Dynamic baseline adjustment for stimulus application system

Abstract

Disclosed herein are systems and methods for applying a stimulus to a target object based on sensed physiological data relating to the target object. A stimulation threshold for when to trigger stimulation of the target object based on a baseline indicating a need for stimulation can be continually adjusted based on the sensed data to more accurately activate stimulation at the appropriate time. To further enhance accuracy, the system can incorporate data from additional sensors, such as an accelerometer, to distinguish between true physiological changes and artifacts caused by patient motion or other bodily activities. This allows for a more robust and reliable calculation of the dynamic baseline.

Description

[0001] DYNAMIC BASELINE ADJUSTMENT FOR STIMULUS APPLICATION SYSTEM

[0002] Priority Claim

[0003] The present application claims priority to and is a continuation of PCT Application No. PCT / IB2024 / 059172 filed September 20, 2024, which is hereby incorporated by reference herein in its entirety.

[0004] Technical Field

[0005] The present disclosure relates to a stimulus application system for applying electrical stimulation to a location in a patient’s body.

[0006] Background

[0007] U.S. Patent Application No. 2017 / 0216607 discloses a neural implant device which comprises a circuit configured to receive an input signal and to generate an electrical signal based on the received input signal. However, this conventional neural implant device has drawbacks that control based on the internal physiological signals is not performed, and thus, stimulus corresponding to the status of the object person may not be applied.

[0008] PCT Publication No. WO 2023 / 1888437 provides a stimulus application system, an implant device, a controller device, a method for controlling a controller device, and a program, capable of applying stimulus corresponding to the status of the object. In particular, the publication discloses an implant device implanted in an object body which is an animal including a human being that can be wirelessly communicably connected to a controller device arranged outside of the object body. The implant device can include circuitry for detection which detects an electrical signal representing a physiological signal at a predetermined part in the object body, circuitry for transmission / reception which transmits detection information representing time variation of the detected electrical signal, and receives, from the controller device, a stimulation instruction representing a stimulus to be applied to the object body, and circuitry for application which applies an electrical stimulus to a predetermined part in the object body, on the basis of the stimulation instruction received by the circuitry for transmission / reception.

[0009] In some applications, systems such as the above have been used for stimulating a nerve. However, it has been found that a baseline of the physiological signal detected from the nerve may vary over time such that determining a stimulation instruction based on a constant threshold may not accurately provide stimulation at the times when it is needed. Furthermore, these baseline variations, or drifts, can be caused by other bodily phenomena, such as patient movement or the activity of nearby organs. These non-target phenomena can introduce artifacts into the detected nerve activity, potentially leading to inaccurate baseline calculations and, consequently, suboptimal stimulation timing.

[0010] Summary

[0011] Disclosed herein are systems and methods for applying a stimulus to a target object based on sensed physiological data relating to the target object. A stimulation threshold for when to trigger stimulation of the target object based on a baseline indicating a need for stimulation can be continually adjusted based on the sensed data to more accurately activate stimulation at the appropriate time.

[0012] In an embodiment, a stimulus application system can include an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction. At least one controller configured to receive detection information pertaining to the physiological signals detected by the sensor unit at each predetermined time interval. For a current time interval, the at least one controller can generate a flat data level comprising averaged processed detection information over a predetermined period of time. The at least one controller can further generate an adjustment threshold based on the flat data level and selectively update a baseline level for the current time interval based on a comparison of the adjustment threshold for the current time interval with the flat data level from a previous time interval. A stimulation instruction can be generated if the processed detection information for the current time interval exceeds the baseline level for the current time interval by a threshold amount.

[0013] In embodiments, the at least one controller is further configured to process the detection information for each predetermined time interval by decoding the detection information.

[0014] In embodiments, the at least one controller is configured to compare the decoded detection information to the baseline level.

[0015] In embodiments, the adjustment threshold is a predetermined percentage higher than the flat data level.

[0016] In embodiments, the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the current time interval if the adjustment threshold for the current time interval is lower than the flat data level for the previous time interval.

[0017] In embodiments, the at least one controller is configured to keep the baseline level for the previous time interval as the baseline level for the current time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is lower than the flat data level for the previous time interval.

[0018] In embodiments, the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the previous time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is not lower than the flat data level for the previous time interval.

[0019] In an embodiment, a stimulus application system can include an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction. At least one controller can be configured to receive detection information pertaining to the physiological signals detected by the sensor unit at each of a plurality of time intervals. A baseline level for a current time interval can be determined by performing a Fast Fourier Transform (FFT) on the detection information to identify a dominant frequency corresponding to an underlying physiological event, defining a search time window having a variable size based on the identified dominant frequency, and identifying the baseline level as a minimum value of the processed detection information within the search time window. A stimulation instruction can be generated if the processed detection information for the current time interval exceeds the baseline level by a threshold amount.

[0020] In embodiments, the at least one controller is configured to generate the stimulation instruction by determining a dynamic peak level for the current time interval and calculating the threshold amount for the current time interval as a function of both the baseline level and the dynamic peak level.

[0021] In embodiments, the dynamic baseline level and the dynamic peak level are determined by performing a Fast Fourier Transform (FFT) on the detection information to identify a dominant frequency, defining a search time window based on the dominant frequency, and identifying the dynamic baseline level as a minimum value and the dynamic peak level as a maximum value within the search time window.

[0022] In embodiments, the at least one controller is further configured to optimize a minimum baseline threshold parameter used to determine the baseline level by iteratively performing the steps of (a) calculating a baseline value for a plurality of physiological cycles using the minimum baseline threshold parameter, (b) in response to determining that the calculated baseline value is unstable within a single physiological cycle, increasing the value of the minimum baseline threshold parameter and (c) in response to determining that the calculated baseline value is the same for two or more consecutive physiological cycles, decreasing the value of the minimum baseline threshold parameter.

[0023] In embodiments, the at least one controller is further configured to optimize the threshold amount by receiving at least one temporal reference marker from a user, the marker corresponding to a patient-sensation event, calculating a timing deviation between a stimulation onset time and a time corresponding to the temporal reference marker for a plurality of threshold amounts and identifying an optimal threshold amount that minimizes the timing deviation.

[0024] In embodiments, the sensor unit further comprises an accelerometer, and wherein the at least one controller is further configured to process data from the accelerometer to distinguish between physiological signals and motion artifacts.

[0025] In embodiments, the at least one controller can be further configured to receive a plurality of input parameters, generate a multivariate predictive model based on the plurality of input parameters, and determine a stimulation threshold for a current time interval based on an output of the predictive model, wherein the stimulation instruction is generated if the processed detection information for the current time interval exceeds the determined stimulation threshold.

[0026] In embodiments, the plurality of input parameters includes at least one parameter selected from the group consisting of: a local maximum of the processed detection information, a local minimum of the processed detection information, a baseline-to-peak time, a peak-to-peak cycle time, a patient-reported sensation, a curve fitting coefficient of the processed detection information, a patient's age, a patient's gender, a liquid intake amount, and a time of day.

[0027] In embodiments, the predictive model comprises a machine learning model. In embodiments, the at least one controller is further configured to receive an input from an external programmer to calibrate the predictive model.

[0028] In embodiments, the predetermined stimulation location and predetermined sensing location is a nerve.

[0029] In embodiments, the nerve is the sacral nerve.

[0030] In an embodiment, a method of applying electrical stimulation to a predetermined stimulation location in a body of a patient can include detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals. At least one controller can process detection information pertaining to the detected physiological signals for each of the plurality of time intervals. For a current time interval, a flat data level can be generated comprising averaged processed detection information over a predetermined period of time . An adjustment threshold can be generated based on the flat data level of the current time interval. A baseline level for the current time interval can be generated based on a comparison of the adjustment threshold for the current time interval with the flat data level from a previous time interval. A stimulation instruction can be generated if the processed detection information for the current time interval exceeds the updated baseline level for the current time interval by a threshold amount. A stimulus circuit unit of the implanted device can apply an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

[0031] In an embodiment, a stimulus application system can include implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction. At least one controller can be configured to receive detection information pertaining to the physiological signals detected by the sensor unit at each predetermined time interval, process the detection information, calculate a rate of change of the processed detection information for a current time interval and generate a stimulation instruction in response to the calculated rate of change exceeding a slope threshold.

