkers of chronic pain severity and distress. We finally show that infrequent, intermittent stimulation often produces prolonged benefits outlasting electrical stimulation, which perform as well as adaptive algorithms.METHODSSTUDY OVERVIEW

[0314] This study is part of the Pain Relief with Individualized Brain Stimulation (PRISM) study, which is a single institution clinical feasibility trial aimed at developing closed-loop brain stimulation to treat refractory chronic neuropathic pain. Participants underwent two brain implant surgeries: we first performed an acute stimulation-response trial using temporary intracranial stereoencephalography (sEEG) electrodes to identify optimal stimulation sites and biomarkers for pain relief over 10d. sEEG is a high-resolution intracranial mapping technique commonly used in epilepsy (REF). Participants that obtained significant relief (>50%) in stage 1 were then implanted with two permanent deep brain stimulation devices (Medtronic Summit RC+S) capable of recording neural signals and providing on-board closed-loop stimulation in response to programmed biomarkers. Participants underwent neuropsychological and psychiatric evaluations prior to enrollment with the study psychiatrist and will be followed for up to two years. This study was approved under an Investigational Device Exemption from the FDA (NCT04144972) and approved by the Committee on Human Research at the University of California, San Francisco. The participant data presented reflects the first two patients that have progressed to ambulatory adaptive stimulation testing.

[0315] PS, PAS and EC were involved in the design of the clinical study. All authors were involved in the collection, storage, analysis, and interpretation of the data and the writing of the manuscript. Medtronic Inc. donated the Summit RC+S device hardware for the permanent implants, but did not have any role in data collection, analysis, interpretation or writing of the manuscript. All authors vouch for the accuracy, ethical conduct, and completeness of the data collection and for the fidelity of the study to the PROTOCOL (redact and link) . Informed consent was obtained from all participants before study enrollment.PARTICIPANTS

[0316] Participant 1 was a right-handed 56-year-old woman at the start of the study. 1 1 years prior, she suffered a right sided middle cerebral artery ischemic stroke, affecting the insula and frontal cortices. She fully regained motor function but weeks after she developed chronic post stroke pain, with constant burning and stabbing pain with allodynia along the entire left side of her body, from face to toe. She scored 29 out of 30 on the Montreal Cognitive Assessment (MoCA). She had been failed by numerous medication trials, pain psychotherapy and deep brain stimulation trial for pain. 1month after permanent bilateral DBS implant, the left sided DBS hardware was explanted due to infection, but the right sided system was unaffected and remained intact.

[0317] Participant 2 was a right-handed 51 -year-old man at the start of the study. He had a 21 -year history of cervical and lumbar degenerative disc disease and a prior right ankle injury resulting in complex regional pain syndrome in both legs. He had undergone a total of 26 surgeries on his spine and ankles and was unable to walk, sit or work for more than 10 minutes at a time. He scored 26 / 30 on the MoCA and had been failed by multiple medication classes, injections and two spinal cord stimulators. Both participants underwent comprehensive neuropsychological assessments and were free from any untreated psychiatric illness (Section S1 ).TEMPORARY AND PERMANENT IMPLANT DEVICES

[0318] Stage 1 : Participants were admitted to the hospital for a 10-day trial period for surgical implant of 8 sEEG depth electrodes (manufactured by PMT and AdTech) and two electrocorticography (ECoG) electrode arrays over sensorimotor cortex in each hemisphere (manufactured by PMT). sEEG electrodes contained 8-14 cylindrical contacts spaced 5 mm apart center-to-center, were 1.1 mm in diameter and terminated in the following targets: bilateral ACC, insula, SGC, ventral (P1) or medial thalamus (P2), ventral striatum, PVG, (FIG. 20A). Depth electrodes were stereotactical ly implanted using a small twist drill in the skull and secured in anchor bolts in the skull. The ECoG rectangular electrode arrays consisted of 20 flat disk-shaped contacts arranged in a 2x10 grid, each spaced 5 mm apart (center-to-center). Each array was inserted through a small parietal burr hole drilled in the and terminated in the subdural space overlying pia so that contacts spanned over S1 and M1. Accessible brain targets were chosen on an individual basis based on safety of the intended trajectories during surgical planning with BrainLab iPlan and Elements software. Externalized electrodes were connected to percutaneous connectors that interfaced with a 256-channel Nihon Kohden clinical neurophysiology recording and stimulation system. The percutaneous connector conducts electrical current to specified stimulation contacts and neural signals from the electrodes to a detachable analog link and cable that allows transmission of brain activity and stimulation signals to and from a computer (FIG. 20B). The participants underwent temporary electrode implant in September 2019 (P1) and May 2021 (P2) and had no complications. Surgical implant was done under general anesthesia and lasted 5 hours. Data collection and stimulation began on postoperative day 1 , and all electrodes were removed on day 10.

