System and method for processing connectivity value between sub-processing regions

Abstract

To obtain a connectivity value using one or more methods, such as imaging.SOLUTION: The present disclosure relates to a system and a method for: accessing a data structure including subject identification information for identifying a subject and a subject connectivity value derived from at least one of a scan or test performed on the subject, the subject connectivity value representing a magnitude of a connection associated with at least one sub-processing region of a nervous system of the subject; comparing the subject connectivity value to a connectivity threshold value to determine a classification of the subject; upon determining that the subject connectivity value exceeds the connectivity threshold value, storing, in the data structure, an association between the subject identification information and a first classification value corresponding to a first classification; or, upon determining that the subject connectivity value is less than the connectivity threshold value, storing, in the data structure, an association between the subject identification information and a second classification value corresponding to a second classification.SELECTED DRAWING: Figure 8

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims the benefit and priority of U.S. Provisional Patent Application No. 62 / 649,469, filed on Mar. 28, 2018, the entire disclosure of which is incorporated herein by reference.

[0002] Government Support This invention was made with government support under grant number 5K23MH099223 awarded by the National Institutes of Health. The government has certain rights in this invention.

Background Art

[0003] Background of the Invention The background of the present technology is described below merely to assist in understanding the present technology, not to explain or constitute the prior art with respect to the present technology.

[0004] Relevance values can be obtained using one or more methods such as imaging. The relevance value can represent a state. Some states can be improved by affecting the change of the relevance value.

Summary of the Invention

[0005] Summary of the Present Technology In one aspect, the present disclosure provides a data processing system having a hardware storage device that stores one or more data structures and executable logic for processing data values stored within the one or more data structures. The data processing system includes a hardware storage device that stores one or more data structures, each data structure storing keyed data, one item of the keyed data including a key representing a subject, and the hardware storage device further storing executable logic including classification rules. The data processing system includes one or more data processing devices that access at least one of the one or more data structures from the hardware storage device and obtain keyed data of a particular subject from at least one of the accessed ones of the one or more data structures, the obtained keyed data including a subject correlation value derived from at least one of a scan or a test performed on the subject represented by the key included in the obtained keyed data, the subject correlation value representing the magnitude of the correlation associated with at least one sub-processing region of the subject's nervous system. The data processing system further includes an executable logic engine configured to execute executable logic on the keyed data and apply the classification rules to the subject correlation value and a correlation threshold to determine a classification of the subject. Storage by the hardware storage device is further configured to store, within the data structure, a relationship between the key represented by the keyed data and a first classification value corresponding to a first classification when the subject correlation value is greater than the correlation threshold. Storage by the hardware storage device is further configured to store, within the data structure, a relationship between the key represented by the keyed data and a second classification value corresponding to a second classification when the subject correlation value is less than the correlation threshold.

[0006] In some embodiments, at least one sub-processing region includes the dorsolateral prefrontal cortex (DPFC) and the amygdala (AMG), the subject correlation value and the correlation threshold represent the effective correlation between the DPFC and the AMG and the effective correlation threshold between the DPFC and the AMG, the first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible.

[0007] In some embodiments, at least one sub-processing region includes the dorsal anterior cingulate cortex (dACC) and the amygdala (AMG), the subject correlation value and the correlation threshold respectively represent the effective correlation between the dACC and the AMG and the effective correlation threshold between the dACC and the AMG, the first classification value indicates that the subject is eligible, and the second classification value indicates that the subject is ineligible.

[0008] In some embodiments, at least one sub-processing region includes the default mode resting state network (DMN), the subject correlation value and the correlation threshold respectively represent the functional correlation within the DMN and the functional correlation threshold related to the DMN, the first classification value indicates that the subject is eligible, and the second classification value indicates that the subject is ineligible.

[0009] In some embodiments, at least one sub-processing region includes the salience resting state network (SAL), the subject correlation value and the correlation threshold respectively represent the functional correlation within the SAL and the functional correlation threshold related to the SAL, the first classification value indicates that the subject is eligible, and the second classification value indicates that the subject is ineligible.

[0010] In some embodiments, at least one sub - processing region includes a left central executive network (LCEN) and a right central executive network (RCEN), the subject - relatedness value and the relatedness threshold respectively represent the integrity between the LCEN and the RCEN and the integrity threshold between the LCEN and the RCEN, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0011] In some embodiments, at least one sub - processing region includes a dorsal default mode resting state network (dDMN) and a ventral default mode resting state network (vDMN), the subject - relatedness value and the relatedness threshold respectively represent the integrity between the dDMN and the vDMN and the integrity threshold between the dDMN and the vDMN, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0012] In some embodiments, at least one sub - processing region includes a left central executive network (LCEN) and a ventral default mode resting state network (vDMN), the subject - relatedness value and the relatedness threshold respectively represent the integrity between the LCEN and the vDMN and the integrity threshold between the LCEN and the vDMN, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0013] In some embodiments, at least one sub - processing region includes a left central executive network (LCEN) and a salience resting state network (SAL), the subject - relatedness value and the relatedness threshold respectively represent the integrity between the LCEN and the SAL and the integrity threshold between the LCEN and the SAL, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0014] In one aspect, the present disclosure provides a method for classifying a subject based on a relevance value. The method includes accessing, by a data processing system including one or more processors, a data structure including subject identification information identifying a subject and a subject relevance value derived from at least one of a scan or a test performed on the subject, the subject relevance value including representing a magnitude of relevance associated with at least one sub-processing region of the subject's nervous system. The method further includes comparing, by the data processing system, the subject relevance value with a relevance threshold and determining a classification of the subject. The method also includes storing, by the data processing system, within the data structure, a relationship between the subject identification information and a first classification value corresponding to a first classification in response to a determination by the data processing system that the subject relevance value is greater than the relevance threshold. The method also includes storing, by the data processing system, within the data structure, a relationship between the subject identification information and a second classification value corresponding to a second classification in response to a determination by the data processing system that the subject relevance value is less than the relevance threshold.

[0015] In some embodiments, the at least one sub-processing region includes a dorsolateral prefrontal cortex (DPFC) and an amygdala (AMG), the subject relevance value and the relevance threshold represent an effective relevance between the DPFC and the AMG and an effective relevance threshold between the DPFC and the AMG, respectively, the first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible.

[0016] In some embodiments, the at least one sub-processing region includes a dorsal anterior cingulate cortex (dACC) and an amygdala (AMG), the subject relevance value and the relevance threshold represent an effective relevance between the dACC and the AMG and an effective relevance threshold between the dACC and the AMG, respectively, the first classification value indicates that the subject is eligible, and the second classification value indicates that the subject is ineligible.

[0017] In some embodiments, at least one sub-processing region includes a default mode network (DMN) at rest, the subject correlation value and the correlation threshold respectively represent the functional correlation within the DMN and the functional correlation threshold related to the DMN, the first classification value indicates that the subject is qualified, and the second classification value indicates that the subject is unqualified.

[0018] In some embodiments, at least one sub-processing region includes a salience network at rest (SAL), the subject correlation value and the correlation threshold respectively represent the functional correlation within the SAL and the functional correlation threshold related to the SAL, the first classification value indicates that the subject is qualified, and the second classification value indicates that the subject is unqualified.

[0019] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a right central executive network (RCEN), the subject correlation value and the correlation threshold respectively represent the integrity between the LCEN and the RCEN and the integrity threshold between the LCEN and the RCEN, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0020] In some embodiments, at least one sub-processing region includes a dorsal default mode network at rest (dDMN) and a ventral default mode network at rest (vDMN), the subject correlation value and the correlation threshold respectively represent the integrity between the dDMN and the vDMN and the integrity threshold between the dDMN and the vDMN, the first classification value indicates that the subject is unqualified, and the second classification value indicates that the subject is qualified.

[0021] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a ventral default mode network (vDMN), the subject correlation value and the correlation threshold respectively represent the integrity between the LCEN and the vDMN and the integrity threshold between the LCEN and the vDMN, the first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible.

[0022] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a salience network at rest (SAL), the subject correlation value and the correlation threshold respectively represent the integrity between the LCEN and the SAL and the integrity threshold between the LCEN and the SAL, the first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible.

[0023] In some embodiments, the subject correlation value is a first subject correlation value, the correlation threshold is a first correlation threshold, and the method further comprises accessing, by a data processing system, a data structure including a second subject correlation value derived from at least one scan or test performed on the subject, the second subject correlation value representing the magnitude of the correlation associated with at least one other sub-processing region of the subject's nervous system. The method also includes comparing, by the data processing system, the second subject correlation value. The method further includes storing, by the data processing system, a relationship between the subject identification information and a first classification value corresponding to the first classification within the data structure in response to a determination by the data processing system that the first subject correlation value is greater than the first correlation threshold and the second subject correlation value is less than the second correlation threshold. The method also includes storing, by the data processing system, a relationship between the subject identification information and a second classification value corresponding to the second classification within the data structure in response to a determination by the data processing system that the first subject correlation value is less than the first correlation threshold and the second subject correlation value is greater than the second correlation threshold.

[0024] In some embodiments, at least one sub-processing region includes the dorsolateral prefrontal cortex (DPFC) and the amygdala (AMG), the first subject correlation value and the first correlation threshold represent the effective correlation between the DPFC and the AMG and the effective correlation threshold between the DPFC and the AMG, another at least one sub-processing region includes the dorsal anterior cingulate cortex (dACC) and the amygdala (AMG), and the second subject correlation value and the second correlation threshold represent the effective correlation between the dACC and the AMG and the effective correlation threshold between the dACC and the AMG, respectively. The first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible.

[0025] In yet another aspect, the present disclosure provides a method for ranking candidates based on correlation values. The method includes obtaining, by a data processing system including one or more processors, a plurality of subject correlation values associated with a plurality of subject identification information corresponding to a plurality of subjects from a data structure, wherein each subject correlation value of the plurality of subject correlation values represents the magnitude of the correlation associated with at least one sub-processing region of the nervous system of each of the plurality of subjects. The method further includes assigning, by the data processing system, a rank to each subject identification information in the data structure based on the subject correlation value. The method also includes selecting, by the data processing system, a subset of the plurality of subjects, wherein each subject of the subset is selected based on the difference between the correlation threshold and the subject correlation value. The method further includes generating, by the data processing system, an ordered list of subject identification information corresponding to the selected subset of the plurality of subjects, wherein the subject identification information is arranged based on the rank assigned to each subject identification information.