[0032] In embodiments, the slope threshold is a predetermined fixed value.

[0033] In embodiments, the predetermined fixed value corresponds to a rate of change indicative of a rapid physiological event.

[0034] In embodiments, the slope threshold is a dynamic threshold calculated based on an average of rate of change values from one or more previous time intervals.

[0035] In an embodiment, a method of applying electrical stimulation to a predetermined stimulation location in a body of a patient can include detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals. At least one controller can process detection information pertaining to the detected physiological signals for each of the plurality of time intervals. The at least one controller can calculate a rate of change of the processed detection information for a current time interval and generate a stimulation instruction if the calculated rate of change exceeds a slope threshold. A stimulus circuit unit of the implanted device can apply an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

[0036] In embodiments, the slope threshold is a predetermined fixed value.

[0037] In embodiments, the slope threshold is a dynamic threshold calculated based on an average of previously calculated rates of change.

[0038] In an embodiment, a method of applying electrical stimulation to a predetermined stimulation location in a body of a patient can include detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals. At least one controller can process detection information pertaining to the detected physiological signals, receive a plurality of input parameters related to the patient or the physiological signals and generate a multivariate predictive model based on the plurality of input parameters. A stimulation threshold can be determined for a current time interval based on an output of the predictive model and a stimulation instruction generated if the processed detection information for the current time interval exceeds the determined stimulation threshold. A stimulus circuit unit of the implanted device can apply an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

[0039] In embodiments, the plurality of input parameters includes at least one of a physiological signal temporal characteristic, a physiological signal morphology parameter, a patient-reported subjective parameter, a patient demographic parameter, or a patient behavioral parameter.

[0040] In a further aspect of the disclosure, the stimulus application system is configured to generate and utilize a multivariate statistical predictive model to determine an optimal stimulation threshold for the timing of the next stimulation. The at least one controller is configured to receive and analyze a plurality of input parameters to build and continuously refine this model. These parameters may include, but are not limited to: (a) direct physiological signal features such as local maximums (peaks) and local minimums (baseline levels); (b) temporal cycle characteristics such as baseline-to-peak time (filling time), peak- to-baseline time (voiding time), and peak-to-peak physiological cycle time; (c) advanced signal morphology characteristics, such as curve fitting coefficients of the nerve activity; (d) patient-reported subjective data, such as sensations of desire or urgency; (e) patientspecific demographic data, such as age and gender; and (f) contextual behavioral data, such as liquid intake and the time of day. By integrating these diverse inputs, the predictive model can anticipate the patient's physiological state and proactively determine the most effective stimulation threshold for the upcoming cycle, ensuring highly personalized and timely therapy. Furthermore, the system may be configured to operate in a closed-loop fashion, wherein it analyzes and evaluates the results of a stimulation event and revises the stimulation threshold for subsequent cycles. This enables the predictive model to continuously learn and adapt to changes in the patient's therapeutic needs over time, further enhancing the personalization and efficacy of the treatment.

[0041] In a further aspect, the disclosure provides systems and methods for analyzing and evaluating the performance and effects of the stimulation therapy. The at least one controller may be configured to analyze the timing of stimulation delivery with respect to physiological cycles, providing visualizations of timing distribution and calculating metrics such as a correct timing ratio. Furthermore, the system can be configured to analyze the immediate, short-term effects of stimulation on physiological parameters, as well as the longer-term, cumulative effects on the stability of those parameters across different therapeutic periods. These analytical tools provide quantitative feedback on the efficacy of the therapy, enabling clinicians to assess and refine treatment strategies.

[0042] The above summary is not intended to describe each illustrated embodiment or every implementation of the subject matter hereof. The figures and the detailed description that follow more particularly exemplify various embodiments.

[0043] Brief Description of Drawings

[0044] Subject matter hereof may be more completely understood in consideration of the following detailed description of various embodiments in connection with the accompanying figures, in which: FIG. 1 is a structural block diagram showing an example of a stimulus application system according to an aspect of the present disclosure.

[0045] FIG. 2 is a functional block diagram showing an example of an implant device according to an aspect of the present disclosure.

[0046] FIG. 3 is a functional block diagram showing an example of a control unit of an implant device according to an aspect of the present disclosure.

[0047] FIG. 4 is a flowchart showing an operation example of a stimulus application system according to an aspect of the present disclosure.

[0048] FIGS. 5A-5B depict decoded nerve activity data overtime according to an aspect of the disclosure.

[0049] FIG. 6 depicts a flowchart of steps in a method of calculating a dynamic baseline for a stimulus application system according to an aspect of the disclosure.

[0050] FIGS. 7A-7B depict examples of the method described with respect to FIG. 6.

[0051] FIG. 8 depicts a flowchart of steps in a method according to an aspect of the disclosure.

[0052] FIG. 9 depicts a flowchart of steps in a method according to an aspect of the disclosure.

[0053] FIG. 10 depicts a flowchart of steps in a method according to an aspect of the disclosure.

[0054] FIG. 11 depicts a flowchart of steps in a method according to an aspect of the disclosure.

[0055] FIG. 12 depicts a flowchart of steps in a method according to an aspect of the disclosure.

[0056] FIG. 13A-13B depict a process for how the slope from multiple datapoints can used to determine the moment to activate stimulation according to an aspect of the disclosure.

[0057] FIG. 14 depicts a flowchart of steps in a closed-loop control process for generating and revising a stimulation threshold according to an aspect of the disclosure.

[0058] FIG. 15 depicts a flowchart of steps in a closed-loop control process for generating and revising a stimulation threshold according to an aspect of the disclosure.

[0059] FIG. 16A depicts an exemplary graphical representation of stimulation timing distribution across several physiological cycles, according to an aspect of the disclosure.

[0060] FIG. 16B depicts a method for quantifying the proportion of correctly timed stimulations within each cycle, according to an aspect of the disclosure. FIG. 16C depicts a method for analyzing the rate of stimulation, according to an aspect of the disclosure.

[0061] FIG. 17A depicts a method for analyzing the immediate effect of stimulation on a measured physiological parameter, according to an aspect of the disclosure.

[0062] FIG. 17B depicts a method for analyzing the cumulative effect of stimulation over a longer period, according to an aspect of the disclosure.

[0063] While various embodiments are amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail. It should be understood, however, that the intention is not to limit the disclosed inventions to the particular embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the subject matter described herein.

[0064] Detailed Description

[0065] The following detailed description should be read with reference to the drawings in which similar elements in different drawings are numbered the same. The drawings, which are not necessarily to scale, depict illustrative embodiments and are not intended to limit the scope of the invention.

[0066] As exemplified in FIG. 1, a stimulus application system 1 according to an aspect of the present disclosure includes an implant device 10 implanted in an object body which is an animal including a human being and a programmer 20 and charging device 30 which can be arranged outside of the object body. Stimulus application system 1 may further communicate with a cloud platform 40.

[0067] Implant device 10, depicted in FIG. 2, can include a transmission / reception unit 11, a power supply unit 12, a stimulus circuit unit 13, a sensor unit 14, and a control unit 15.

[0068] The transmission / reception unit 11 of the implant device 10 can transmit data to the programmer 20 arranged outside of the object body, in accordance with instructions input from the control unit 15. Further, the transmission / reception unit 11 can receive data from the programmer 20 and output the data to the control unit 15. In various embodiments, a widely known data transmission / reception method can be adopted, such as NFC, Wi-Fi, Bluetooth (registered trademark), RFID wireless communication standard, and the like. According to an embodiment, the transmission / reception unit 11 can receive wireless power transfer from the recharging device 30 and output the received power to the power supply unit 12.