[0319] Stage 2: Both participants achieved >50% pain reduction during stage 1 and were implanted with the Summit RC+S after waiting a minimum of 6 weeks after sEEG removal. Each participant underwent stereotactic brain surgery under general anesthesia for placement of 4, cylindrical quadripolar DBS leads (Medtronic model 3387, 1.5 mm contact length and 2.0 mm intercontact spacing), targeted to two brain regions in each hemisphere: one electrode targeted the best stimulation site and the other targeted the optimal recording site on each side. Two electrodes on each side were connected to 60 cm lead extenders (Medtronic model 37087), tunneled under the skin of the neck, and connected to the implantable pulse generators (IPGs, Medtronic Summit RC+S), which was implanted in the infraclavicular location (one device on each side. Impedances were checked immediately after to ensure integrity of electrical connections. Participant 1 experienced an infection of the left sided system, which was explanted 8 weeks after initial implant.STIMULATION AND RECORDING SETTINGS

[0320] To determine the optimal stimulation targets and parameters for each participant (stage 1 ), We performed exploratory iEEG mapping by delivering continuous bipolar stimulation to contact pairs residing in 10 predetermined targets of interest including bilateral ACC, SGC, insula, motor / sensory cortex, VS, thalamus, PVG and regions that were incidentally sampled along the path of electrodes (i.e., caudate, vmPFC, frontal operculum) (FIG. 20A, Section S4). We systematically tested the clinical effects of a set of stimulation parameters used in prior DBS pain studies (10, 50 or 100Hz; 100-300 ps; 0-4 mA, biphasic square wave pulses). After performing initial safety testing at each site for 3s duration, regions of interest were stimulated in successive stimulation blocks for durations of 30-300 seconds in pseudorandomized order with sham stimulation. Pain score reports were collected within 60 seconds of stimulation termination. At the end of each stimulation block (each block consisted of 7-20 brain target / parameter combinations), we chose the best 3-6 candidate targets to advance for testing in the following block based on pain symptom surveys. Later stimulation blocks included 5 minutes of washout duration to allow pain scores to return to baseline to confirm efficacy. Chronic, ambulatory stimulation with the Summit RC+S device used biphasic, bipolar stimulation with active recharge (see Section S7 for details).

[0321] Neural intracranial electroencephalogram (iEEG) data during the temporary trial was acquired in a similar manner to that previously described during presurgical mapping for drugresistant epilepsy17. Signals were referenced to a single contact located in a confirmed white matter tract and sampled at 10kHz. Visually confirmed noisy channels and data segments containing artifacts were excluded from further analysis. We used standard iEEG processing with offlinebandpass filtering from 1-250 Hz in Python, notch filtering to remove line noise and harmonics, downsampling to 500Hz and performing Laplacian referencing between all channels. Neural data collection in the ambulatory setting was self-initiated by participants by connecting and streaming to their tablets up to 3 times daily for a minimum of 5 minutes per session with the Summit RC+S platform. We recorded four time domain and eight power domain channels per device, sampled at 250 Hz, using an embedded 100 Hz lowpass filter16.PAIN SYMPTOM REPORTING

[0322] Participants provided ecological momentary assessments of their pain multiple times daily by completing surveys and scales on a touch-sensitive digital tablet (iPad Pro, 11 -inch, gen. 2), Each pain report included pain intensity numerical rating score (NRS) and visual analog score (VAS), pain unpleasantness VAS (unpVAS), short-form McGill Pain Questionnaire (MPQ) and pain relief VAS (reliefVAS, pain relief only answered during stimulation) using the Research Electronic Data Capture, v10.6.19) platform. VAS sliders were horizontal and started at 50 / 100, but patients were instructed to move the slider to 0 first before completing the rating and received practice for 1 day. Pain body diagrams were captured using an iPad and Apple Pencil.SEEG analysis Preprocessing

[0323] To preprocess the data, first notch filtering was applied at the harmonics of 60 Hz up to 300 Hz to remove any electrical noise. A 'firwin' bandpass filter with cutoff frequencies between 0.1 and 250 Hz was then applied to isolate the frequency range of interest. The data were then resampled at 1000 Hz to reduce computational load and facilitate subsequent analysis. Finally, laplacian referencing was performed to reduce the effect of common noise sources, such as reference electrode drift or common mode noise. This preprocessing pipeline was implemented using the MNE- Python package

[0001] .Automated noise detection

[0324] An automated noise detection program was applied to detect large variations in the neural time-series that could be attributed to factors such as movement. The raw sEEG traces were first normalized by calculating the z-score across all channels for every trial using the mean and standard deviation. Epochs that exceeded 7 standard deviations for a minimum of 1 second were marked as noise. To minimize the impact of short-term fluctuations, noise epochs that occurred within threeseconds of each other were merged into a single epoch. Finally, noise epochs were extended by 1 .5 seconds before and after the marked period.Decoding

[0325] To determine the most relevant neural features for decoding pain, we compared multiple models that included a variety of time epochs preceding the survey, as well as pain variables from the survey. These models included 40-second, 5-minute, 10-minute, and 30-minute stimulation-free data epochs. If noise was detected during any of these epochs, the corresponding trials were excluded. We used linear discriminant analysis (LDA) to classify high and low pain states, with high pain defined as the top 30% and low pain as the bottom 30% of data. We calculated a total of 20 models, incorporating five pain scales, including 'intensity_vas_s0', 'nrs_s0', 'unpleasantness_vas_sO', and 'relief_sO'. For each temporal epoch, spectral features were extracted by using the Welch method to estimate the PSD, and then averaging power in the canonical frequency bands (delta, theta, alpha, beta, low-gamma, and high-gamma) across time. Epochs with spectral features above 3 standard deviations were excluded. The data were split into low- and high- pain groups, and LDA was performed using leave-one-out cross-validation with the solver set to 'Isqr' and the shrinkage set to 'auto' (need to rewrite this). Models with AUCs in the top 95thpercentile were deemed significant based on permutation testing, in which trial labels were shuffled 1000 times.