[0026] In some embodiments, at least one sub - processing region includes the dorsolateral prefrontal cortex (DPFC) and the amygdala (AMG), the plurality of subject - relatedness values and relatedness thresholds represent the effective relatedness between the DPFC and the AMG and the effective relatedness threshold between the DPFC and the AMG for each of the plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are less than the relatedness threshold.

[0027] In some embodiments, at least one sub - processing region includes the dorsal anterior cingulate cortex (dACC) and the amygdala (AMG), the plurality of subject - relatedness values and relatedness thresholds represent the effective relatedness between the dACC and the AMG and the effective relatedness threshold between the dACC and the AMG for each of the plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are greater than the relatedness threshold.

[0028] In some embodiments, at least one sub - processing region includes the default - mode resting - state network (DMN), the plurality of subject - relatedness values and relatedness thresholds represent the functional relatedness within the DMN and the functional relatedness threshold within the DMN for each of the plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are greater than the relatedness threshold.

[0029] In some embodiments, at least one sub - processing region includes the salience resting - state network (SAL), the plurality of subject - relatedness values and relatedness thresholds represent the functional relatedness within the SAL and the functional relatedness threshold within the SAL for each of the plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are greater than the relatedness threshold.

[0030] In some embodiments, at least one sub - processing region includes a left central execution network (LCEN) and a right central execution network (RCEN), a plurality of subject - relatedness values and relatedness thresholds represent the integrity between LCEN and RCEN and the integrity threshold between LCEN and RCEN for each of a plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are less than the relatedness threshold.

[0031] In some embodiments, at least one sub - processing region includes a dorsal default mode resting state network (dDMN) and a ventral default mode resting state network (vDMN), a plurality of subject - relatedness values and relatedness thresholds represent the integrity between dDMN and vDMN and the integrity threshold between dDMN and vDMN for each of a plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are less than the relatedness threshold.

[0032] In some embodiments, at least one sub - processing region includes a left central execution network (LCEN) and a ventral default mode resting state network (vDMN), a plurality of subject - relatedness values and relatedness thresholds represent the integrity between LCEN and vDMN and the integrity threshold between LCEN and vDMN for each of a plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are less than the relatedness threshold.

[0033] In some embodiments, at least one sub - processing region includes a left central execution network (LCEN) and a salience resting state network (SAL), a plurality of subject - relatedness values and relatedness thresholds represent the integrity between LCEN and SAL and the integrity threshold between LCEN and SAL for each of a plurality of subjects, and the subject - relatedness values related to a selected subset of the plurality of subjects are less than the relatedness threshold.

[0034] In yet another aspect, the present disclosure provides a method for determining the effectiveness of treatment based on a correlation value. The method includes obtaining, by a data processing system including one or more processors, a first correlation value of a subject, where the first correlation value represents a first magnitude of correlation associated with at least one sub-processing region of the subject's nervous system at a first time point. The method further includes obtaining, by the data processing system, a second correlation value of the subject, where the second correlation value represents a second magnitude of correlation associated with at least one sub-processing region of the subject's nervous system at a second time point after the subject has received cognitive-emotional training. The method also includes obtaining, by the data processing system, a difference between the first correlation value and the second correlation. The method further includes storing, by the data processing system, in a data structure including subject identification information for identifying the subject, a relationship between the subject identification information and a first classification value corresponding to a first classification in response to a determination that the difference is greater than a threshold. The method further includes storing, by the data processing system, in the data structure including the subject identification information, a relationship between the subject identification information and a second classification value corresponding to a second classification in response to a determination that the difference is less than the threshold.

[0035] In some embodiments, the at least one sub-processing region includes the dorsolateral prefrontal cortex (DPFC) and the amygdala (AMG), and the first correlation value and the second correlation value represent the effective connectivity between the subject's DPFC and AMG at the first time point and the second time point, respectively.

[0036] In some embodiments, the at least one sub-processing region includes the dorsal anterior cingulate cortex (dACC) and the amygdala (AMG), and the first correlation value and the second correlation value represent the effective connectivity between the subject's DPFC and AMG at the first time point and the second time point, respectively.

[0037] In some embodiments, at least one sub-processing region includes a default mode network (DMN), and the first and second correlation values represent the functional connectivity within the subject's DMN at the first and second time points, respectively.

[0038] In some embodiments, at least one sub-processing region includes a salience network (SAL), and the first and second correlation values represent the functional connectivity within the subject's SAL at the first and second time points, respectively.

[0039] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a right central executive network (RCEN), and the first and second correlation values represent the integration between the subject's LCEN and RCEN at the first and second time points, respectively.

[0040] In some embodiments, at least one sub-processing region includes a dorsal default mode network (dDMN) and a ventral default mode network (vDMN), and the first and second correlation values represent the integration between the subject's dDMN and vDMN at the first and second time points, respectively.

[0041] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a ventral default mode network (vDMN), and the first and second correlation values represent the integration value between the subject's LCEN and vDMN at the first and second time points, respectively.

[0042] In some embodiments, at least one sub-processing region includes a left central executive network (LCEN) and a salience network (SAL), and the first and second correlation values represent the integration value between the subject's LCEN and SAL at the first and second time points, respectively.

[0043] In yet another aspect, the present disclosure provides a method for increasing the effective connectivity between the DPFC and the AMG and decreasing the effective connectivity between the dACC and the AMG over an effective amount of time in a subject suffering from an affective disorder.

[0044] In yet another aspect, the present disclosure provides a method for decreasing the functional connectivity in at least one of the dDMN and the SAL over an effective amount of time in a subject suffering from an affective disorder.

[0045] In yet another aspect, the present disclosure provides a method for increasing the integration within at least one of the LCEN and the RCEN, the dDMN and the vDMN, the LCEN and the vDMN, or the LCEN and the SAL over an effective amount of time in a subject suffering from an affective disorder.

[0046] In any of the above-described embodiments of the methods disclosed herein, the affective disorder can be major depressive disorder (MDD), bipolar disorder, post-traumatic stress disorder (PTSD), generalized anxiety disorder, social phobia, obsessive-compulsive disorder, treatment-resistant depression, or borderline personality disorder. BRIEF DESCRIPTION OF THE DRAWINGS

[0047]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

[0048] Detailed Description of the Invention It will be appreciated that some aspects, embodiments, implementations, variations and features of the method of the present invention are described below with varying degrees of detail for the purpose of a complete understanding of the present technology.

[0049] There is a pressing need for more effective treatments for mental disorders characterized by negative effects (e.g., Affective Disorder (AD)) such as major depressive disorder (MDD), post-traumatic stress disorder and anxiety disorders. Such AD occurs frequently, makes it difficult to work and is costly. In fact, it is estimated that 350 million people worldwide suffer from depression, which is a major factor rendering Americans aged 15 - 44 unemployable.

[0050] Major depressive disorder (MDD) is one of the leading causes of disability worldwide, associated with both severe impairment in work capacity and quality of life (WHO, 2001). MDD is a very common mental disorder, with approximately 17% of the general population affected at some point in their lifetime, often following a relapsing and chronic course (Kessler et al., JAMA. 289, 3095 - 3105 (2003)). Although treatment options are available, it is estimated that only about one - third of MDD patients achieve remission (Trivedi et al., Am J Psychiatry 163, 28 - 40 (2006); Rush et al., Am J Psychiatry 163, 1905 - 1917 (2006)). Therefore, there is a need for methods and systems for appropriate treatment intervention for each patient. [Non - Patent Document 1] Kessler et al., JAMA. 289, 3095 - 3105 (2003) [Non - Patent Document 2] Trivedi et al., Am J Psychiatry 163, 28 - 40 (2006) [Non - Patent Document 3] Rush et al., Am J Psychiatry 163, 1905 - 1917 (2006)

[0051] The present disclosure shows that cognitive-emotional training is involved in short-term plasticity changes in brain networks affected by mood disorders such as MDD, bipolar disorder, post-traumatic stress disorder (PTSD), generalized anxiety disorder, social phobia, obsessive-compulsive disorder, treatment-resistant depression, or borderline personality disorder. Fourteen MDD patients were given cognitive-emotional training (such as Emotional Faces Memory Task (EFMT) training) as monotherapy over 6 weeks. The patients were scanned at baseline and after treatment to examine changes in resting-state functional connectivity and effective connectivity during emotional working memory processing. Compared to baseline, connectivity within the resting-state network involved in self-reference and salience processing decreased in patients after treatment, while the integration across the resting-state functional connectome improved. Furthermore, after treatment, an increase in modulation of connectivity induced by EFMT was observed between the cortical control regions and the limbic regions of the brain, and clinical improvement was noted. These results indicate that cognitive-emotional training improves the functional integration of the resting-state network and the effective connectivity from the cortical control regions of the brain to the regions involved in emotional responses, and that changes in these connectivity parameters are related to symptom improvement. Cognitive-emotional training also changes the short-term plasticity of brain networks affected by other disorders. For example, the disorders may include anxiety disorders such as generalized anxiety disorder (GAD), social phobia, borderline personality disorder, and post-traumatic stress disorder (PTSD).

[0052] The systems and methods disclosed herein are useful for rapidly and accurately detecting emotional disorders based on the effective and / or functional relatedness between selected sub-treatment regions of a patient. Further, the systems and methods of the present technology enable a clinician to quickly perform appropriate treatment interventions (such as cognitive-emotional training) on patients with emotional disorders. The systems and methods described herein also assist in alleviating delays in clinical decision-making regarding the maintenance or modification of the treatment regimen of patients suffering from emotional disorders (e.g., changing, replacing, or discontinuing a certain treatment course, or incorporating another therapy), thereby ensuring the safety of the patient and reducing the overall suicide risk.