[0069] The power supply unit 12 can be provided with a battery B and supply power to each unit of the implant device 10. Battery of power supply unit 12 can be wirelessly recharged, as will be described in more detail below.

[0070] The stimulus circuit unit 13 is under control of the control unit 15 and can apply a stimulus to the object body through electrodes arranged at predetermined parts (hereinbelow, referred to as stimulation parts) in the object body. Herein, the stimulation part where the electrode is arranged is a part to which stimulus to a nerve etc., can be applied, as used in Spinal Cord Stimulation, Sacral Neuro Modulation, Vagus Nerve Stimulation, Deep Brain Stimulation, and the like. The stimulation parts can be selected in accordance with the type of stimulus to be applied to the object body. For the arrangement of the electrodes of the stimulus circuit unit 13, arrangements widely known as the arrangements used for the above-mentioned various stimulation methods can be adopted. Therefore, detailed explanation therefor is omitted here. Further, the stimulus may be a periodical electric signal, a single pulse signal, etc., and an amplitude, a frequency, a duration, a pulse width, and the like of the signal can be controlled by the control unit 15.

[0071] The sensor unit 14 can detect electric signals representing physiological signals of the object body (for example, signals representing the physiological signals at the detection parts of the object body, by the magnitude of potential thereof), through the electrodes arranged at predetermined parts (hereinbelow, referred to as detection parts) in the object body. In an embodiment, the sensor unit 14 may further comprise an accelerometer configured to detect patient movement. The data from the accelerometer can be used by the control unit 15 to differentiate between changes in the detected electrical signal due to nerve activity and those due to motion artifacts. In other embodiments, the sensor unit 14 could also include sensors to monitor the activity of nearby organs, such as pressure sensors, to further refine the baseline calculation by accounting for physiological signals not directly related to the targeted nerve activity. The physiological signal can include one or more of membrane potential, Nerve action potential, Organ pressure, Tissue impedance, temperature, and other signals acting as biomarkers, in the object body, and the physiological signal can be selected in accordance with a rule predetermined depending on the type of stimulus to be applied. Further, the stimulation part and the detection part may be different parts, adjacent (comparatively close) parts, or the same part.

[0072] The control unit 15 can include a program controller device such as a CPU, and a storage device such as a memory. The control unit 15 can process detection information representing the electrical signal detected by the sensor unit 14 (for example, in case that the electrical signal represents the physiological signal at the detection part in the object body by the magnitude of its potential, the detection information is information representing the magnitude of the potential). The control unit 15 can analyze the detection information representing the detected electrical signal every time that the electrical signal is detected, or can store the information representing the electrical signals detected for a plurality of times in a memory, etc., and then process the information representing the detection information stored in the memory, at a predetermined time. Namely, for example, the detection information may include information representing an electrical signal as a result of one-time detection, or may include information representing a plurality of electrical signals as a result of a plurality of times of detection (representing detection information representing the time variation of the detected electrical signal).

[0073] Further, the control unit 15 can provide instructions (stimulation instruction) to the stimulation circuit 13. The control unit 15 can determine parameters such as the frequency and the intensity (amplitude) of the electrical signal to be applied to the object body as a stimulus, the pulse width (in case that the electrical signal is a pulse signal), and in addition, the time when the stimulus is to be applied, the duration of the stimulus, and the like. Then, the control unit 15 can control the stimulus circuit unit 13 so that the stimulus of the electrical signal defined by the determined parameters is applied.

[0074] The programmer 20 can be arranged at a position outside of the object body and wirelessly communicable with the implant device 10. In some embodiments, the programmer 20 can provide initial or updated programming of parameters for the implant 10 to follow and the processor of the implant 20 then carries out detection, analysis and stimulation functions independently. For example, a physician may program the implant 10 with the programmer 20 during an office visit, with the implant 10 then operating without need for further external instruction. The programmer 20 can also receive historical therapy data stored by the implant. Such data can include, for example, sensor data detected by sensor unit 14, stimulation data provided by stimulation circuit 13, etc. The programmer 20 can also transmit data to and receive data from artificial intelligence system 40, as will be described in more detail below.

[0075] Recharging device 30 can be a wearable recharger configured to transmit power to recharge the battery in the power supply unit 12 of the implant. Wearable recharger 30 can be selectively worn by the user when the implant needs to be charged and then removed from the user as desired. In embodiments, recharging device 30 can provide an indication of a charging status and / or battery level of the power supply unit 12 of the implant. Wearable recharger 30 can itself be recharged by connection to a power source. Further details regarding wireless power technology that can be used with aspects of the disclosure can be found in U.S. Patent Publication No. 2023 / 0344273, which is hereby incorporated by reference herein in its entirety.

[0076] Cloud platform 40 can employ, for example, any available cloud computing service. Programmer 20 can communicate information to and receive information from cloud platform 40 over WiFi communications or other known communication modalities. Cloud platform 40 can include data storage that stores data relating to system 1. In embodiments, programmer 20 can transmit historical data received from implant 10 for a plurality of patients to cloud platform 40. Cloud platform may further include an Artificial Intelligence (Al) system that can analyze the historical data and provide suggestion for improvements to therapy provided by implant 10. For example, cloud platform 40 may store a separate file for each patient using an implant 10. The Al system may analyze the data across all patients, a group of patients, etc. and provide suggestions for modifications to improve therapy for one or more individual patients based on the analysis.

[0077] Referring now to F igure 3 , an operation example of the control unit 15 of the implant device 10 will be explained. According to an embodiment, the control unit 15 executes the programming received from the programmer 20 and stored in a storage device. To carry out the programming, the control unit 15 functionally comprises a reception unit 151, a stimulus determination unit 152, and an instruction transmission unit 153.

[0078] The reception unit 151 can receive detection information detected by the sensor unit 14, the detection information being a detection result of the electrical signal detected by the sensor unit 14 as a physiological signal of the object body.

[0079] On the basis of the detection information received by the reception unit 151, the stimulus determination unit 152 can determine the details of the stimulus to be applied to the object body by the stimulation circuit 13. As mentioned above, the detection information represents an electrical signal at a predetermined detection part in the object body, detected by the sensor unit 14. The stimulus determination unit 152 can use this detection information to acquire information regarding the time variation of the physiological signal at the predetermined part in the object body. For example, when the detection information includes information representing one electrical signal, as a result of detection for one time, the stimulus determination unit 152 can accumulate and store the detection information for a plurality of times to acquire information representing the time variation of the electrical signal (physiological signal at a predetermined part in the object body).

[0080] For example, the stimulus determination unit 152 determines the type of stimulus with reference to stimulus setting information in which a plurality of mutually different stimulus applying conditions are associated with information representing the details of the stimulus corresponding to each stimulus applying condition. The stimulus setting information can be set in advance and transmitted to implant 10 by programmer 20 and stored in the storage device. The stimulus determination unit 152 can output the information representing the details of the stimulus to the instruction transmission unit 153.

[0081] The instruction transmission unit 153 can transmit the stimulation instruction representing the details of the stimulus determined by the stimulus determination unit 152, to be carried out by the stimulation circuit 13. The stimulus setting information used here may be defined on the basis of the detection result of the electrical signal representing the physiological signal, which is acquired by the sensor unit 14.

[0082] Figure 4 depicts a flowchart showing an operation example of a stimulus application system according to an aspect of the present disclosure. In the following example, the implant device 10 is implanted in the object body, i.e., the body of a human being, and electrodes for providing stimuli are arranged at the stimulation parts and the detection parts used for Sacral Neuro Modulation. The sensor unit 14 ofthe implant device 10 senses nerve activity of the sacral nerve. In other examples, stimulation application system can be employed in a similar manner to stimulate other nerves or areas of the body.