[0326] Following the model calculation, we moved forward with computing the decoding feature prevalence of anatomical brain regions. This process began with averaging across the power band features within each channel for every significant model. We then averaged these coefficients across different frequencies, which allowed us to isolate the channels showcasing the top and bottom 10 coefficient values for each model. These identified channels served as the best decoding electrodes, providing the most significant insight into the model's predictive functionality. Subsequently, we combined these coefficients within their respective anatomical brain regions, enabling us to determine the "feature prevalence" for each anatomical area. The "feature prevalence" was determined by summing up the appearances of each feature in the top 10 features across all models.PAIN PREDICTION MODELING

[0327] Neural activity was represented as the spectral power within canonical frequency bands (e.g., delta, theta, alpha etc.) recorded at each electrode contact. To explore the timescale on which painrelevant brain activity may be most predictive of pain, we constructed separate models for brain data divided into 40s, 5 min, 10 min, and 30 min epochs recorded immediately before participants reportedpain metrics. Spectral power values averaged over these timescales served as input features in separate 1) regression models to predict continuous pain scores and 2) classification models on dichotomized pain scores (comparable to requirements of the embedded device). Each model assigns different importance weights to different power features and contacts, according to how important a particular feature is for predicting each pain metric. By counting the number of times a particular contact (and associated brain region) was represented among the top 10 neural features from any single model, we ranked brain regions by importance to choose an optimal brain target for sensing biomarkers in each hemisphere (FIGS. 21A-21 D).STAGE 2: Chronic recording and stimulation with summit RCSData Collection

[0328] Data were collected with the Summit RC+S, an investigational sensing-enabled neurostimulator, capable of recording neural recordings of 4 time-domain channels from 2 connected electrodes at a sampling rate of 250 Hz. In addition, the spectral power from each channel was recorded through the device onboard Fourier module that converts digitized time-domain recording into spectral domain features. The onboard Fourier Transform was configured with a 256-pt Fast Fourier Transform (FFT) with 100% Hanning Window on every 250 data points with 125 data points moving window (overlap).Data Processing

[0329] The neural recordings from the Summit RC+S neurostimulator were extracted from standard JSON format using Python 3.8. For every redcap survey pain report, we extract onboard spectral features from 5-minute before to 5-minute after each report timestamp. Up to 1200 FFTs can be extracted from this 10-minute window based on the configured window size and overlap size. However, due to occasional telemetry error, the actual number of FFTs could be lower. A threshold of minimum 240 FFTs (2-minutes data) was set, and recording sessions with less than 240 FFTs within each pain report window were dropped. Feature matrix of size (# of pain reports x # of frequency) was created by averaging all accepted FFTs in each accepted window. Outlier removal was done by performing z-score normalization, and instances with features outside of 3 standard deviations from mean were removed.

[0330] To identify the features that discriminate best between high and low pain states, a linear discriminant analysis (LDA) classifier was used. A single-band feature vector was constructed byaveraging spectral power within + / - 2.5Hz bandwidth with a moving window of 0.5Hz. A label vector was created by thresholding pain scores below 30th percentile and above 70th percentile (highest and lowest 30% pain values). A leave-one-out cross-validation approach is used for each feature vector up to 50Hz by obtaining the prediction probability for each pain report session predicted by LDA trained with all pain reports except the testing report. Area-under-curve (AUG) for the feature was calculated by performing trapezoid integration on the receiver-operating characteristic (ROC) curve obtained from thresholding the prediction probability. The best feature that discriminates pain states was defined as the center frequency at which the peak AUG was obtained. The statistical significance of the obtained AUG was obtained by repeating the leave-one-out cross-validation process with 5000 surrogates with random permuted label vectors.Closed-Loop Development

[0331] Summit RC+S embedded closed-loop system can be designed with up to 2 independent LDA classifiers with up to 4 features input with 2 thresholds for up to 9 combined states. The primary chronic pain decoder will be implemented in the first LDA classifier with a single threshold detection. The input feature coefficient will be set to a standard constant with either positive or negative sign, depending on if the high-pain feature is higher than low-pain feature or not, to ensure that increase above threshold is transitioning toward stimulation state. In addition to the coefficient and threshold, the onset duration, which defines time-above-threshold to initiate stimulation, and offset duration, which defines time-below-threshold to stop stimulation, will be estimated based on simulation using at-home recording without stimulation. The simulation is done to choose the parameters at which prevent self-triggering of cyclic stimulation and control the stimulation duty cycles. The ramping time, which defines the speed of stimulation changes when closed-loop state is updated, is programmed based on in-person testing of stimulation tolerance.