[0053] Definition Unless otherwise defined, all technical and scientific terms used herein generally have the same meaning as commonly understood by one of ordinary skill in the art to which the present technology belongs. The singular forms such as "a", "an", and "the" used in this specification and the appended claims also include the plural form unless the context clearly dictates otherwise. For example, the term "cell" includes combinations of two or more cells. In general, the academic terms used herein, as well as the test methods in cell culture, molecular genetics, organic chemistry, analytical chemistry, nucleic acid chemistry, and hybridization described below, are well-known and widely adopted in the field of the present technology.

[0054] As used herein, the term "about" with respect to a number generally includes numbers within a range of plus or minus 1%, 5%, or 10% of that number (except when such a number is less than 0% or greater than 100% of a possible value) unless otherwise stated or apparent from the context.

[0055] As used herein, the term "brain activity" refers to the electrical activity of one or more neurons within at least one brain region observed in a subject in response to internal or external stimuli, and the corresponding metabolic changes.

[0056] As used herein, "cognitive-emotional training" refers to the implementation of cognitive-oriented or emotion-oriented tasks for the purpose of improving psychotic symptoms by inducing activity within specific brain regions and modulating the activity patterns over time within / among regions (controlling brain plasticity).

[0057] As used herein, "control" is an alternative sample used in an experiment for comparison purposes. The control can be "positive" or "negative". For example, if the purpose of the experiment is to examine the correlation of the effect of a therapeutic intervention for a specific type of disease, a positive control (an intervention known to exhibit the desired therapeutic effect) and a negative control (subjects or samples that do not receive treatment or receive a placebo) are typically used.

[0058] As used herein, "effective connectivity" or "EC" refers to the influence that one nervous system or brain region exerts on, or has on, another nervous system or brain region. EC does not report the correlations observed (as in FC) for the influence between brain regions / nervous systems, but rather depends on a priori defined models and tests these. To determine whether one brain region affects another, the correlation of neural activity between the two brain regions is measured during a particular behavior or cognitive task known or thought to activate that network or system (such as working memory). In some embodiments, effective connectivity may be measured from fMRI data. Methods for measuring and interpreting EC are described in Friston, Human Brain Mapping 2:56 - 78 (1994). Effective connectivity may be regarded as an indicator of short - term neural network - level plasticity (Stephan et al., Biological Psychiatry 59, 929 - 939 (2006); Friston, Brain Connect. 1(1):13 - 36 (2011)). This short - term plasticity represents a fundamental mechanism that enables the brain to change or contextualize its connectivity and functionality in response to external or internal cues (Salinas & Sejnowski, Neuroscientist 7, 430 - 440 (2001)). [Non - Patent Document 4] Friston, Human Brain Mapping 2:56 - 78 (1994) [Non - Patent Document 5] Stephan et al., Biological Psychiatry 59, 929 - 939 (2006) [Non - Patent Document 6] Friston, Brain Connect. 1(1):13 - 36 (2011)) [Non - Patent Document 7] Salinas & Sejnowski, Neuroscientist 7, 430 - 440 (2001)

[0059] As used herein, "functional connectivity" or "FC" refers to the temporal correlation between spatially separated neurophysiological events (activities in distinct brain regions). FC is a description of the observed correlation. Functional connectivity may be measured using functional magnetic resonance imaging (fMRI) in which the blood oxygen level dependent (BOLD) signal is measured as a quantitative indicator of neural activity in a particular target brain region. Any neural activity signal (such as an electrophysiological signal like EEG) can be used in the calculation of FC. To estimate FC, the correlation between signals from distinct brain regions is calculated. Methods for the measurement and interpretation of FC are described in Friston, Human Brain Mapping 2:56-78 (1994). [Non-Patent Document 8] Friston, Human Brain Mapping 2:56-78 (1994)

[0060] As used herein, "integrity" refers to the degree of functional or effective connectivity within a single network or brain region, or between multiple networks or brain regions.

[0061] As used herein, a "neuron" is an electrically excitable cell of the nervous system that communicates information with other cells via synapses. A typical neuron includes a cell body, multiple short branched projections (dendrites), and a single long projection (axon).

[0062] As used herein, the term "plasticity" refers to the time-dependent strengthening or weakening of the synapses of a neuron.

[0063] As used herein, a "synapse" is a special region between the tip of a neuron's axon and an adjacent neuron or target effector cell (such as a muscle cell), through which stimuli (i.e., electrical and / or chemical signals) are transmitted. The stimuli may be transmitted by neurotransmitters, or may be transmitted through gap junctions connecting the cytoplasm of the presynaptic cell and the cytoplasm of the postsynaptic cell.

[0064] As used herein, the terms "individual", "patient", or "subject" can be an individual organism, vertebrate, mammal, or human. In some embodiments, the individual, patient, or subject is human.

[0065] As used herein, an "effective amount of time" is the time based on one or more combinations of the frequency, length, and / or content (such as the number or quality of images) of cognitive-affective training sessions capable of achieving a desired endpoint of effective relatedness, functional relatedness, or integration in at least one lower processing region of the subject's nervous system.

[0066] Abnormal brain activity in MDD Numerous studies have shown that patients with MDD present a persistent lack of cognitive control (the ability to hold and manipulate information) in the presence of emotionally salient stimuli, and that such lack is associated with the severity of the illness (Hamilton et al., Am J Psychiatry. 169(7):693 - 703(2012); Bora et al., Psychol Med. 43(10):2017 - 26(2013)). Functional magnetic resonance imaging (fMRI) studies in MDD have shown that regions of the dorsal cortex known to facilitate cognitive control are hypoactive, while regions involved in emotional processing, particularly the amygdala (AMG), are hyperactive (Deiner et al., Neuroimage. 61(3):677 - 85(2012); Fitzgerald et al., Hum Brain Mapp. 29, 683 - 695(2008)). These abnormalities have been observed in multiple tasks, but most commonly studied using working memory (Wang et al., Prog Neuropsychopharmacol Biol Psychiatry. 56:101 - 8(2015)) and emotional processing paradigms (Stuhrmann et al., Biol Mood Anxiety Disord. 1:10(2011); Delvecchio et al., Eur Neuropsychopharmacol. 22(2):100 - 13(2012)). These local brain activity abnormalities also apply to the functional connectivity between the dorsal cortex region and the AMG, which is characterized by a decrease in "top - down" control input from the cortical region to the AMG (Zhang et al., Neurosci Bull. 32(3):273 - 85(2016)).

Non - Patent Document 9

Non - Patent Document 10

Non - Patent Document 11

Non - Patent Document 12

Non - Patent Document 13

Non - Patent Document 14

Non - Patent Document 15

Non - Patent Document 16

[0067] MDD is also associated with abnormalities in resting - state functional connectivity in networks related to cognitive control (central executive network; CEN), salience (salience network; SAL), and self - reference processing (default mode network; DMN) (Kaiser et al., JAMA Psychiatry 72:603 - 611 (2015)). Compared with healthy individuals, patients with MDD show reduced connectivity within the CEN region and increased connectivity between the medial brain regions that form part of the dorsal DMN (Kaiser et al., JAMA Psychiatry 72:603 - 611 (2015)). These alterations in internal network coupling are thought to occur in association with a decrease in the functional integration between resting - state networks (Kaiser et al., JAMA Psychiatry 72:603 - 611 (2015)). Collectively, these task - related abnormalities and resting - state abnormalities in brain functional connectivity represent network - level correlates of emotional dysregulation commonly observed in the MDD population.

Non - Patent Document 17

[0068] Cognitive-behavioral therapy Cognitive-affective therapy is very promising as a therapeutic intervention for MDD because it is theoretically possible to target and alleviate neural network abnormalities. Examples of various forms of cognitive-affective therapy include, but are not limited to, the Emotional Facial Memory Task (EFMT), the Wisconsin Card Sorting Task, the Affective Stroop Task, the Iowa Gambling Task, the Dot-Probe Task, the Face Perception Task, and the Delay Discounting Task.

[0069] The EFMT was developed as a cognitive-affective training aimed at promoting cognitive control (and thereby improving emotional control) for emotional information processing in MDD by targeting both the cognitive control network and the emotional processing network. The EFMT intervention combines a working memory (n-back) task and a facial expression recognition task, which have been shown to specifically elicit activity in the dorsolateral prefrontal cortex (DPFC) and the AMG, respectively. The EFMT requires participants to remember the emotions observed on a series of faces presented one at a time on a computer screen, and then to answer whether the emotion observed on a given face matches the expression shown on the face N (times) before. The difficulty level (N) of the task is adjusted according to the performance of each participant so that the target neural network can consistently work hard. The EFMT training regimen manipulates emotionally salient stimuli in working memory, and thus it is thought that participants are exerting cognitive control while participating in the task and performing emotional information processing. In a sample of healthy volunteers, one version of this task activated both the DPFC and the AMG simultaneously.

[0070] The Wisconsin Card Sorting Task examines the abstract reasoning and problem-solving abilities of subjects in a changing environment. The Wisconsin Card Sorting Task is useful for activating the frontal lobe (such as the prefrontal cortex region), as described in Chen & Sun, C.W., Sci Rep 7, 338 (2017); Teubner-Rhodes et al., Neuropsychologia 102, 95-108 (2017).

Non-Patent Document 18

Non-Patent Document 19

[0071] The emotional Stroop task is a cognitive interference task that measures the emotional processing ability of subjects when conflicting information is also presented by examining the length of time it takes for participants to name the color of a word presented in the presence of emotional interference information. The emotional Stroop task is useful for activating the precentral gyrus and the anterior cingulate gyrus, as described in Ben-Haim et al., J Vis Exp. 112 (2016); Song et al., Sci Rep 7, 2088 (2017).