[0083] In this process, first, the sensor unit 14 of the implant device 10 detects the electrical signals representing the physiological signals in the human body by the electrodes arranged in the detection parts, and generates detection information representing the detected electrical signals(Sl 1). The sensor unit 14 transmits the generated detection information to the control unit 15 of the implant device 10, at a predetermined time (for example, every time that the detection is performed) (SI 2). Then, the implant device 10 examines whether or not an instruction is received from the control unit 14 within a predetermined time (S 13). If no instruction is received (S13: No), the process returns to Step Si l, and is continued.

[0084] In Step S12, the control unit 15 receives and stores the detection information transmitted by the sensor unit 14 (S21). The detection information stored in Step S21 is processed to decode the information to obtain data of interest (S22), such as a frequency range of interest and a reference frequency range (described in more detail below), after a predetermined number of data points are stored. In one embodiment, the information is decoded after digital data points are stored. Further details on one example of this decoding process can be found in copending PCT Application No. PCT / IB24 / 059174 entitled SYSTEMS AND METHODS FOR STIMULATION APPLICATION ACCOUNTING FOR LEAD MIGRATION, filed on the same day as the present application, which is hereby incorporated by reference in its entirety.

[0085] Once the decoded nerve activity data of interest is obtained at Step S22, the baseline level is obtained from decoded data (S23). The baseline is set as the minimum nerve activity intensity of each data gathering cycle. The stimulation threshold can then be obtained from the baseline data (S24). It is then determined if the nerve activity data is above the stimulation threshold (S25), and if so details of a stimulation to be applied are determined (S26). If the data is not above the baseline by a threshold amount at Step S25, the system reverts to Step S21. Further details regarding calculation of and comparison with the baseline are detailed below.

[0086] The controller unit 15 transmits an instruction representing the details of the stimulus determined in Step S26 to the stimulation circuit 13 (S27). If the stimulation circuit 13 receives the instruction from the control unit 15 in Step S13 (S13: Yes), the stimulation circuit 13 applies parameters, such as a frequency, an amplitude, etc., of the electrical signal as a stimulus to be applied to the human body in which the implant device 10 is implanted, and controls a current to be applied to the stimulation part through the electrode so that the stimulus of the electrical signal with the determined parameters is to be applied (S 14). The stimulus application system 1 according to the present aspect repeats the operations from Step SI 1 to Step S14, and the operations from Step S21 to Step S27. Accordingly, according to an example of the present aspect, the stimulus to be applied is varied in accordance with the status of the object human body, etc., and thus, a stimulus suitable for the status of the object can be applied.

[0087] To address baseline drifts caused by non-target phenomena, the control unit 15 can be configured to process data from multiple sensors. For instance, when an accelerometer is included in the sensor unit 14, the control unit 15 can receive both nerve activity data and motion data. The control unit 15 can then utilize an adaptive filtering algorithm. During periods of significant motion detected by the accelerometer, the control unit 15 can temporarily reduce the weight given to the nerve activity signal when calculating the baseline, or it can apply a filter to remove the motion-induced artifacts from the nerve signal before the baseline is calculated. This ensures that the baseline is not erroneously skewed by patient movement. Similarly, if a sensor for organ activity is included, its data can be used to subtract the influence of that organ's activity from the nerve signal, providing a cleaner signal for baseline determination

[0088] In an exemplary application, this stimulus application system can be used for suppressing the symptoms of overactive bladder (OAB) by delivering stimulation timed to the patient's physiological (e.g., urination) cycle. The accurate determination of this cycle's duration and its key phases is crucial for optimizing the system's therapeutic parameters, including the dynamic baseline and the stimulation threshold ratio. The duration of the physiological cycle can be determined through several methods, which may be used independently or in combination:

[0089] 1. Automated Signal Analysis: The cycle can be defined by analyzing the detected physiological signal, such as nerve activity. In this method, the start of the cycle, referred to as the baseline or onset, is identified as a local minimum in the signal's amplitude. The end of the cycle can be correlated with a local maximum, which may represent a voiding event. The duration is the time elapsed between these two algorithmically detected points.

[0090] 2. Patient-Reported Events: The cycle duration can be defined by markers actively provided by the patient. For example, the patient can use an external device (such as the programmer 20) to log the time of a "first desire to urinate" and the time of a subsequent urination or incontinence event. The system then calculates the duration between these manually entered markers.

[0091] 3. Hybrid Approach: A combination of automated signal analysis and patient- reported events can be used to create the most accurate model. This method correlates the objective physiological data with the patient's subjective sensations. For instance, the system can use the patient's report of a "desire to urinate" as a key marker to help the control unit's algorithm more accurately identify the corresponding baseline or onset in the nerve activity signal, thereby refining the cycle's start time and improving the overall accuracy of the therapy timing.

[0092] Referring now to Figure 5 A, one example of decoded nerve activity data 50 over time is depicted. As noted above, to determine when stimulation should be delivered a baseline 52 of the nerve activity data is set and therapy is delivered upon the nerve activity 50 exceeding the baseline 52 by a stimulation threshold amount 54. The stimulation threshold 54, may be, for example 10% above the baseline. In embodiments, stimulation threshold can be predetermined by a physician and programmed into implant device 10 using programmer 20. Alternatively and / or additionally, the stimulation threshold can also be automatically determined by programmer 20 based on the analyzed patient data and programmed into implant device 20.

[0093] Setting the stimulation threshold too high would risk giving stimulation too late to manage the patient’s symptoms and setting the stimulation threshold too low would make the stimulation similar to constant stimulation that does not provide stimulation specifically when needed. To balance these concerns, for example, nerve activity data can be recorded for multiple physiological cycles and reviewed by the physician to determine an appropriate percentage for the threshold for activating stimulation to best manage a given patient’s symptoms.

[0094] The example depicted in Figure 5A sets the baseline 52 as a constant value based on the minimum nerve activity during the initial physiological cycle. However, it has been found that the baseline nerve activity for a given physiological cycle tends to fluctuate over time. For example, in the nerve activity 52 depicted in Figure 5A, the minimum nerve activity in each physiological cycle is continually going lower. As such, therapy using a constant baseline set during the initial cycle is not provided at the optimal time during a given subsequent physiological cycle. Methods and systems disclosed herein therefore provide a dynamic baseline 62 as depicted in Figure 5B that enables a dynamic stimulation threshold 64 (e.g., 10% above the baseline level for a given physiological cycle). This baseline 62 can be recalculated for each physiological cycle resulting in the ability to provide therapy at the optimal time within a given cycle.

[0095] Figure 6 depicts a flowchart of steps in a method of calculating a dynamic baseline for a stimulus application system according to an aspect of the disclosure. Initially the decoded nerve activity is obtained (S61). Flat data of averaged decoded nerve activity data obtained in step S61 is then generated (S62). The same decoded nerve activity data is maintained for aperiod of time (e .g . , 10 seconds) to minimize data fluctuations and improve detection accuracy. After 10 seconds, the flat data is updated with the next decoded nerve activity result, which is maintained for the following 10 seconds, and so on.

[0096] Next, an adjustment threshold to trigger a baseline data adjustment is generated (S63). The adjustment threshold can be, e.g., 20% higher in amplitude than the flat data generated in step S62. This adjustment threshold will be used to define the baseline data for each physiological cycle. In embodiments, this threshold can similarly be predetermined by a physician and programmed into the implant device 10 with the programmer 20 based on an analysis of recorded nerve activity data for a given patient. The adjustment threshold can be alternatively or additionally be automatically determined by programmer 20 based on the analyzed data and programmed into implant device 10.