[0332] After the primary decoder is set up, an additional LDA classifier should be set based on individual patient customization. For example, if the primary chronic pain decoder displayed large stimulation artifacts when stimulation is triggered, the second LDA classifier should include a stimulation detector that forces change stimulation state back to Off state after a fixed stimulation period (responsive closed-loop stimulation). In addition, if the primary neural feature changes significantly between awake and sleep, the second LDA classifier should also include a sleep detector to turn off or on stimulation at night depending on the patient’s need.BLINDING

[0333] Participants and investigators (except for stimulator operator) were blinded both to the identity of stimulation settings on all days (i.e., verum or sham, regions, etc.) and to the onset and duration of stimulation in the last 3 days of the temporary trial period. We delivered sham-control stimulation using 0 mA at 100Hz delivered to contacts outside regions of interest. Sham stimulation produced identical sounds and light flashes as real stimulation; the indicator LED light was covered with opaque tape and the computer monitor displaying acquired signals was shielded with a privacy screen.

[0334] In the ambulatory phase, sham testing and closed-loop testing, patients were blinded to stimulation condition by using a software interface that only identified stimulation groups as one of four rectangular boxes labeled A,B,C or D without any other visible parameters. Stimulation groups visible to participants did not necessarily correspond to stimulation groups on the device (i.e., Medtronic software groups A-D) and parameters for open vs closed loop were programmed in the software by an investigator different from the one that activated any program, or that routinely interacts with the participant each week.STATISTICS

[0335] Group wise pain scores were compared using Wilcoxon’s rank sum test without correction for multiple comparisons. Significance of decoding models was individually tested by generating surrogate null models obtained by randomly shuffling the training pain scores or high / low class labels 1000 times. Analysis was done using Python and Matlab 2022b programming languages (section SX).RESULTSIdentifying optimal stimulation targets for pain relief

[0336] During a 10-day trial period, we systematically delivered electrical stimulation to adjacent bipolar contacts at preselected brain regions of interest in each participant (FIGS. 22A-22C), to identify optimal candidate targets for pain relief. Our primary goal was to find one optimal target in each cerebral hemisphere. We first determined the best stimulation frequency for each tested site by stimulating at 10, 50 or 100 Hz for 30 seconds, under the assumption that a short bout of stimulation could produce rapid pain relief. In P1 , double-blinded stimulation of the right ACC, right VPthal and left ACC produced improvements in multiple pain metrics (NRS 3(1.5); VAS 25 (1 ),unpVAS 42(14), MPQ 20(2) and reliefVAS 90(14)) that were significantly different from sham stimulation (75(14)), alternative sites (negative control) and baseline pain scores (FIG. 23). At multiple instances throughout the 10-day trial, P1 reported complete resolution of all usual chronic pain symptoms in the left hemibody from post-stroke pain. In P2, blinded stimulation in the right CM and left body of the caudate (Caud) significantly improved his reported pain scores compared to other sites, sham, and baseline.

[0337] Because every stimulation epoch was not always associated with immediate benefit, it was difficult to recognize whether a particular site was effective in the moment. However, the stimulation targets most associated with analgesia (or mean improvements in pain scores) produced immediate and clinical meaningful pain relief during the majority of stimulation epochs.Identifying optimal sites to record pain biomarkers

[0338] To rigorously identify brain regions that harbor biomarkers of chronic pain, we aggregated results across different machine learning models that used neural activity to predict individual pain metrics (i.e., NRS, VAS, reliefVAS). Neural activity averaged over five minutes provided the most accurate classification of VAS for both participants.

[0339] For P1 , iEEG contacts in the right and left ACC most often contained important features that tracked pain symptoms. For P2, although the left insula ranked the highest in biomarker importance, surgical planning revealed a significantly higher risk of hemorrhage with this target due to intervening blood vessels that could be impacted by the rigid cannula required with permanent DBS implant. Therefore, we chose to implant the left ACC and right VMPFC for chronic biomarker sensing in P2.Ambulatory intermittent stimulation reproduces pain relief from temporary iEEG trial

[0340] Based on stimulation mapping and biomarker detection results from the temporary trial, we implanted permanent DBS electrodes targeting one optimal stimulation and sensing target in each hemisphere for each participant (FIGS. 22A-22C). After waiting 3 weeks for resolution of inflammation and acute tissue reaction post-surgery, participants were instructed connect to their tablets at home and perform up to 3 streaming sessions per day, with associated momentary assessments of pain metrics without stimulation (NRS, VAS, MPQ). Initial data collection without active stimulation allowed us to validate previously discovered pain biomarkers. We turned on stimulation after 100 pain reports were collected, and found that using similar parameters and bipolar contacts that corresponded to those used during the temporary trial again produced profound pain relief. In P1 , continuous stimulation with 50% duty cycle of the right ACC at 50 Hz significantlyreduced NRS, VAS and MPQ compared both to preimplant and pre-stimulation baseline scores (STATS HERE). For P2, similar continuous stimulation was also associated with significant reductions in VAS, NRS and MPQ (STATS HERE).