Non-Patent Document 20

[0072] The Iowa Gambling Task (IGT) is a decision-making task that involves the complex interaction of motivational processing, cognitive processing, and response processing in choice behavior and is useful for activating the amygdala and the ventral-medial prefrontal cortex (vmPFC). The IGT is also thought to be an indicator of dopaminergic activity, as described in Fukui et al., Neuroimage 24, 253 - 259 (2005) and Ono et al., Psychiatry Res 233, 1 - 8 (2015).

Non-Patent Document 21

Non-Patent Document 22

[0073] The dot-probe task measures the selective attention of a subject to an emotional stimulus and is useful for activating the anterior cingulate cortex (ACC) and the amygdala. The dot-probe task is described in Gunther et al., BMC Psychiatry 15, 123 (2015).

Non-Patent Document 23

[0074] The face recognition task measures the ability of a subject to understand and interpret facial expressions and activates the fusiform face area (FFA), the amygdala, and the superior temporal sulcus (fSTS). See Dal Monte et al., Nat Commun 6, 10161 (2015); Hortensius et al., Philos Trans R Soc Lond B Biol Sci 371 (2016); Taubert et al., Proc Natl Acad Sci U S A 115, 8043 - 8048 (2018).

Non-Patent Document 24

Non-Patent Document 25

Non-Patent Document 26

[0075] The delay discounting task assesses the ability of a subject to set and achieve a goal, particularly a reward in the immediate future, compared to a larger but delayed reward. The delay discounting task activates the orbital frontal cortex (OFC), as described by Altamirano et al., Alcohol Clin Exp Res 35, 1905 - 1914 (2011).

Non - Patent Document 27

[0076] System and method FIG. 8 is a block diagram of an exemplary data processing system 800. The data processing system 800 can be used for the treatment of the above - mentioned emotional disorders and the analysis of data related to the treatment. For example, the data processing system may be used for the implementation of cognitive - behavioral therapy and the analysis of data related to the therapy. The data processing system 800 may include one or more processing units 802, a user interface 804, a network interface 806, a storage device 808, a memory 810, and a system bus 830.

[0077] The processing unit 802 is any logic circuit that processes the instructions fetched from the memory 810 in response thereto. In many embodiments, the processing unit 802 is provided by a microprocessor unit such as, for example, a microprocessor unit manufactured by Intel Corporation, Mountain View, California; a microprocessor unit manufactured by Motorola Corporation, Schaumburg, Illinois; an ARM processor and TEGRA system-on-chip (SoC) POWER7 processor manufactured by Nvidia, Santa Clara, California; a microprocessor unit manufactured by International Business Machines, White Plains, New York; or a microprocessor unit manufactured by Advanced Micro Devices, Sunnyvale, California. The processing unit 802 may be based on any of these processors or on any other processor operable as described herein. The processing unit 802 may utilize instruction-level parallelism, thread-level parallelism, different cache levels, and multi-core processors. A multi-core processor may include two or more processing units on a single computing element. Examples of multi-core processors include AMD PHENOM IIX2, INTEL CORE i5, and INTEL CORE i7.

[0078] The user interface 804 may include a display, input / output devices, and peripheral devices that can enable communication with one or more users. The user interface may include a display device, a touch screen display, a mouse, a keyboard, a gesture-sensitive device, and the like. The network interface 806 may enable an interface with an external network such as the Internet, an Ethernet network, or any other local area network or wide area network. The storage device 808 may include non-volatile memory such as a disk drive, a flash drive, a ROM, an EPROM, an EEPROM, and the like. The system bus 830 may provide communication among various components of the data processing system 800.

[0079] The memory 810 may store data, data executable by the processing unit 802 to execute one or more functions, and one or more software modules. Specifically, the memory 810 may include an executable logic engine that can be executed by the processing unit 802. For example, the memory 810 may include a subject data structure 812, a ranking engine 814, a validity engine 816, a cognitive emotion engine 818, a classification engine 820, and an analysis engine 824. These components of the memory 810 will be further described below. The memory 810 may be a hardware storage device and may include volatile and / or non-volatile memory. As an example, the memory 810 may include volatile memory such as RAM, DRAM, SRAM, and non-volatile memory such as those described above with respect to the storage device 808.

[0080] FIG. 9 shows an exemplary data structure 900. As an example, the data structure 900 may be stored in the memory 810 of the data processing system 800 shown in FIG. 8. The data processing system 800 may maintain the data structure 900 to manage associating one or more association values with one or more subjects. The data structure may include a plurality of columns with a plurality of fields for storing data values assigned to each column. However, of course, any kind of single or multiple data structures can be used. The data structure 900 stores a key field 902 that stores subject identification information for identifying a subject. The subject identification information may include alphanumeric and / or binary digits that can uniquely identify the subject. The data structure 900 may include keyed data corresponding to the key or subject identification information in the key field 902. For example, the keyed data associated with certain subject identification information may include data values in the row corresponding to this subject identification information. As an example, the data structure 900 may further include a field 904 for storing data values representing a set of first association values associated with one or more subject identification information. The first association value may represent, for example, the magnitude of the association related to at least one sub-processing area of the subject's nervous system. In some examples, the first association value may represent the magnitude measured at a first point in time.

[0081] The data structure 900 may include a field 906 for storing data values representing ranks associated with one or more subject identification information. The rank may be based on a first correlation value and may represent the position of the subject identification information in relation to the first correlation value associated with other subject identification information. For example, the rank may represent the position of the subject identification information based on an increase in the first correlation value associated with all subject identification information. The data structure 900 may further include a field 908 for storing data values representing a classification associated with one or more subject identification information. The classification may be further based on the first correlation value. The classification may include, for example, a first classification value such as "qualified" and a second classification value such as "unqualified". The classification may indicate, for example, whether the subject identified by the subject identification information is a suitable or good candidate for cognitive-emotional training.

[0082] Similar to the first correlation value, the data structure 900 may further include a field 910 for storing data values representing a second correlation value that can represent the magnitude of the correlation associated with at least one sub-processing region of the subject's nervous system. In some examples, the second correlation value may represent the magnitude measured, for example, at a second time point after the subject has received cognitive-emotional training. In some examples, the second correlation value may represent the magnitude of the correlation associated with at least one sub-processing region that is the same as the sub-processing region associated with the first correlation value. For example, both the first correlation value and the second correlation value may represent the magnitude of the correlation between the subject's DPFC and AMG.

[0083] The data structure 900 may further include another classification field 912 for storing data values representing the effectiveness related to one or more subject identification information. The effectiveness may be indicated based on first and second correlation values measured before and after each implementation of the treatment, such as the effectiveness of a treatment such as cognitive-emotional training for a subject. The data structure 900 may indicate further correlation values, classifications, and ranks. In some examples, the data structure 900 may include values related to the correlation of at least one sub-processing region of the subject's nervous system. For example, the data structure may include a first correlation value and / or a second correlation value that measures the correlation between the subject's dACC and AMG.

[0084] It should be understood that the correlation value may correspond to the effective correlation, functional correlation, or integration of at least one sub-processing region of the subject's nervous system. In some embodiments, the correlation value may be obtained from a device configured to perform fMRI on a subject. In some embodiments, the correlation value may be estimated from one or more outputs provided by fMRI. It should be understood that the correlation value may be estimated or specified from any device configured to determine the effective correlation, functional correlation, and integration of at least one sub-processing region.

[0085] FIG. 10 is a flowchart of an exemplary process 1000 representing a classification engine 820 available for determining a subject's eligibility for treatment. Specifically, flowchart 1000 is available to update data structure 900 with a classification value indicating whether a relevant subject is eligible or ineligible for treatment, such as cognitive-emotional training. Process 1000 includes accessing (1002) a subject-relatedness value associated with the subject. Specifically, process 1000 may include accessing a data structure including subject identification information identifying the subject and a subject-relatedness value derived from at least one of a scan or test performed on the subject, the subject-relatedness value representing the magnitude of relatedness associated with at least one sub-processing region of the subject's nervous system. The data structure may be, for example, data structure 900 shown in FIG. 9. The subject-relatedness value may represent, for example, a first relatedness value or a second relatedness value within data structure 900. At least one sub-processing region of the nervous system may non-exclusively include at least one of DPFC, AMG, dACC, DMN, SAL, LCEN, RCEN, dDMN, and vDMN. As an example, data processing system 800 may access the first relatedness value 10 within data structure 900 associated with a subject identified by subject identification information "Subject 1".

[0086] Process 1000 executes executable logic including a comparison rule and compares (1004) the subject-relatedness value to a threshold. Specifically, process 1000 may include the data processing system comparing the subject-relatedness value to a relatedness threshold to determine the classification of the subject. As an example, data processing system 800 may store the relatedness threshold in memory 810 and compare the first relatedness value 10 stored in data structure 900 to this threshold. The threshold may represent a baseline number that may be based on the measurement environment and settings in which the subject is placed and may vary in response to changes in the measurement environment or settings. For example, different relatedness values may result from using different measurement instruments on the same subject. Therefore, the threshold may be selected based on the measurement environment and settings.

[0087] Process 1000 includes storing a first classification value in a data structure (1006) when the subject relevance value is greater than a threshold. Specifically, in response to the subject relevance value being greater than the relevance threshold, data processing system 800 stores the relationship between the subject identification information and the first classification value corresponding to the first classification in the data structure. As an example, the first classification may indicate ineligibility, and the first classification value may include the entry "ineligible". For example, referring to data structure 900 shown in FIG. 9, data processing system 800 compares the first relevance value 6 of subject 2 with an exemplary threshold 5 and may store the entry "ineligible" in the classification column (field 908) of data structure 900 associated with the subject identification information "subject 2".

[0088] Process 1000 also includes storing a second classification value in the data structure (1006) when the subject relevance value is less than the threshold. Specifically, in response to the subject relevance value being less than the relevance threshold, the data processing system stores the relationship between the subject identification information and the second classification value corresponding to the second classification in the data structure. As an example, the second classification may indicate eligibility, and the second classification value may include the entry "eligible". For example, in data structure 900, data processing system compares the first relevance value 15 of subject 3 with an exemplary threshold 20 and may store the entry "eligible" in the classification column (field 908) of data structure 900 associated with the subject identification information "subject 3".