[0097] The baseline data is then generated. To do so, first it is determined if the new adjustment threshold data determined at step S63 from the current cycle is lower than the previous flat data from the prior cycle (S64). If the determination at step S64 is that the new adjustment threshold data is lower than the previous flat data, then the new flat data is set as the new baseline data (S65). If the determination at step S64 is that the new adjustment threshold data is not lower than the previous flat data, it is next checked if the previous baseline data from the prior cycle is lower than the previous flat rate data (S66). If the previous baseline data is lower than the previous flat rate data in step S66, then the previous baseline data is kept as the new baseline data (S67). If at step S66, the previous baseline data is not lower than the previous flat rate data, then the previous flat data is set as the new baseline data (S68). These steps will then repeat each cycle to provide a dynamically adjusted baseline. Repeating the process will resulting in the dynamically changing baseline 62 depicted in Figure 5B. The corresponding stimulation threshold 64 then automatically adjusts along with the baseline 62 as described above. The changes in the baseline can be due to other bodily phenomena, such as patient movement or organ activity, hence, the drifts in baseline can be addressed using another sensor such as an accelerometer, allowing to perform the calculation without using the nerve activity.

[0098] Examples of the method described with respect to Figure 6 can be seen with reference to Figures 7A-7B. There are essentially three results that can be obtained by following this method:

[0099] A) If the new threshold data in a cycle is lower than the flat data from the previous cycle, then the new flat data in the current cycle is set as the new baseline data for that cycle,

[0100] B) If 1) the new threshold data in a cycle is not lower than the flat data from the previous cycle and 2) the baseline data from the previous cycle is lower than the flat data from the previous cycle, then the baseline data from the previous cycle is kept as the new baseline data in the current cycle, and

[0101] C) If 1) the new threshold data in a cycle is not lower than the flat data from the previous cycle and 2) the baseline data from the previous cycle is not lower than the flat data from the previous cycle, then the flat data from the previous cycle is set as the new baseline data for the current cycle.

[0102] One example of result (A) can be seen in Figure 7A, which depicts decoded nerve data 70, threshold data 72, flat data 74 and baseline data 76 over time. In the third physiological cycle depicted between 310 seconds and 320 seconds, the threshold data 723 falls below the flat data 742 from the previous cycle between 300 seconds and 310 seconds. As such, the baseline data 763 in the current third cycle is set equal to the flat data 743 in the current cycle. The same result occurs in the fourth depicted time period between 320 and 330 seconds, where the threshold data 724 is below the is below the previous flat data 743, so the baseline data 764 for the current cycle is set equal to the flat data 744 for that cycle.

[0103] Examples of result (B) can be seen in Figure 7B, which similarly depicts decoded nerve data 80, threshold data 82, flat data 84 and baseline data 86 over time. Across each time period in this graph, the threshold data 82 is not lower than the previous flat data 84. In addition, in the third time period between 70 seconds and 80 seconds, the threshold data 823 is not lower than the previous flat data 842 and the baseline data 862 is lower than the previous flat data 842. As such, the baseline data from the previous time periods 862 remains the same baseline data for the current time period 863.

[0104] An example of result (C) can also be seen in Figure 7B. At the time of the first transition in the data just before 61 seconds, the threshold data 821 is not lower than the previous flat data 841 and the baseline data 861 is not lower than the previous flat data 821. The baseline data at 862 is then set equal to the previous flat data 841.

[0105] Systems and methods disclosed herein can further include a process for optimizing the parameters used for the minimum baseline threshold estimation, as depicted in Figure 8. The purpose of this method is to calibrate the sensitivity of the baseline detection algorithm to the unique physiological signal characteristics of the patient. This ensures the system can reliably identify a distinct baseline for each physiological cycle, even when the baseline fluctuates overtime, without being overly sensitive to intra-cycle noise.

[0106] The process begins with obtaining decoded nerve activity data (S 81) and calculating a baseline value using an initial or default minimum baseline threshold (S82). The system then evaluates the stability and responsiveness of this calculation across one or more physiological cycles (S83).

[0107] The algorithm operates based on two conditions. First, it determines if the calculated baseline changes for one physiological cycle (S84). Such instability suggests the threshold is too low, making the algorithm overly sensitive to minor signal fluctuations. If this occurs, the value of the minimum baseline threshold is increased to reduce sensitivity (S85), and the process is repeated.

[0108] Conversely, the system determines if two or more consecutive physiological cycles yield the exact same baseline value (S86). This condition suggests the threshold is too high, making the algorithm insensitive to genuine, subtle changes in the patient's physiological state between cycles. If this occurs, the value of the minimum baseline threshold is decreased to increase sensitivity (S87).

[0109] This iterative adjustment continues until a threshold value is established that provides a stable baseline within each individual cycle while remaining responsive to changes between cycles. This final, optimized value is then saved as the minimum baseline threshold for subsequent therapeutic operation (S88).

[0110] Systems and methods disclosed herein can further optimize the stimulation threshold ratio to ensure stimulation is delivered at the most therapeutically effective time for the patient. Figure 9 depicts one example of such an optimization process. This method leverages patient-reported events, such as the sensation of a "desire to urinate," as a temporal reference marker.

[0111] First, decoded nerve activity data is obtained (S91) and the baseline is determined (S92). The system then receives patient-reported markers indicating the timing of each "desire to urinate" event. The median nerve activity amplitude at the time of these events is calculated and used as an ideal stimulation reference timing (S93). A timing deviation, or "loss," is defined as the difference between this reference timing and the actual onset of stimulation that would occur at a given threshold ratio.

[0112] To find the optimal ratio, the controller can iteratively test a plurality of possible stimulation threshold ratios (S94) and compute the corresponding loss for each one (S95). The results can be analyzed (S96) to identify the stimulation threshold ratio that yields the lowest loss, thereby indicating the proper timing for stimulation. This process allows the system to be calibrated to the individual patient's perception, and can be configured to deliver stimulation at, or even slightly before, the onset of desire in an anticipatory manner.

[0113] To further improve the accuracy of the baseline detection, the system can employ an adaptive methodology based on the signal's own characteristics, as depicted in Figure 10. After collecting the raw signal (SI 01), the controller performs a Fast Fourier Transform (FFT) on the data to identify the dominant frequency of the underlying physiological event (S102). This dominant frequency corresponds to the primary cadence of the physiological / biological cycle (e.g., the urination cycle).

[0114] Based on this dominant frequency, the controller calculates an adaptive search time window with a variable size (S 103), making the search for the baseline more robust to signal variations. Within this adaptive window, the controller identifies candidate baseline points by locating the minimum signal value (S 104). To ensure reliability, this minimum value can be validated against the values in adjacent windows (SI 05). If validated, the minimum value is set as the baseline for that cycle (SI 06).

[0115] Similarly, as depicted in Figure 11, the system can employ a parallel adaptive peak methodology. After obtaining the decoded data (Si l l) and using FFT to determine the dominant frequency (SI 12) and the corresponding adaptive search window (SI 13), the controller searches for the maximum signal value within the window to identify the peak of the physiological event (SI 14). This peak value is likewise validated (SI 15) before being saved as the peak for the current cycle (SI 16). Together, these adaptive baseline and peak detection methods allow the system to accurately track the full dynamic range of the physiological signal, even as it changes overtime.

[0116] In the embodiments described above, the stimulation threshold is primarily described as being a fixed threshold amount (e.g., 10%) above the baseline for a given physiological cycle at which stimulation will be delivered that can be predetermined by a physician. In a further embodiment, the system can move beyond a predetermined stimulation threshold amount and instead calculate a fully adaptive threshold for each physiological cycle, as depicted in Figure 12. This method leverages the dynamic baseline and peak detection capabilities described previously.