[0341] By performing systematic wash-in and wash-out testing, we aimed to estimate the latency to clinically meaningful pain relief after stimulation began, and the latency for pain scores to return to baseline after cessation of stimulation, respectively (Section S8).Adaptive DBS provides significant pain relief with lower electrical dosage.Decoding results:

[0342] After removing noise, for RCS02, 159, 115, 77, and 59 surveys were included in the decoding analysis for 40-second, five-minute, ten-minute, and thirty-minute chunks, respectively. For RCS05, 237, 93, 73, and 50 surveys were included for the 40-second, five-minute, ten-minute, and thirty- minute chunks, respectively.

[0343] In the decoding analysis for both patients, RCS05 and RCS02, we assessed feature prevalence by examining the top 10 features from each model based on the coefficients. For the intensity_vas_s0 metric, RCS05 included one_minute, five_minute, ten minute, and thirty_minute models, while RCS02 included one_minute, five_minute, and ten minute models. For the nrs_s0 metric, both patients included one_minute and five_minute models. In the case of the unpleasantness_vas_sO metric, both patients included one_minute, five_minute, ten minute, and thirty_minute models. For the relief_sO metric, no models were deemed significant. This evaluation provided insights into the most relevant neural features for decoding pain in both patients.

[0344] For patient RCS02, the feature prevalence analysis unveiled some notable trends. The left anterior cingulate cortex (L ACC) demonstrated the highest feature prevalence with 20 instances. Following this, the left medial frontal gyrus (L MFG) showed a count of 19, and the left posterior insula (L PINS) showed a prevalence of 17. In contrast, regions such as the left hypothalamus (L Hypothal) and the left triangular part of the inferior frontal gyrus (L IFGtri) did not contribute to any significant model. It's important to note that the majority of contributing features were predominantly from the left hemisphere. Turning to patient RCS05, we see a somewhat different pattern of feature prevalence. Similar to RCS02, the left anterior cingulate cortex (L ACC) showed the highest prevalence with 18 instances. The right caudate exhibited a significant contribution with 11 counts, followed closely by the right dorsolateral prefrontal cortex (R DLPFC) and the right medial frontalgyrus white matter (R MFG WM), each showing a count of 13. Unlike RCS02, several features were more evenly distributed between the two hemispheres.

[0345] Drawing comparisons between RCS02 and RCS05, we observed that both patients showed a strong contribution from the left anterior cingulate cortex (L ACC), confirming the critical role of this region in pain processing. Additionally, the left medial frontal gyrus (L MFG) also exhibited high feature prevalence in both patients.References

[0346] 1. Dahlhamer J. Prevalence of Chronic Pain and High-Impact Chronic Pain AmongAdults — United States, 2016. MMWR Morb Mortal Wkly Rep [Internet] 2018 [cited 2019 May 12];67.

[0347] 2. Apkarian AV, Baliki MN, Farmer MA. Predicting transition to chronic pain. Curr OpinNeurol 2013;26(4):360-7.

[0348] 3. Apkarian AV, Sosa Y, Sonty S, et al. Chronic Back Pain Is Associated withDecreased Prefrontal and Thalamic Gray Matter Density. J Neurosci 2004; 24(46):10410-5.

[0349] 4. Baliki MN, Petre B, Torbey S, et al. Corticostriatal functional connectivity predicts transition to chronic back pain. Nat Neurosci 2012;15(8):1117-9.

[0350] 5. Hashmi JA, Baliki MN, Huang L, et al. Shape shifting pain: chronification of back pain shifts brain representation from nociceptive to emotional circuits. Brain 2013;136(9):2751 -68.

[0351] 6. Mazars G, Roge R, Mazars Y. [Results of the stimulation of the spinothalamic fasciculus and their bearing on the physiopathology of pain]. Rev Neurol (Paris) 1960;103:136-8.

[0352] 7. Frizon LA, Yamamoto EA, Nagel SJ, Simonson MT, Hogue O, Machado AG. DeepBrain Stimulation for Pain in the Modern Era: A Systematic Review. Neurosurgery [Internet] [cited 2019 Feb 28]; Available from: academic.oup.com / neurosurgery / advance- article / doi / 10.1093 / neuros / nyy552 / 5364269

[0353] 8. Hosobuchi Y, Adams JE, Rutkin B. Chronic thalamic and internal capsule stimulation for the control of central pain. Surg Neurol 1975;4(1 ):91 -2.

[0354] 9. Coffey RJ. Deep brain stimulation for chronic pain: results of two multicenter trials and a structured review. Pain Med Malden Mass 2001 ;2(3) :183-92.

[0355] 10. Wager TD, Atlas LY, Lindquist MA, Roy M, Woo C-W, Kross E. An fMRI-BasedNeurologic Signature of Physical Pain. N Engl J Med 2013;368(15):1388-97.