[0089] Process 1000 may also be repeatedly executed to update data structure 900 as new relevance values are received. Thus, process 1000 may update field 908 with an appropriate first or second classification value based on the change in the subject's relevance value relative to the relevance threshold.

[0090] In some examples, the sub-processing region may include the dorsolateral prefrontal cortex (DPFC) and the amygdala (AMG), and the correlation value and the correlation threshold represent the effective correlation between the DPFC and the AMG and the effective correlation threshold between the DPFC and the AMG. The first classification value may indicate that the subject is ineligible, and the second classification value may indicate that the subject is eligible. That is, when the effective correlation value is less than the threshold, the relevant subject is eligible for cognitive-emotional training, and the data structure 900 may be updated accordingly.

[0091] In some examples, the sub-processing region may include the dorsal anterior cingulate cortex (dACC) and the amygdala (AMG), and the correlation value and the correlation threshold represent the effective correlation between the dACC and the AMG and the effective correlation threshold between the dACC and the AMG. The first classification value may indicate that the subject is eligible, and the second classification value may indicate that the subject is ineligible. That is, when the effective correlation value is greater than the threshold, the relevant subject is eligible for cognitive-emotional training, and the data structure 900 may be updated accordingly.

[0092] In some examples, the sub-processing region may include the default mode resting state network (DMN), and the correlation value and the correlation threshold represent the functional correlation within the DMN and the functional correlation threshold related to the DMN. The first classification value may indicate that the subject is eligible, and the second classification value may indicate that the subject is ineligible. That is, when the functional correlation value is greater than the threshold, the relevant subject is eligible for cognitive-emotional training, and the data structure 900 may be updated accordingly.

[0093] In some examples, the sub-processing region may include a default mode network (DMN) or a salience network (SAL), and the relevance value and the relevance threshold represent the functional relevance within the DMN or SAL and the functional relevance threshold associated with the DMN or SAL. The first classification value may indicate that the subject is eligible, and the second classification value indicates that the subject is ineligible. That is, when the functional relevance value is greater than the threshold, the associated subject is eligible for cognitive-emotional training, and the data structure 900 may be updated accordingly.

[0094] In some examples, the sub-processing region may include one of LCEN and RCEN, dDMN and vDMN, LCEN and vDMN, and LCEN and SEL. The relevance value and the relevance threshold respectively represent the integration within the selected pair of sub-processing regions. The first classification value indicates that the subject is ineligible, and the second classification value indicates that the subject is eligible. That is, when the integration value is less than the threshold (e.g., threshold 0), the associated subject is eligible for cognitive-emotional training, and the data structure 900 may be updated accordingly.

[0095] In some examples, the classification 820 engine determines the eligibility of the subject by considering the relevance values between or within two or more sets of sub-processing regions. For example, the classification engine 820 determines the eligibility by considering the combination of the effective relevance between DPFC and AMG and the effective relevance between dACC and AMG. When the effective relevance between DPFC and AMG is less than the threshold and the effective relevance between dACC and AMG is greater than the threshold, the classification engine 820 may determine that the subject is eligible and update the data structure 900 accordingly.

[0096] FIG. 11 is a flowchart of an exemplary process 1100 representing a ranking engine 814 of the data processing system shown in FIG. 8 that can be used to rank subjects based on their measured subject correlation values. Process 1100 may include accessing (1102) subject correlation values related to a subject. Specifically, process 1100 may include obtaining from a data structure a plurality of subject correlation values related to a plurality of subject identification information corresponding to a plurality of subjects, where each subject correlation value of the plurality of subject correlation values represents the magnitude of the correlation related to at least one sub-processing region of the nervous system of each subject of the plurality of subjects. For example, the ranking engine 814 may access a plurality of first correlation values listed in a data structure 900 related to a plurality of subject identification information from "subject 1" to "subject n".

[0097] Process 1100 includes performing ranking (1102) based on the correlation values. Specifically, process 1100 may include assigning a rank to each subject identification information in the data structure based on the subject correlation values. For example, referring to the data structure 900 of FIG. 9, the ranking engine 814 may assign ranks to a rank column based on an increase in the first correlation value, with the lowest correlation value assigned the lowest rank and the highest correlation value assigned the highest rank. In some cases, the ranking may be assigned in reverse order.

[0098] Process 1100 may include determining (1106) the difference between the subject correlation value and a threshold value. Specifically, the ranking engine 814 may determine the difference between the first correlation value shown in FIG. 9 and the threshold value. In some cases, the threshold value may be the same as the threshold value selected by the classification engine 820 described above. Process 1100 includes determining (1108) whether the difference is greater than 0, i.e., whether the correlation value is greater than or equal to the threshold value.

[0099] Process 1100 includes selecting (1100) a subset of a plurality of subjects, where each subject in the subset is selected based on the difference between the relevance value and the subject relevance value. As an example, the ranking engine 814 may select m subjects having the m largest differences between the relevance value and the threshold value. After the selection of the subjects, the ranking engine 814 can generate (1112) a numbered list of the selected subject identification information based on each rank.

[0100] Process 1100 may also be repeatedly executed to update the data structure 900 as new relevance values are received. Thus, process 1000 may update field 906 at an appropriate rank based on the first relevance value of the received additional subject or based on a new first relevance value associated with the existing subject identification information in the data structure 900.

[0101] In some examples, at least one sub-processing region may include the DPFC and the AMG or the dACC and the AMG, and the subject relevance value may represent the effective relevance between the sub-processing regions. In the case of the DPFC and the AMG, the subject relevance value associated with the selected subset of a plurality of subjects is less than the relevance threshold value. In the case of the dACC and the AMG, the subject relevance value associated with the selected subset of a plurality of subjects is greater than the relevance threshold value.

[0102] In some examples, at least one sub-processing region may be the dDMN or the SAL, and the subject relevance value may represent the functional relevance within the DMA or the SAL. In some such examples, the subject relevance value associated with the selected subset of a plurality of subjects is greater than the relevance threshold value.

[0103] In some examples, pairs of sub - processing regions may be LCEN and RCEN, dDMN and vDMN, LCEN and vDMN, or LCEN and SAL. The subject - related value may represent the functional correlation within these pairs of sub - processing regions. In some such examples, the subject - related value associated with a selected subset of multiple subjects is greater than a correlation threshold.

[0104] Process 1100 may also include not selecting subjects having a correlation value smaller than a threshold. In some examples, this includes excluding candidates with low responsiveness to treatments such as cognitive - emotional training. Thus, by removing those subjects from the list of all subjects, the time for performing the treatment may be improved.

[0105] FIG. 12 shows a flowchart of an exemplary process 1200 that represents the effectiveness engine shown in FIG. 8 and that can be used to determine the effectiveness of cognitive - emotional training for a subject. Process 1200 includes obtaining (1202) a first correlation value and a second correlation value related to the subject. Specifically, the effectiveness engine 816 obtains a first correlation value of the subject, which represents the first magnitude of the correlation related to at least one sub - processing region of the subject's nervous system at a first point in time, and a second correlation value of the subject, which represents the second magnitude of the correlation related to at least one sub - processing region of the subject's nervous system at a second point in time after the subject has received cognitive - emotional training. For example, referring to FIG. 9, the effectiveness engine 816 may access the data structure 900 to obtain the first correlation value and the second correlation value of the subject, for example, related to the subject identification information "Subject 1".

[0106] Process 1200 includes obtaining (1204) the difference between the first correlation value and the second correlation value. For example, the effectiveness engine 816 of "Subject 1" may obtain the difference between the first correlation value and the second correlation value as 10.

[0107] Process 1200 includes determining (1206) whether the difference is less than or greater than a threshold value when compared. Specifically, the validity engine 816 can access the stored value, or the operator provides the threshold value. The threshold value may represent the smallest difference in the correlation value that needs to be observed to indicate that there is effective improvement in the subject as a result of the treatment.

[0108] Process 1200 includes classifying (1208) the subject as effective when the difference is greater than the threshold value and classifying (1210) the subject as ineffective when the difference is less than the threshold value. Specifically, in response to a determination that the difference is greater than the threshold value, the validity engine 816 stores in the data structure the relationship between the subject identification information and a first classification value corresponding to the first classification. As an example, the first classification may be that there is an effect, and the first classification value may be the entry "Y". Further, in response to a determination that the difference is less than the threshold value, the validity engine 816 may store in the data structure the relationship between the subject identification information and a second classification value corresponding to the second classification. For example, the second classification may be that there is no effect, and the second classification value may be the entry "N". Naturally, entries other than "Y" and "N" can also be used.

[0109] Process 1200 may also be repeatedly executed to update the data structure 900 as new correlation values are received. Thus, process 1000 may update field 912 with appropriate entries based on the first and / or second correlation values, or based on new first and second correlation values of new keyed data related to new subject identification information.

[0110] In some examples, the sub-processing regions may include the DPFC and the AMG, or the dACC and the AMG. The first correlation value and the second correlation value represent the effective correlation between the subject's DPFC and the AMG, or between the dACC and the AMG. In some examples, the sub-processing regions may include the dDMN or the SAL. The first correlation value and the second correlation value represent the functional correlation within the subject's dDMN or SAL. In some examples, the sub-processing regions may include pairs of LCEN and RCEN, dDMN and vDMN, LCEN and vDMN, or LCEN and SAL. The first correlation value and the second correlation value represent the integrity within a particular pair of the subject's sub-processing regions.