[0117] After the decoded nerve activity is obtained (S 121), the system detects the dynamic baseline and the dynamic peak for the current cycle (S122), for instance by using the FFT- based adaptive window methods. The stimulation threshold for the current cycle is then dynamically calculated as a function of both the baseline and the peak (S123). For example, the threshold can be set at a specific ratio of the signal's dynamic range (i.e., the difference between the peak and the baseline) above the current baseline level.

[0118] This ratio can be preset or can itself be adaptive, adjusting in real-time based on historical data from previous baselines and peaks. After calculation, the threshold value is set and used for the current cycle (S 124). This adaptive thresholding ensures that stimulation is triggered at a consistent relative point within the signal's activity range, making the therapy more robust and better controlled, especially when the overall amplitude of the nerve activity fluctuates.

[0119] In a further embodiment, the determination of when to generate a stimulation instruction can be based on the rate of change of the decoded nerve activity, rather than or in addition to its absolute amplitude relative to a baseline. This approach can be particularly effective for detecting rapid onsets of physiological events, such as a sudden change in bladder activity.

[0120] In such an embodiment, after the decoded nerve activity is obtained, the at least one controller can be configured to calculate the rate of change, or slope, of the decoded nerve activity's amplitude over a series of time points. This slope value represents the velocity of the change in nerve activity. The controller can then use this slope information to generate a stimulation instruction based on one or more criteria.

[0121] An example of one such implementation is shown in Figure 13A. The controller performs a slope calculation in a given time, for example every 1, 2 or 5 seconds, and calculates the running average each 10 calculations, for example. A stimulation instruction can then be generated when a newly calculated slope value for the current time interval exceeds this running average by a predetermined amount or percentage. This indicates that the nerve activity is increasing at an accelerated rate compared to its recent trend.

[0122] In an alternative implementation shown in Figure 13B, the controller can compare the calculated slope to a fixed, predetermined slope threshold. A stimulation instruction can be generated if the calculated slope value surpasses this threshold. This threshold can be defined, for example, as a slope corresponding to a specific angle, such as 45 degrees on a normalized plot of amplitude versus time, which signifies a rapid increase in activity. This method provides a direct trigger based on a significant, predefined rate of change in the physiological signal.

[0123] These slope-based methods can be used independently or in conjunction with the dynamic baseline and peak amplitude methods described previously to create a more robust and responsive stimulation system.

[0124] In some embodiments, auxiliary sensors are configured to capture multivariate information related to contextual, physiological and behavioral state of the patient, and may include but are not limited to at least one of an accelerometer, an Inertial Measurement Unit (IMU), a gyroscope, a magnetometer, or a pressure sensor. Beyond real-time adaptive adjustments, the stimulus application system can be configured to employ a predictive paradigm to further optimize therapy. In this embodiment, as illustrated in the block diagram of Figure 14, the at least one controller is configured to generate and utilize a multivariate statistical predictive model. This model leverages a comprehensive analysis of diverse data inputs 1410 to forecast the patient's physiological needs and proactively determine the stimulation threshold for the next start timing of stimulation.

[0125] The input parameters 141 for this predictive model are derived from multiple sources, providing a holistic view of the patient's state. As shown in Figure 14, these inputs 1410 can be categorized as follows:

[0126] Direct Physiological Signal Features: The model directly analyzes key amplitude markers from the decoded nerve activity data, including:

[0127] Peaks (local maximums): The highest amplitude points within a physiological cycle.

[0128] Baseline level (local minimums): The lowest amplitude points, which are foundational to the dynamic baseline methods described herein.

[0129] Temporal Signal and Cycle Characteristics: The model analyzes the timing and duration of different phases of the physiological cycle, such as the urination cycle. These include:

[0130] Baseline-to-peak time (filling time): The duration from a cycle's minimum point to its subsequent maximum point, often corresponding to a bladder filling phase.

[0131] Peak-to-baseline time (voiding time): The duration from a cycle's maximum point to the next minimum, potentially corresponding to a voiding event.

[0132] Peak-to-peak cycle: The total time elapsed between two consecutive peaks.

[0133] Minimum-to-minimum cycle: The total time elapsed between two consecutive minimums, representing the full cycle duration.

[0134] Advanced Signal Morphology Parameters: To capture more subtle changes in nerve activity, the model can analyze the shape of the signal's curve. This includes:

[0135] Curve fitting coefficient for nerve activity: The controller may fit the nerve activity data to a mathematical function (e.g., polynomial, exponential) and use the resulting coefficients as inputs. A change in these coefficients can indicate a change in the rate or pattern of nerve firing.

[0136] Patient-Reported Subjective Data: The system is configured to receive direct feedback from the patient via the programmer 20. This is a critical input for calibrating the model to the patient's actual experience and includes:

[0137] Patient-reported sensations: Timestamps or severity ratings for events like a first desire to urinate, urgency, or an incontinence event.

[0138] Patient-Specific Demographic Data: This includes static data about the patient that can be used to initialize or weight the model, such as:

[0139] Patient age.

[0140] Patient gender.

[0141] Contextual and Behavioral Data: This includes real-time or daily inputs that provide context for the physiological signals:

[0142] Time of the day: To account for circadian rhythms.

[0143] Liquid intake: Entered by the patient, as it directly impacts bladder function.

[0144] The decoded nerve activity data 1405, along with these multivariate inputs 1410, are fed into the multivariate predictive statistical model 1420. The model, which can be a machine learning algorithm such as a regression model or neural network, is continuously updated. The diagram also illustrates a feedback loop where the stimulation can be controlled or adjusted by a physician via the programmer, allowing for human-in-the-loop calibration of the model's output. The ultimate output is a precisely calculated stimulation threshold 1430 optimized for the next physiological cycle, enabling a proactive and highly personalized therapeutic strategy in which the programmer can control stimulation 1440 according to the precisely calculated threshold 1430.

[0145] Figure 15 provides a detailed flowchart of an adaptive control process that utilizes the multivariate inputs 1510 to not only generate an initial threshold but also to continuously refine it through a closed-loop feedback mechanism. This process transforms the system into a learning device that adapts to the patient over time.

[0146] The process begins with the collection of a comprehensive set of Inputs 1510, which are categorized for clarity, previously defined above and described in flowchart from Figure 14:

[0147] Physiological: These include direct measurements and temporal characteristics derived from the decoded nerve activity, such as peaks, baseline levels, baseline-to-peak time, peak-to-peak cycle, and minimum-to-minimum cycle, as previously described.

[0148] Behavioral: This category includes patient-provided information such as liquid intake and patient-reported sensations (e.g., desire, urgency).

[0149] Advanced signal parameters: This includes morphological data such as the curve fitting coefficients for the nerve activity signal.

[0150] Contextual: This includes environmental or situational data such as the time of the day and activity change (e.g., transitioning from sitting to standing), the latter of which can be provided by an accelerometer.

[0151] Demographic: This includes static patient data such as age and gender.

[0152] These diverse inputs are then fed into an Analysis at step 1520. This step represents the core of the multivariate predictive model, where algorithms process the inputs to understand their complex interdependencies and their relationship to the patient's physiological state.

[0153] Based on the output of the analysis, the controller proceeds to generate a stimulation threshold at step 1530. This is the initial, proactively determined threshold that the system predicts will be most effective for the upcoming physiological cycle.

[0154] The newly generated threshold is then used in the control stimulation step 1540. In this operational phase, the system actively monitors the real-time physiological signal and triggers stimulation when the signal exceeds the generated threshold at step 1540.

[0155] A key aspect of this embodiment is the feedback loop that enables continuous learning and optimization. After a stimulation event is controlled at step 1540, the process enters an Analysis and Evaluation of stimulation result step 1550. In this step, the system assesses the effectiveness of the applied stimulation. This evaluation can involve correlating the timing of the stimulation with subsequent physiological data (e.g., did the nerve activity return to baseline as expected?) or with subsequent patient-reported outcomes (e.g., was an urgency or incontinence event avoided?).