[0356] 11 . Davis KD, Aghaeepour N, Ahn AH, et al. Discovery and validation of biomarkers to aid the development of safe and effective pain therapeutics: challenges and opportunities. Nat Rev Neurol 2020;16(7):381 -400.

[0357] 12. Boccard SGJ, Prangnell SJ, Pycroft L, et al. Long-Term Results of Deep BrainStimulation of the Anterior Cingulate Cortex for Neuropathic Pain. World Neurosurg 2017;106(Supplement C):625-37. 13. Hollingworth M, Sims-Williams HP, Pickering AE, Barua N, Patel NK. Single Electrode Deep Brain Stimulation with Dual Targeting at Dual Frequency for the Treatment of Chronic Pain: A Case Series and Review of the Literature. Brain Sci [Internet] 2017 [cited 2020 Jul 23];7(1). Available from: ncbi.nlm.nih.gov / pmc / articles / PMC5297298 /

[0358] 14. Kohoutova L, Atlas LY, Buchel C, et al. Individual variability in brain representations of pain. Nat Neurosci 2022;25(6):749-59.

[0359] 15. Bartolomei F, Lagarde S, Wendling F, et al. Defining epileptogenic networks: Contribution of SEEG and signal analysis. Epilepsia 2017;58(7):1131-47.

[0360] 16. Gilron R, Little S, Perrone R, et al. Long-term wireless streaming of neural recordings for circuit discovery and adaptive stimulation in individuals with Parkinson’s disease. Nat Biotechnol 2021 ;1-8.

[0361] 17. Scangos KW, Makhoul GS, Sugrue LP, Chang EF, Krystal AD. State-dependent responses to intracranial brain stimulation in a patient with depression. Nat Med 2021 ;27(2) :229-31 .

Claims

What is claimed is:1 . A method for treating chronic pain in a subject, the method comprising: positioning a first electrode at a first location of the brain of the subject to deliver electrical stimulation to a target stimulation site to relieve the chronic pain; positioning a second electrode at a second location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; detecting the brain electrical signal at the sensorimotor cortex region of the brain the subject using the second electrode; analyzing the brain electrical signal using a binary pain classification model that classifies the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state; and applying electrical stimulation to the target stimulation site in the brain of the subject using the first electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal exceeds a specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state.

2. The method of claim 1 , wherein the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.

3. The method of claim 1 or 2, further comprising calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.

4. The method of any one of claims 1 -3, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.

5. The method of claim 4, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.

6. The method of claim 5, wherein the OFC region is a contralateral OFC region.

7. The method of any one of claims 4-6, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.

8. The method of any one of claims 1 -7, wherein the target stimulation site is in an anterior cingulate cortex (ACC) region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.

9. The method of any one of claims 1 -8, further comprising mapping the brain of the subject to identify an optimal location in the sensorimotor cortex region to detect the brain electrical signal associated with the chronic pain.

10. The method of any one of claims 1 -9, further comprising mapping the brain of the subject to identify an optimal target stimulation site to deliver the electrical stimulation to relieve the chronic pain.

11. The method of claim 10, wherein said mapping comprises identifying the optimal target stimulation site in each cerebral hemisphere.

12. The method of any one of claims 1 -11 , wherein the chronic pain is caused by a pain- associated disorder, wherein applying the electrical stimulation relieves the pain.

13. The method of claim 12, wherein the pain-associated disorder is post-stroke pain, complex regional pain syndrome, back pain, failed back surgery syndrome, spinal cord injury, spinal stenosis, post-surgical pain, arachnoiditis, angina, nerve-related pain, peripheral vascular disease, pain after an amputation, visceral abdominal pain, perineal pain, multiple sclerosis, arthritis, or chronic leg pain, neck pain, foot pain, or arm pain.

14. The method of claim 13, wherein the nerve-related pain is caused by diabetic neuropathy, cancer-related neuropathy, or nerve damage caused by radiation, surgery, or chemotherapy.

15. The method of claim 13, wherein the pain-associated disorder is said post-stroke pain, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a right ACC region and a left ACC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right ACC region, a right VPthal region, and a left ACC region.

16. The method of claim 13, wherein the pain-associated disorder is said complex regional pain syndrome, wherein said detecting the brain electrical signal comprises detecting the brain electrical signal at a left ACC region and a right VMPFC region; and wherein said applying electrical stimulation comprises applying electrical stimulation to a right CM region and a left body of the caudate nucleus region.

17. The method of any one of claims 1-16, wherein the brain electrical signal is a local field potential.

18. The method of any one of claims 1-17, further comprising using a control algorithm to automate said applying electrical stimulation when the brain electrical signal exceeds the specified threshold indicating the subject is transitioning to the high pain severity state or is in the high pain severity state.

19. The method of claim 18, wherein the control algorithm uses a machine learning algorithm for pain classification.

20. The method of claim 19, wherein the machine learning algorithm is a supervised machine learning algorithm, an unsupervised machine learning algorithm, or a semi-supervised machine learning algorithm.

21. The method of any one of claims 18-20, wherein the control algorithm further modulates one or more programmed stimulation parameters to relieve the chronic pain.