[0111] In some examples, the analysis engine 824 may perform an analysis of data collected before, during, or after treatment. For example, the analysis engine 824 may perform data analysis such as neuroimaging preprocessing and quality assurance, resting state network connectivity analysis, task-based fMRI (connectivity) analysis, statistical analysis, analysis of changes in resting state functional connectivity, effective connectivity, and integrity of the post-treatment data described above. In some examples, the analysis engine may perform an analysis of neuroimaging data to obtain effective correlation values, functional correlation values, and integrity values. As an example, the analysis engine 824 may include software based on the procedures described in "Activity and connectivity of brain mood regulating circuit in depression: a functional magnetic resonance study" by Anand, A., Li, Y., Wang, Y., Wu, J., Gao, S., Bukhari, L., Mathews, V.P., Kalnin, A., and Lowe, M.J., Biol Psychiatry 57, 1079-1088 (2005) and "A default mode of brain function" by Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., and Shulman, G.L., Proc Natl Acad Sci U S A 98, 676-682. Non-Patent Document 28 Anand, A., Li, Y., Wang, Y., Wu, J., Gao, S., Bukhari, L., Mathews, V. P., Kalnin, A., and Lowe, M. J., "Activity and connectivity of brain mood regulating circuit in depression: a functional magnetic resonance study", Biol Psychiatry 57, 1079-1088 (2005) Non-Patent Document 29 Raichle, M. E., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A., and Shulman, G. L., "A default mode of brain function", Proc Natl Acad Sci U S A 98, 676-682

[0112] In some examples, the cognitive-emotional training engine 818 may perform one or more cognitive-emotional therapies such as, for example, the above-described EFMT, Wisconsin Card Sorting Task, Emotional Stroop Task, Iowa Gambling Task, Dot Probe Task, Face Perception Task, and Delay Discounting Task.

[0113] System The system can include any script, file, program, instruction set, or computer-executable code configured to enable a computing device to perform an emotion recognition task and a working memory task on a user. The computing device may include a display or be connected to a display. The computing device can be a laptop, desktop computer, mobile phone, tablet, or other computing device. The system may include a database containing a plurality of face images. The faces depicted by the face images may be in different emotional states, such as happiness, sadness, fear, excitement, etc. Each of the face images may be labeled in the database according to its corresponding emotional state. In some examples, a reference database containing portions of face images in different emotional states may be used to generate face images on demand. For example, a face image in a happy state may be divided into sub-images containing only a single face part, such as eyes, mouth, etc. The system may combine these sub-images on demand to form a unique face image. The system may also include a plurality of training guidelines in the database. The training guidelines may manage the length and order of time for presenting each image to the user.

[0114] The system may present a series of images to the user. During the task using the system, the participant differentiates the emotions observed by the participant from a series of face images presented one by one on the display at a time. Each image may be displayed for about 1 second, followed by a fixation cross for 1 second. When the image is presented to the user, the system may instruct the user to remember the order of the emotions the user observed. Using the N-back working memory paradigm, the participant may be prompted to answer whether the expression of the face just observed is the same as the expression shown by the face N times before after each face is presented. In some examples, each training session includes 15 task blocks, and the training policy may indicate that the level of N changes according to the participant's performance between these, that is, as the participant's accuracy improves or decreases during the block, the difficulty level increases or decreases (respectively) during the block. The first training session starts at a difficulty level of N = 1, and the difficulty level at the start of the next session is determined by the performance of the previous session. Since EFMT utilizes an increasingly difficult working memory paradigm, the task is adjusted to the participant's ability level, enabling consistent effort throughout each training session. The increasingly difficult N-back working memory has been shown to improve working memory capacity.

Example

[0115] Example The technology of the present application will be further described using the following examples, but these examples should not be considered limiting in any way.

Example

[0116] Example 1: Experimental materials and methods Subjects. Twenty-five drug-naive MDD participants currently experiencing a major depressive episode (MDE) were recruited and had fMRI scans performed before and after completing a 6-week emotional face memory task (EFMT) or sham-control training (CT) as part of an investigational protocol (NCT01934491). Participants were recruited online and through local newspaper advertisements for depression research studies. All participants were between 18 and 55 years old and were evaluated by a trained clinician using the Structured Clinical Interview for DSM-IV-TR Axis I Disorders (SCID) (see First, M.B., et al., “Structured clinical interview for DSM-IV axis disorders (SCID),” New York State Psychiatric Institute, Biometrics Research, New York, NY (1995)). Diagnosis of other Axis I comorbid disorders was permitted only if the participant's MDD diagnosis was considered the primary diagnosis (excluding mental disorders, bipolar disorder, and substance abuse or dependence within the past 6 months). The severity of MDD, measured by the Hamilton Depression Rating Scale-17-item version (HAM-D) (see Hamilton, J.P., et al., Am J Psychiatry 169:693-703 (1960)), had to be at least “moderate” (Ham-D ≥ 16). Participants with very severe MDD (HAM-D ≥ 27) were excluded from the study and offered medical consultation. Participants who reported taking any antidepressant medications during the current MDE and those with a history of no treatment response (more than 2 unsuccessful trials of standard antidepressant medications) were excluded from the study. Participants who received cognitive behavioral therapy within 6 weeks prior to the study or at any point during the study were also excluded from the study based on the protocol. Participants with visual or motor impairments that were thought to interfere with the implementation of EFMT training were also excluded.

[0117] During the initial pre-screening interview, eligible potential participants were informed about the research procedures, signed the informed consent form, and completed the screening and baseline procedures. Eligible and formally registered participants in the phase II clinical trial investigating the efficacy of EFMT were then approached for registration into the fMRI protocol and provided informed consent if they chose to participate. Participants were compensated for their time and travel expenses at the completion of each research session.

[0118] Procedure. The investigational intervention (EFMT) was administered over 20 research visits. At the first visit, the SCID and Ham-D were administered to the subjects to confirm the diagnosis of MDD and determine the severity of symptoms. Subsequently, a baseline assessment including a pre-treatment fMRI scan was performed. The research coordinator randomly assigned the participants to the EFMT group or the CT group using a predetermined randomization sequence for grouping. The participants were tasked with completing all 18 training sessions over 6 weeks (each session approximately 20 - 35 minutes, three times a week). Participants who were unable to complete at least two training sessions in any given week or who missed more than three training sessions during the course of the investigation had their training terminated based on the clinical trial protocol. Weekly, severity (Ham-D) assessments were performed by multiple doctoral or master's level clinicians who were unaware of the participants' group assignments. The Ham-D assessors received extensive training in performing the assessments, with an intra-class correlation coefficient (ICC) > 0.8 in two separate training interviews. Outcome assessments were performed within one week of the completion of the training sessions, at which time the baseline assessment and fMRI scan procedures were repeated.

[0119] Cognitive training intervention. EFMT has been described in detail in past publications (see Iacoviello, B.M. et al., Eur Psychiatry 30:75-81 (2015); Iacoviello, B.M. et al., Depress Anxiety 31:699-706 (2014)). EFMT is designed to promote cognitive control for emotional information processing in MDD by targeting both the cognitive control and the emotion processing networks, and is achieved by combining an emotion discrimination task and a working memory task. In the EFMT task, participants were asked to identify the emotions observed in a series of face images presented one by one on the computer screen, and then to remember the order of the observed emotions. Figure 1 shows an exemplary test sequence in the EFMT task. Using the n-back working memory paradigm, participants were prompted to answer whether the expression of the face just observed was the same as the expression shown by the face n times before after each face was presented. Thus, the EFMT task was presumed to be involved in the exertion of cognitive control over emotional information processing and to simultaneously activate the amygdala (AMG) and the dorsolateral prefrontal cortex (DPFC). As the CT condition, the same n-back paradigm as EFMT was used for working memory training, except that neutral shapes were used as stimuli instead of emotional faces. Each EFMT session or CT session took about 15-25 minutes to complete, and the investigation regimen was to complete 18 EFMT sessions or CT sessions in 6 weeks (3 sessions per week over 6 weeks).

[0120] Neuroimaging data acquisition. Imaging data were acquired at the Mount Sinai School of Medicine using a 3T Skyra scanner (Siemens, Erlangen, Germany) with a 32-channel receive coil. Participants were scanned at the time of study enrollment (baseline) and immediately following the completion of 6 weeks of EFMT training. Anatomical rest and task-based fMRI data were acquired. The tasks included a shortened version and a modified version of the EFMT session. Twelve blocks of 10 trials were initiated with a 2.5-second cue indicating the target type (0-back, 1-back, or 2-back). In the 0-back trials, subjects viewed target images (faces expressing emotions) and answered whether each subsequent stimulus was the exact same image as the target image. In the 1-back and 2-back trials, participants answered whether each face expressed the same emotion as it had been presented “one trial ago” or “two trials ago.” Anatomical rest and task-acquired data were identical for all participants at baseline and post-treatment.

[0121] Resting-state fMRI data and task fMRI data were acquired with a T2 * single-shot echo-planar gradient echo imaging sequence with the following parameters: echo time / repetition time = 35 / 1000 milliseconds (ms), isotropic resolution: 2.1 mm, 70 contiguous axial slices of the whole brain, field of view (FOV): 206×181×147 mm 3 , matrix size: 96×84, flip angle: 60 degrees, multiband (MB) factor: 7, blipped CAIPIRINHA (high-speed parallel imaging) phase-encoding shift = FOV / 3, bandwidth at ramp sampling: ~2 kHz / pixel, echo spacing: 0.68 ms, and echo train length: 57.1 ms. The acquisition time at rest was 10 minutes, and the acquisition time for the WM task was 7 minutes and 34 seconds. Structural images were acquired with a T1-weighted 3D turbo fast scan (MPRAGE) sequence (FOV: 256×256×179 mm 3Matrix size: 320×320, isotropic resolution: 0.8 mm, TE / TR = 2.07 / 2400 ms, inversion time (TI) = 1000 ms, flip angle with binomial (1, -1) fat suppression: 8 degrees, bandwidth: 240 Hz / pixel, echo spacing: 7.6 ms, in-plane acceleration (Generalized Autocalibrating Partial Parallel Acquisition) factor: 2, total acquisition time: 7 minutes) were used for acquisition.