[0156] Based on this evaluation, the controller proceeds to revise the stimulation threshold at step 1560. If the evaluation indicates a suboptimal outcome (e.g., stimulation was too late, or the patient still reported urgency), the model adjusts the threshold accordingly for the next cycle. This revision may be a simple incremental adjustment or a more complex update to the internal weightings of the predictive model.

[0157] This revised threshold is then fed back to the 'Control stimulation' step 1540 for use in the next therapeutic intervention, completing the closed-loop process. This iterative cycle of generating, controlling, evaluating, and revising allows the system to autonomously learn and adapt, creating a continuously self-optimizing therapeutic system that is tailored to the individual patient's evolving needs.

[0158] In addition to generating and refining stimulation instructions, the stimulus application system 1 can be configured to provide detailed analysis and reporting on the performance and effects of the therapy. These analytics can be presented to a clinician or patient on the programmer 20 or a connected device, allowing for a comprehensive assessment of the treatment's efficacy. These analytical methods are applicable to all stimulation control methods described herein.

[0159] In various embodiments, the at least one controller 15 is configured to evaluate the timing of stimulation delivery with respect to recurring physiological cycles. A physiological cycle may be, for example, a urination cycle, a respiratory cycle, or a cardiac cycle, as determined from sensor data. The analysis provides quantitative and qualitative feedback on whether the stimulation is being applied within a desired therapeutic window of said cycle.

[0160] FIG. 16A depicts an exemplary graphical representation of stimulation timing distribution across several physiological cycles, according to one embodiment. The method involves segmenting each physiological cycle into a plurality of temporal portions. In this non-limiting example, each cycle on the horizontal axis 1610 is divided into four quartiles (0-25%, 25-50%, 50-75%, 75-100%). A therapeutic window of interest is defined. For instance, the desired timing for stimulation may be the latter half of the cycle (50-100%), indicated by green-shaded portions 1620. Stimulations occurring in the first half of the cycle may be considered early, indicated by red-shaded portions 1630. For each cycle, the number of stimulation events falling within each temporal portion is quantified and displayed, for example, as a stacked bar chart on the vertical axis 1640. This visualization allows a user to rapidly inspect the distribution of stimulation timing and identify any systematic deviations from the intended therapeutic window.

[0161] FIG. 16B depicts a method for quantifying the proportion of correctly timed stimulations within each cycle. A "correct timing ratio" 1650 is calculated for each physiological cycle. This ratio is the number of stimulations delivered within the predefined therapeutic window (e.g., after the 50% mark of the cycle) divided by the total number of stimulations in that cycle. The calculated ratio for each cycle is displayed, for example, as a bar chart. A threshold 1660 (e.g., a dashed line at a ratio of 0.5) can be displayed to indicate a minimum acceptable performance level. Cycles with bars 1670 that meet or exceed this threshold are considered to have acceptable stimulation timing.

[0162] FIG. 16C depicts a method for analyzing the rate of stimulation. To account for the variability in the duration of physiological cycles, the number of stimulations is normalized by time. The stimulation rate, such as the number of stimulations per minute, is calculated for each cycle 1680. This normalized rate is plotted for each cycle, providing a measure of stimulation intensity that is independent of cycle length. This allows for the evaluation of whether the current system settings trigger an appropriate frequency of stimulation in response to the patient's physiological state.

[0163] In further embodiments, the system is configured to analyze the effects of the stimulation on physiological, behavioral, or contextual data collected from one or more sensors.

[0164] FIG. 17A depicts a method for analyzing the immediate effect of stimulation on a measured parameter, such as nerve activity. The method involves measuring a physiological parameter at specific time points relative to a stimulation event. Measurements are taken immediately before (e.g., point 'O' at 1710) and immediately after (e.g., point 'I' at 1720) the stimulation trigger. Additional measurements 1730 may be taken at other preceding (e.g., point '-1') or succeeding (e.g., point '2') time points. These measurements are averaged across multiple stimulation events to yield an average response profile, which may be displayed with indicators of statistical variance such as standard error of the mean (SEM). This analysis quantifies the immediate, short-term impact of a single stimulation pulse or burst, for example, as a percentage change in the measured parameter. This method is extendable to other types of sensor data, including but not limited to other physiological signals, patient-reported outcomes, or contextual data, to evaluate the immediate effect of stimulation on various aspects of the patient's condition.

[0165] FIG. 17B depicts a method for analyzing the cumulative effect of stimulation over a longer period. A statistical property of a measured parameter, such as the variance of nerve activity, is analyzed across different epochs. The analysis compares the parameter's variance during a first sensing period 1740 before a session of closed-loop stimulation, a second period 1750 during the closed-loop stimulation, and a third sensing period 1760 after the stimulation has ceased. A graphical representation, such as a box plot, can be used to visualize the distribution of the parameter's variance in each epoch. This allows for an evaluation of how the stimulation therapy modulates the stability of the physiological signal and whether there are lingering therapeutic effects after the stimulation period ends. This provides insight into the longer-term or cumulative impact of the therapy. As with the immediate effect analysis, this method can be applied to a wide range of sensor datatypes.

[0166] With regard to the above detailed description, like reference numerals used therein may refer to like elements that may have the same or similar dimensions, materials, and configurations. While particular forms of embodiments have been illustrated and described, it will be apparent that various modifications can be made without departing from the spirit and scope of the embodiments herein. Accordingly, it is not intended that the invention be limited by the forgoing detailed description.

[0167] In an embodiment, a stimulus application system can include an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction. The system can further include at least one controller configured to receive detection information pertaining to the physiological signals detected by the sensor unit at each predetermined time interval, process the detection information for each predetermined time interval for comparison to a baseline level, with the baseline level dynamically changeable for each predetermined time interval, and generate a stimulation instruction if the processed detection information for a current time interval exceeds the baseline level for the current time interval by a threshold amount.

[0168] In embodiments, the at least one controller is further configured to process the detection information for each predetermined time interval by decoding the detection information.

[0169] In embodiments, the at least one controller is configured to compare the detection information to the baseline level by comparing the decoded detection information to the baseline level.

[0170] In embodiments, the at least one controller is further configured, for each current time interval, to generate a flat data level comprising averaged processed detection information over a predetermined period of time, generate an adjustment threshold based on the flat data level and selectively change the baseline level based on a comparison of the adjustment threshold to the flat data level.

[0171] In embodiments, the adjustment threshold is a predetermined percentage higher than the flat data level.

[0172] In embodiments, the at least one controller is configured to compare the adjustment threshold data to the flat data level by comparing the adjustment threshold for the current time interval to the flat data level for the previous time interval. In embodiments, the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the current time interval if the adjustment threshold for the current time interval is lower than the flat data level for the previous time interval.

[0173] In embodiments, the at least one controller is configured to keep the baseline level for the previous time interval as the baseline level for the current time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is lower than the flat data level for the previous time interval.

[0174] In embodiments, the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the previous time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is not lower than the flat data level for the previous time interval.

[0175] In embodiments, the predetermined stimulation location is a nerve.

[0176] In embodiments, the predetermined detection location is a nerve.

[0177] In embodiments, the nerve is the sacral nerve.

[0178] In embodiments, the detection information comprises nerve activity data.

[0179] In embodiments, the at least one controller is part of the implant device.

[0180] In an embodiment, the system can further include a wearable recharging device configured to recharge a battery of the implant device.

[0181] In embodiments the system can further include an external programmer configured to wirelessly communicate with the implant device and the at least one controller is part of the external programmer.

[0182] In an embodiment a method includes applying electrical stimulation to a predetermined stimulation location in a body of a patient using the implant device and / or the system as described above.

[0183] The entirety of each patent, patent application, publication, and document referenced herein is hereby incorporated by reference. Citation of the above patents, patent applications, publications and documents is not an admission that any of the foregoing is pertinent prior art, nor does it constitute any admission as to the contents or date of these documents.