22. The method of any one of claims 18-21 , wherein the control algorithm further uses linear discriminant analysis (LDA) to adjust stimulation amplitude, or frequency of the electrical stimulation.

23. The method of any one of claims 1-22, wherein the electrical stimulation is applied unilaterally or bilaterally.

24. The method of any one of claims 1 -23, wherein the brain electrical signal comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations.

25. The method of claim 24, wherein the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz.

26. The method of claim 25, wherein the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.

27. The method of any one of claims 1 -26, wherein the control algorithm further modulates one or more programmed stimulation parameters based on a level of power of the brain electrical signal.

28. The method of claim 27, wherein the level of power is a level of delta power from a contralateral OFC region.

29. The method of claim 27 or 28, wherein changes in power features associated with the chronic pain are sustained for at least 4 seconds.

30. The method of any one of claims 1 -29, wherein the first electrode is placed within or on a surface of the ACC region, the subgenual cingulate gyrus (SGC) region, the insula region, the motor / sensory cortex region, the ventral striatum (VS) region, the thalamus region, the thalamusventral posterior complex (VPthal) region, the centromedian thalamus (CM) region, the periaqueductal gray region, the periventricular gray (PVG) region, the caudate nucleus region, the ventromedial prefrontal cortex (VMPFC) region, or the frontal operculum region.31 . The method of any one of claims 1 -30, wherein the second electrode is placed within or on a surface of the OFC region, the ACC region, the insula region, the vmPFC region, the BG region, or the CM region.

32. The method of any one of claims 1 -31 , wherein the first electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

33. The method of any one of claims 1 -32, wherein the second electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

34. The method of any one of claims 1 -33, wherein the second electrode is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.

35. The method of any one of claims 1 -34, wherein the method further comprises assessing effectiveness of the treatment in the subject.

36. The method of claim 35, wherein said assessing comprises using a visual analog scale (VAS), a pain intensity numerical rating score (NRS), or a verbal rating scale.

37. A computer implemented method for programming a deep brain stimulation (DBS) device to treat chronic pain in a subject, the computer performing steps comprising: a) receiving recorded brain electrical signal data from a sensorimotor cortex region of the brain of the subject; b) analyzing the recorded brain electrical signal data using a binary pain classification model that classifies the brain electrical signal data as indicating that the subject is in either a high pain severity state or a low pain severity state;c) adjusting one or more programmed stimulation parameters based on the recorded brain electrical signal data according to an algorithm control law; and d) instructing the DBS device to apply an electrical stimulation to a target stimulation site in the brain of the subject using an electrode in a manner effective to treat the chronic pain in the subject when the brain electrical signal data indicates the subject is transitioning to the high pain severity state or is in the high pain severity state.

38. The computer implemented method of claim 37, wherein the brain electrical signal data comprises field potential data.

39. The computer implemented method of claim 37 or 38, wherein the binary pain classification model uses a linear discriminant analysis (LDA) classifier to classify the brain electrical signal as indicating that the subject is in either a high pain severity state or a low pain severity state.

40. The computer implemented method of any one of claims 37-39, further comprising calculating pain scores, wherein the high pain severity state corresponds to pain scores at or above 70thpercentile, and wherein the low pain severity state corresponds to pain scores at or below 30thpercentile.41 . The computer implemented method of any one of claims 37-40, further comprising splitting the recorded brain electrical signal data into consecutive time epochs.

42. The computer implemented method of claim 41 , further comprising calculating average power values for one or more neurophysiological activity frequency bands for each time epoch, wherein the average power values are used to calculate pain scores for each time epoch.

43. The computer implemented method of claim 41 or 42, further comprising calculating pain scores using brain electrical signal data during a stimulation-free data epoch to distinguish the pain scores that are at or above the 70thpercentile corresponding to the high pain severity state and the pain scores that are at or below the 30thpercentile corresponding to the low pain severity state.

44. The computer implemented method of claim 43, wherein the stimulation-free data epoch comprises at least 5 minutes, 10-minutes, 30-minutes or 40 minutes of time of the recorded brain electrical signal data.

45. The computer implemented method of any one of claims 37-44, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.

46. The computer implemented method of claim 45, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.

47. The computer implemented method of claim 46, wherein the OFC region is a contralateral OFC region.

48. The computer implemented method of any one of claims 45-47, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.

49. The computer implemented method of any one of claims 37-48, wherein the target stimulation site is in an ACC region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.

50. The computer implemented method of any one of claims 37-49, further comprising: a) ranking predicted stimulation effectiveness for available settings of the DBS device based on classifier scores for stimulation effectiveness of each setting using a linear classification model; b) selecting stimulation settings predicted to have highest stimulation effectiveness based on the linear classification model;c) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness; d) analyzing the recorded brain electrical signal data to evaluate neural response of the subject to the electrical stimulation; e) updating the linear classification model based on the neural response of the subject to the electrical stimulation to generate an updated linear classification model; f) updating the ranking of predicted stimulation effectiveness for the available settings of the DBS device using the updated linear classification model; g) selecting stimulation settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; h) receiving recorded brain electrical signal data from the sensorimotor cortex region of the brain of the subject after applying the electrical stimulation with the DBS device to the target stimulation site of the brain of the subject using the settings predicted to have the highest stimulation effectiveness based on the updated linear classification model; and i) repeating e) - h) to adjust the available settings of the DBS device to optimize stimulation effectiveness.51 . The computer implemented method of claim 50, wherein the linear classification model uses linear discriminant analysis (LDA) to adjust amplitude of current and frequency of the electrical stimulation.