[0122] Neuroimaging preprocessing and quality assurance. The task fMRI data and resting-state fMRI (rs-fMRI) data acquired at baseline and after treatment were preprocessed separately in the same way. All analyses were performed using Statistical Parametric Mapping software version 12 (SPM12; see www.fil.ion.ucl.ac.uk / spm / software / spm12 / ) and the Data Processing and Analysis for Brain Imaging Toolbox (see Yan, C.G. et al., Neuroinformatics 14:339-351 (2016)). Each fMRI dataset was motion-corrected using rigid-body transformation; image registration between the functional scan and the anatomical T1 scan; spatial normalization of the functional scan to the Montreal Neurological Institute's stereotactic standard space; and spatial smoothing within a functional mask with a full-width at half maximum Gaussian kernel of 6 mm. For head motion correction, the resting-state data were further preprocessed in the following steps: wavelet spike removal (removal of signal transients associated with small-amplitude (<1 mm) head motion) (see Patel, A.X. et al., Neuroimage 95:287-304 (2014)); trend removal; regressing out the motion parameters and their derivatives (24-parameter model) (see Friston, K.J. et al., Magn.Reson.Med. 35:346-355 (1996)) and the time series trends of white matter (WM), cerebro-spinal fluid (CSF), and their linear trends. The WM signal and CSF signal were calculated using a component-based noise removal method (CompCor, 5 principal components) (see Behzadi, Y. et al., Neuroimage 37:90-101 (2007)). Finally, band-pass filtering ([0.01-0.1] Hz) was performed. Individual task fMRI datasets and rs-fMRI datasets were excluded if the volume / volume head motion exceeded 3 mm or 1 degree. There were no significant differences in maximum or mean head motion between the baseline scan and the follow-up scan (all p>0.2).

[0123] Resting-state network connectivity analysis. The rs-fMRI data obtained at baseline and after treatment were analyzed separately using the same method described below. The method was implemented with a focus on the resting-state networks most highly associated with MDD. Specifically, the ventral default mode network (vDMN) and dorsal default mode network (dDMN), left central executive network (LCEN) and right central executive network (RCEN), and salience network (SAL) were examined (Figure 2A). To ensure the reproducibility of the analysis, these networks were defined using a validated template freely provided by the Functional Imaging in Neuropsychiatry Disorders Lab at Stanford University (<http: / / findlab.stanford.edu / functional_ROIs.html>) (see Shirer, W.R. et al., Cerebral Cortex 22:158-165 (2012)). For each participant, the functional connectivity within and between networks reflecting functional coupling and segregation, respectively, was calculated for each network. For functional connectivity within a network, the mean time series of the voxels within each network region was calculated, and then the Pearson correlation coefficient between pairs of network regions was calculated and averaged. For functional connectivity between networks, first the mean time series within each network was calculated (by averaging all the time series of the voxel portion of the network), and then the Pearson correlation coefficient of the time series between each pair of networks was calculated. Fisher's Z-transform was further performed on both the within-network and between-network measurement results.

[0124] Task-based fMRI (correlation) analysis. The task fMRI data acquired at baseline and after treatment were analyzed separately in the same way described below. This analysis was carried out with a focus on the effective connectivity calculated using Dynamic Causal Modeling (DCM; see Friston, K.J. et al., Neuroimage 19:1273-1302 (2003)). In DCM, the intrinsic connectivity represented the coupling strength between regions unrelated to the task, while the modulatory effect represented the change in connectivity between regions induced by the task (Id.). Next, the modeled neuronal dynamics were associated with the observed blood oxygenation level-dependent (BOLD) signal using the hemodynamic forward model (see Stephan, K.E. et al., Neuroimage 38:387-401 (2007)). After established procedures (see Dima, D. et al., Human Brain Mapping 36:4158-4163 (2015); Dima, D. et al., Transl. Psychiatry 6:e706 (2016); Moser, D.A. et al., Mol. Psychiatry 23:1974-1980 (2018)), bilateral 5-mm spherical volumes of interest (VOIs) were defined centered on the MNI coordinates of the maximal set of working memory load-dependent modulations at baseline (left inferior parietal cortex (PAR): -42, -48, 44, right PAR: 44, -38, 42; dACC left dorsal anterior cingulate cortex (dACC): -6 24 44, right dACC: 6 22 44; left DPFC: -28 4 60, right DPFC: 28 8 58; left AMG: -26 -4 -20, right AMG: 26 -2 -20). To ensure continuity between analyses, the same VOIs were used in the post-treatment DCM. The time series of the regions were summarized by the first eigenvariate of all (p<0.01) activated voxels within the participant-specific VOIs. Using the VOIs defined above, the basic 8-region DCMs of all participants were identified. The inter-regional connectivity between these regions both within and between hemispheres was defined. The effect of the working memory load (driving input) extended to the PAR on both the left and right sides.Starting from this basic layout, a structured model space was derived by considering the modulation effect of working memory load on the strength of inter-regional coupling. Since random effects Bayesian Model Averaging (BMA) is applicable to the uncertainty of the model when estimating the coherence and strength of associations, BMA was then executed to obtain the average association estimate of all models for each participant (see Penny, W.D. et al., PLoS Comput Biol 6: e1000709 (2010); Stephan, K.E. et al., Neuroimage 49: 3099 - 3109 (2010)). Using the post-hoc means from the average DCM from the resulting baseline dataset and the post-treatment dataset, the change in the modulation effect of working memory load on inter-regional coherence was tested. To be complete, the differences in the working memory load-dependent modulation of brain activity at baseline and after treatment were examined using a general linear model, and the results are shown in Figure 4.

[0125] Statistical analysis. The effect size of repeated measures based on Cohen's d was calculated, and the post-treatment change in a given functional measurement was estimated using Equation (1):

[0126]

Number

[0127] Where m1 and m2 are the baseline and post-treatment mean values, respectively, of a given measurement; s is the baseline and post-treatment mean standard deviation of a given measurement, and γ is the baseline and post-treatment value of a given correlation value measurement. Based on the normal interpretation of Cohen's d, only results with an effect size greater than 0.3 are shown as being more likely to be significant (see "Statistical power and analysis for the behavioral sciences." by Cohen, J., Hillsdale, N.J., Lawrence Erlbaum Associates, Inc., (1988)). The Pearson correlation coefficient was used to evaluate the relationship between changes in symptom levels and changes in brain imaging measurement results. Since the investigation had the nature of a diagnostic examination, the threshold for statistical inference was p < 0.05, without correction. Different clinical measurement results between the baseline scan and the post-treatment scan were compared based on the paired t-test.

Example

[0128] Example 2: Observation after treatment of resting-state functional connectivity and effective connectivity Twenty-five participants provided signed consent to participate in this investigation. Two participants underwent baseline fMRI scans but discontinued their participation in the parent clinical trial prior to randomization. Of the participants, 16 were assigned to the EFMT condition of the parent clinical trial and 7 participants were assigned to the control (CT) condition. Five participants dropped out and did not complete the clinical trial protocol or the resulting fMRI scans (2 were participants in the EFMT group and 3 were participants in the CT group). The current investigation sample included 14 participants who completed the EFMT regimen and had available valid pre / post fMRI data and behavioral data. Four sham control participants also had valid pre / post image data and behavioral data. The current report included 14 participants who had valid pre / post image data and behavioral data. Figure 5 shows the population statistics and clinical characteristics of 14 MDD participants who received EFMT treatment in this investigation.

[0129] In the parent clinical trial from which the current study participants were derived, EFMT was observed to significantly improve MDD symptoms from baseline to the study outcome compared to CT (see Iacoviello, B.M. et al., npj Digital Medicine 1:21 (2018)). Fourteen participants in the current sample also showed an average clinical response to the EFMT intervention (Ham-D improved from a mean score of 19.14 (SD = 2.6) at baseline to a mean score of 11.43 (SD = 5.12) at the study outcome; t(13) = 6.88, p <.001) (Figure 5).

[0130] Changes in resting-state functional connectivity. As shown in Figure 2B, within-network connectivity was observed to decrease after treatment in the dDMN (d = -0.38) and SAL (d = -0.36). On the other hand, connectivity increased between LCEN and RCEN (d = 0.30), between vDMN and dDMN (d = 0.32), and between LCEN and both vDMN (d = 0.45) and SAL (d = 0.53) (Figure 2B). However, the correlation between the changes in resting-state connectivity after treatment and the changes in symptoms was generally low and not statistically significant.

[0131] Effective connectivity. After treatment, effective connectivity from dACC to AMG decreased bilaterally (left: d = -0.44; right: d = -0.32), and top-down connectivity from DPFC to AMG increased on the right side (d = 0.33) (Figure 3). The post-treatment changes in effective connectivity from both DPFC and DACC to AMG were correlated with the reduction in depressive symptoms measured by the total score of HAM-D, which was significant at an uncorrected threshold (r = 0.51, p = 0.05).

[0132] Figure 6 shows the resting-state community analysis of 42 healthy control subjects. These data indicate that in healthy subjects, the medial temporal lobe network (hippocampus - amygdala - temporal pole) is well integrated. Figures 7A - 7B show the resting-state community analysis of 7 MDD participants from the EFMT study before treatment (left panel) and after treatment (right panel). These results show that MDD participants before EFMT treatment had low integration of the medial temporal lobe network, as can be seen from the absence of black dots within the left and right brain regions surrounded by the dashed circle, but that EFMT treatment partially restored the medial temporal lobe network in MDD participants.

[0133] Figure 6 shows the resting-state community analysis of 42 healthy control subjects. These data indicate that in healthy subjects, the medial temporal lobe network (hippocampus - amygdala - temporal pole) is well integrated. Figure 7 shows the resting-state community analysis of 7 MDD participants from the EFMT study before treatment (left panel) and after treatment (right panel). These results show that MDD participants before EFMT treatment had low integration of the medial temporal lobe network, but that EFMT treatment partially restored the medial temporal lobe network in MDD participants.

[0134] These results indicate that EFMT training alters neuroplasticity in MDD patients. Past studies have suggested that the degree of lateralization of prefrontal cortex dysfunction in MDD with abnormalities in the right DPFC is mainly related to the spontaneous control of emotional processing (Grimm et al., Biol Psychiatry. 63(4):369 - 76(2008)). The results described herein show that the modulation of the connectivity from the right DPFC to the right AMG induced by working memory increased after EFMT, accompanied by symptom improvement.

[0135] Changes after further treatment are associated with a reduction in the functional coupling / effective connectivity between the dACC and the AMG. Past studies have shown that in MDD, the dACC does not show expected changes in its connectivity for multiple different tasks (Shine et al., Neuron 92:544-554 (2016)), suggesting a lack of adaptive flexibility from very early stages (Ho et al., Neuropsychopharmacology. 42(12):2434-2445 (2017)). After EFMT training, a reduction in connectivity from the dACC to the AMG was observed along with symptom improvement. These data suggest that the reduction in effective connectivity of the dACC after EFMT training in MDD patients may reflect an improvement in dACC function. The symptom improvement observed in this investigation is thought to be related to the restoration of regulatory control in the limbic region, as indicated by an increase in the connectivity between the AMG and the DPFC and a decrease in the connectivity between the AMG and the dACC. See FIGS. 3A-3B.

[0136] A decrease in the functional connectivity of dDMA and SAL after treatment, as well as an increase in the integration among networks involved in cognitive control, self-reference, and salience processing, were also observed. Hypo-connectivity and reduced integration of the fronto-parietal resting-state network have been found to be highly correlated with MDD (Kaiser et al., JAMA Psychiatry 72:603-611 (2015)). Therefore, it should be noted that most of the changes in network resting-state functional connectivity after EMFT are involved in the CEN, which is considered to be an important network for cognitive control (Smith et al., Proc Natl Acad Sci U S A. 106(31):13040-5 (2009)). The CEN was more integrated between the left and right hemispheres and between the DMN and SAL. Such an increase in the integration among networks for cognitive control, self-reference, and salience processing may facilitate a more coordinated and coherent response to emotional stimuli in patients with MDD. Furthermore, there is also evidence of a decrease in the functional connectivity of the dDMN after treatment, and this phenomenon has also been observed after good treatment with antidepressants (Brakowski et al., J Psychiatr Res. 92:147-159 (2017)).

[0137] Equivalent The technology of the present application is not limited to the specific embodiments described in the present application, which are intended as examples of individual aspects of the technology of the present application. As will be apparent to those skilled in the art, many modifications and variations of the technology of the present application may be made without departing from its spirit and scope. In addition to those listed in the present application, functionally equivalent methods and apparatuses within the scope of the technology of the present application will be apparent to those skilled in the art from the above description. Such modifications and variations are intended to be included within the scope of the technology of the present application. It should be understood that the technology of the present application is not limited to a particular method, reagent, compound, composition, or biological system, and these may of course vary. It should also be understood that the terms used herein are for the purpose of describing particular embodiments only and are not intended to be limiting.

[0138] In addition, when the features or aspects of the present disclosure are described in Markush group format, it will be understood by those skilled in the art that the description of the present disclosure also applies to any individual element of the Markush group or a subset of the elements of the Markush group.

[0139] As will be understood by those skilled in the art, for any and all purposes, particularly with respect to providing a written specification, all ranges disclosed herein include any and all of their partial ranges and combinations of partial ranges. It is well described that any of the recited ranges can be divided, for example, at least into halves, thirds, fourths, fifths, tenths, etc., and it will be readily understood that the ranges can be divided in such a way. By way of non-limiting example, each range described herein can be easily divided, for example, into the lower one-third portion, the middle one-third portion, and the upper one-third portion. Those skilled in the art will also understand that all terms such as "(up to)", "at least", "greater than", "less than", etc. include the recited numbers, and that these terms mean ranges that can later be divided into partial ranges as described above. Finally, as will be understood by those skilled in the art, ranges include individual elements. Thus, for example, a group containing 1 to 3 cells refers to a group containing 1, 2, or 3 cells. Similarly, a group containing 1 to 5 cells refers to a group containing 1, 2, 3, 4, or 5 cells.

[0140] All patents, patent applications, patent provisional applications, and publications described or cited herein are incorporated by reference in their entirety, including all of their drawings and tables, to the extent that they do not conflict with the explicit teachings herein.

Claims

1. determining, with a data processing system including one or more processors, a first connectivity value for a subject, said first connectivity value representing a first magnitude of connectivity associated with at least one sub-processing region of the subject's nervous system at a first time point, said first connectivity value corresponding to a first scan at said first time point; determining, by the data processing system, a second connectivity value for the subject, the second connectivity value representing a second magnitude of the connectivity associated with the at least one sub-processing region of the nervous system of the subject at a second time point after the subject has undergone cognitive-emotional training, the second connectivity value being derived from a second scan at the second time point after the cognitive-emotional training; determining, by the data processing system, a difference between the first association value and the second association value; in response to determining that the difference is greater than a threshold, storing, by the data processing system, an association between the subject identification and a first classification value corresponding to a first classification in a data structure including a subject identification identifying the subject; A method comprising:

2. 2. The method of claim 1, wherein the at least one sub-processing region includes a dorsolateral prefrontal cortex (DPFC) and amygdala (AMG), and the first connectivity value, the second connectivity value, and the threshold represent an effective connectivity between the DPFC and the AMG and an effective connectivity threshold between the DPFC and the AMG, respectively.

3. 2. The method of claim 1, wherein the at least one sub-processing region includes an anterior cingulate cortex (dACC) and an amygdala (AMG), and the first connectivity value, the second connectivity value, and the threshold represent an effective connectivity between the dACC and the AMG and an effective connectivity threshold between the dACC and the AMG, respectively.

4. 2. The method of claim 1, wherein the at least one sub-processing region comprises a default mode resting state network (DMN), and the first connectivity value, the second connectivity value, and the threshold represent functional connectivity within the DMN and a functional connectivity threshold associated with the DMN, respectively.

5. The at least one sub-processing region includes a salience resting state network (SAL).

2. The method of claim 1, wherein the first connectivity value, the second connectivity value, and the threshold represent functional connectivity within the SAL and a functional connectivity threshold associated with the SAL, respectively.

6. 2. The method of claim 1, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a right central executive network (RCEN), and the first association value, the second association value, and the threshold value represent an integrity between the LCEN and the RCEN and an integrity threshold between the LCEN and the RCEN, respectively.

7. 2. The method of claim 1, wherein the at least one sub-processing region includes a dorsal default mode resting state network (dDMN) and a ventral default mode resting state network (vDMN), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the dDMN and the vDMN and an integration threshold between the dDMN and the vDMN, respectively.

8. 2. The method of claim 1, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a ventral default mode resting state network (vDMN), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the LCEN and the vDMN and an integration threshold between the LCEN and the vDMN, respectively.

9. 2. The method of claim 1, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a salience resting state network (SAL), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the LCEN and the SAL and an integration threshold between the LCEN and the SAL, respectively.

10. 2. The method of claim 1, wherein the cognitive-affective training comprises at least one of the Emotional Face Memory Task (EFMT), the Wisconsin Card Sorting Task, the Emotional Stroop Task, the Iowa Gambling Task, the Dot Probe Task, a Face Perception Task, or a Delay Discounting Task.

11. 1. A system comprising:

1. A data processing system comprising: determining a first connectivity value for the subject, the first connectivity value representing a first magnitude of connectivity associated with at least one sub-processing region of the subject's nervous system at a first time point and corresponding to a first scan at the first time point; determining a second connectivity value for the subject, the second connectivity value representing a second magnitude of the connectivity associated with the at least one sub-processing region of the nervous system of the subject at a second time point after the subject has received cognitive-emotional training, the second connectivity value being derived from a second scan at the second time point after the cognitive-emotional training; determining a difference between the first association value and the second association value; responsive to determining that the difference is greater than a threshold, storing an association between the subject identification and a classification value corresponding to the classification in a data structure including a subject identification identifying the subject. DATA PROCESSING SYSTEM, INCLUDING ONE OR MORE PROCESSORS CONFIGURED TO A system comprising:

12. 12. The system of claim 11, wherein the at least one sub-processing region includes a dorsolateral prefrontal cortex (DPFC) and amygdala (AMG), and the first connectivity value, the second connectivity value, and the threshold represent an effective connectivity between the DPFC and the AMG and an effective connectivity threshold between the DPFC and the AMG, respectively.

13. The at least one sub-processing region is the anterior cingulate cortex (dACC) and the amygdala (AMG).

12. The system of claim 11, comprising: the first association value, the second association value, and the threshold value, each representing an effective association between the dACC and the AMG and an effective association threshold between the dACC and the AMG, respectively.

14. 12. The system of claim 11, wherein the at least one sub-processing region comprises a default mode resting state network (DMN), and the first connectivity value, the second connectivity value, and the threshold represent functional connectivity within the DMN and a functional connectivity threshold associated with the DMN, respectively.

15. 12. The system of claim 11, wherein the at least one sub-processing region includes a salience resting state network (SAL), and the first connectivity value, the second connectivity value, and the threshold represent functional connectivity within the SAL and a functional connectivity threshold associated with the SAL, respectively.

16. 12. The system of claim 11, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a right central executive network (RCEN), and the first association value, the second association value, and the threshold value represent an integrity between the LCEN and the RCEN and an integrity threshold between the LCEN and the RCEN, respectively.

17. 12. The system of claim 11, wherein the at least one sub-processing region includes a dorsal default mode resting state network (dDMN) and a ventral default mode resting state network (vDMN), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the dDMN and the vDMN and an integration threshold between the dDMN and the vDMN, respectively.

18. 12. The system of claim 11, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a ventral default mode resting state network (vDMN), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the LCEN and the vDMN and an integration threshold between the LCEN and the vDMN, respectively.

19. 12. The system of claim 11, wherein the at least one sub-processing region includes a left central executive network (LCEN) and a salience resting state network (SAL), and the first connectivity value, the second connectivity value, and the threshold represent an integration between the LCEN and the SAL and an integration threshold between the LCEN and the SAL, respectively.

20. 12. The system of claim 11, wherein the cognitive-affective training comprises at least one of the Emotional Face Memory Task (EFMT), the Wisconsin Card Sorting Task, the Emotional Stroop Task, the Iowa Gambling Task, the Dot Probe Task, a Face Perception Task, or a Delay Discounting Task.

Images