52. The computer implemented method of any one of claims 37-51 , wherein the recorded brain electrical signal data comprises alpha frequency, beta frequency, gamma frequency, delta frequency, or theta frequency neural oscillations.

53. The computer implemented method of claim 52, wherein the delta frequency neural oscillations are in a range from 1 Hz to 4 Hz, the theta frequency neural oscillations are in a range from 4 Hz to 8 Hz, the alpha frequency neural oscillations are in a range from 8 Hz to 12 Hz, the beta frequency neural oscillations are in a range from 12 Hz to 30 Hz, and the gamma frequency neural oscillations are in a range from 30 Hz to 100 Hz.

54. The computer implemented method of claim 53, wherein the gamma frequency neural oscillations are in a range from 30 Hz to 70 Hz or 70 Hz to 100 Hz.

55. The computer implemented method of any one of claims 37-54, wherein the algorithm control law further modulates one or more programmed stimulation parameters based on a change in a level of power of a brain electrical signal.

56. The computer implemented method of claim 55, wherein the level of power is a level of delta power from a contralateral OFC region.

57. The computer implemented method of claim 55 or 56, wherein the change in the level of power is sustained for at least 4 seconds.

58. The computer implemented method of any one of claims 37-57, further comprising storing a user profile for the subject comprising information regarding the recorded brain electrical signal data associated with the chronic pain.

59. The computer implemented method of any one of claims 37-58, further comprising storing a user profile for the subject comprising information regarding the programmed stimulation parameters used to apply electrical stimulation to the target stimulation site in the brain of the subject to treat the chronic pain in the subject based on the recorded brain electrical signal data.

60. A non-transitory computer-readable medium comprising program instructions that, when executed by a processor in a computer, causes the processor to perform the method of any one of claims 37-59.

61. A kit comprising the non-transitory computer-readable medium of claim 60 and instructions for treating chronic pain in a subject with a deep brain stimulation device.

62. A system for treating chronic pain in a subject, the system comprising: a first electrode adapted for positioning at a location to deliver electrical stimulation to a target stimulation site to relieve the chronic pain in the subject;a second electrode adapted for positioning at a location in a sensorimotor cortex region of the brain of the subject to detect a brain electrical signal associated with the chronic pain; and a processor programmed according to the computer implemented method of any one of claims 37-59 to adjust one or more stimulation parameters based on the recorded brain electrical signal data and instruct the first electrode to apply an electrical stimulation to the target stimulation site in the brain of the subject in a manner effective to treat the chronic pain in the subject.

63. The system of claim 62, wherein the brain electrical signal data comprises field potential data.

64. The system of claim 62 or 63, wherein the first electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

65. The system of any one of claims 62-64, wherein the second electrode is a non-brain penetrating surface electrode array or a brain-penetrating electrode array.

66. The system of any one of claims 62-65, wherein the second electrode is a stereoencephalography (sEEG) electrode, an electroencephalogram (EEG) electrode array, an electrocorticogram (ECoG) electrode array, or a subgaleal mounted, burrhole mounted, or cranially mounted electrode.

67. The system of any one of claims 62-66, wherein the sensorimotor cortex region is an orbitofrontal cortex (OFC) region, an anterior cingulate cortex (ACC) region, an insula region, a ventromedial prefrontal cortex (vmPFC) region, a basal ganglia (BG) region, or a centromedian thalamus (CM) region.

68. The system of claim 67, wherein the sensorimotor cortex region is contralateral relative to the side of body where the subject feels pain.

69. The system of claim 68, wherein the OFC region is a contralateral OFC region.

70. The system of any one of claims 67-69, wherein the sensorimotor cortex region is a right OFC region, a left OFC region, a right ACC region, a left ACC region, a right VMPFC region, or a left VMPFC region, or a combination thereof.71 . The system of any one of claims 62-70, wherein the target stimulation site is in an ACC region, a subgenual cingulate gyrus (SGC) region, an insula region, a motor / sensory cortex region, a ventral striatum (VS) region, a thalamus region, a thalamus ventral posterior complex (VPthal) region, a centromedian thalamus (CM) region, a periaqueductal gray region, a periventricular gray (PVG) region, a caudate nucleus region, a ventromedial prefrontal cortex (VMPFC) region, or a frontal operculum region of the brain of the subject.

72. The system of any one of claims 62-71 , wherein the system further comprises a user interface comprising an input electronically coupled to the processor for instructing the first electrode to apply an electrical stimulation to the target stimulation site to relieve the chronic pain in the subject.

73. The system of claim 72, wherein the user interface is password protected and is operable by a health care practitioner.

Images