[0184] Modifications may be made to the foregoing embodiments without departing from the basic aspects of the technology. Although the technology may have been described in substantial detail with reference to one or more specific embodiments, changes may be made to the embodiments specifically disclosed in this application, yet these modifications and improvements are within the scope and spirit of the technology. The technology illustratively described herein may suitably be practiced in the absence of any element(s) not specifically disclosed herein. The terms and expressions which have been employed are used as terms of description and not of limitation and use of such terms and expressions do not exclude any equivalents of the features shown and described or portions thereof and various modifications are possible within the scope of the technology claimed. Although the present technology has been specifically disclosed by representative embodiments and optional features, modification and variation of the concepts herein disclosed may be made, and such modifications and variations may be considered within the scope of this technology.

Claims

What is claimed is:

1. A stimulus application system, comprising: an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction; and at least one controller configured to: receive detection information pertaining to the physiological signals detected by the sensor unit at each predetermined time interval; for a current time interval, generate a flat data level comprising averaged processed detection information over a predetermined period of time; generate an adjustment threshold based on the flat data level; selectively update a baseline level for the current time interval based on a comparison of the adjustment threshold for the current time interval with the flat data level from a previous time interval; and generate a stimulation instruction if the processed detection information for the current time interval exceeds the baseline level for the current time interval by a threshold amount.

2. The system of claim 1, wherein the at least one controller is further configured to process the detection information for each predetermined time interval by decoding the detection information.

3. The system of claim 2, wherein the at least one controller is configured to compare the decoded detection information to the baseline level.

4. The system of claim 1, wherein the adjustment threshold is a predetermined percentage higher than the flat data level.

5. The system of claim 1, wherein the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the current time intervalif the adjustment threshold for the current time interval is lower than the flat data level for the previous time interval.

6. The system of claim 1, wherein the at least one controller is configured to keep the baseline level for the previous time interval as the baseline level for the current time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is lower than the flat data level for the previous time interval.

7. The system of claim 1, wherein the at least one controller is configured to set the baseline level for the current time interval to the flat data level for the previous time interval if the adjustment threshold for the current time interval is not lower than the flat data level from the previous time interval and the baseline level for the previous time interval is not lower than the flat data level for the previous time interval.

8. A stimulus application system, comprising: an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction; and at least one controller configured to: receive detection information pertaining to the physiological signals detected by the sensor unit at each of a plurality of time intervals; determine a baseline level for a current time interval by: performing a Fast Fourier Transform (FFT) on the detection information to identify a dominant frequency corresponding to an underlying physiological event; defining a search time window having a variable size based on the identified dominant frequency; and identifying the baseline level as a minimum value of the processed detection information within the search time window; and generate a stimulation instruction if the processed detection information for the current time interval exceeds the baseline level by a threshold amount.

9. The system of claim 8, wherein the at least one controller is configured to generate the stimulation instruction by: determining a dynamic peak level for the current time interval; and calculating the threshold amount for the current time interval as a function of both the baseline level and the dynamic peak level.

10. The system of claim 9, wherein the dynamic baseline level and the dynamic peak level are determined by: performing a Fast Fourier Transform (FFT) on the detection information to identify a dominant frequency; defining a search time window based on the dominant frequency; and identifying the dynamic baseline level as a minimum value and the dynamic peak level as a maximum value within the search time window.

11. The system of any of claims 1-10, wherein the at least one controller is further configured to optimize a minimum baseline threshold parameter used to determine the baseline level by iteratively performing the steps of:(a) calculating a baseline value for a plurality of physiological cycles using the minimum baseline threshold parameter;(b) in response to determining that the calculated baseline value is unstable within a single physiological cycle, increasing the value of the minimum baseline threshold parameter; and(c) in response to determining that the calculated baseline value is the same for two or more consecutive physiological cycles, decreasing the value of the minimum baseline threshold parameter.

12. The system of any of claims 1-11, wherein the at least one controller is further configured to optimize the threshold amount by: receiving at least one temporal reference marker from a user, the marker corresponding to a patient-sensation event; calculating a timing deviation between a stimulation onset time and a time corresponding to the temporal reference marker for a plurality of threshold amounts; andidentifying an optimal threshold amount that minimizes the timing deviation.

13. The system of any of claims 1-12, wherein the sensor unit further comprises an accelerometer, and wherein the at least one controller is further configured to process data from the accelerometer to distinguish between physiological signals and motion artifacts.

14. The system of any of claims 1-13, wherein the at least one controller is further configured to: receive a plurality of input parameters; generate a multivariate predictive model based on the plurality of input parameters; and determine a stimulation threshold for a current time interval based on an output of the predictive model, wherein the stimulation instruction is generated if the processed detection information for the current time interval exceeds the determined stimulation threshold.

15. The system of claim 14, wherein the plurality of input parameters includes at least one parameter selected from the group consisting of: a local maximum of the processed detection information, a local minimum of the processed detection information, a baseline- to-peak time, a peak-to-peak cycle time, a patient-reported sensation, a curve fitting coefficient of the processed detection information, a patient's age, a patient's gender, a liquid intake amount, and a time of day.

16. The system of claim 14, wherein the predictive model comprises a machine learning model.

17. The system of claim 14, wherein the at least one controller is further configured to receive an input from an external programmer to calibrate the predictive model.

18. The system of any preceding claim, wherein the predetermined stimulation location and predetermined sensing location is a nerve.

19. The system of claim 18, wherein the nerve is the sacral nerve.

20. A method of applying electrical stimulation to a predetermined stimulation location in a body of a patient, the method comprising: detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals; processing, with at least one controller, detection information pertaining to the detected physiological signals for each of the plurality of time intervals; for a current time interval, generating a flat data level comprising averaged processed detection information over a predetermined period of time; generating an adjustment threshold based on the flat data level of the current time interval; selectively updating a baseline level for the current time interval based on a comparison of the adjustment threshold for the current time interval with the flat data level from a previous time interval; generating a stimulation instruction if the processed detection information for the current time interval exceeds the updated baseline level for the current time interval by a threshold amount; and applying, with a stimulus circuit unit of the implanted device, an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

21. A stimulus application system, comprising: an implant device configured to be implanted in a body of a patient including a sensor unit configured to detect physiological signals at a predetermined detection location in the body at predetermined time intervals and a stimulus circuit unit configured to apply an electrical stimulus to a predetermined stimulation location in the body upon receiving a stimulation instruction; and at least one controller configured to: receive detection information pertaining to the physiological signals detected by the sensor unit at each predetermined time interval; process the detection information; calculate a rate of change of the processed detection information for a current time interval; and generate a stimulation instruction in response to the calculated rate of change exceeding a slope threshold.

22. The system of claim 21, wherein the slope threshold is a predetermined fixed value.

23. The system of claim 22, wherein the predetermined fixed value corresponds to a rate of change indicative of a rapid physiological event.

24. The system of claim 21, wherein the slope threshold is a dynamic threshold calculated based on an average of rate of change values from one or more previous time intervals.

25. A method of applying electrical stimulation to a predetermined stimulation location in a body of a patient, the method comprising: detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals; processing, with at least one controller, detection information pertaining to the detected physiological signals for each of the plurality of time intervals; calculating, with the at least one controller, a rate of change of the processed detection information for a current time interval; generating, with the at least one controller, a stimulation instruction if the calculated rate of change exceeds a slope threshold; and applying, with a stimulus circuit unit of the implanted device, an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

26. The method of claim 25, wherein the slope threshold is a predetermined fixed value.

27. The method of claim 25, wherein the slope threshold is a dynamic threshold calculated based on an average of previously calculated rates of change.

28. A method of applying electrical stimulation to a predetermined stimulation location in a body of a patient, the method comprising: detecting, with a sensor unit of an implanted device, physiological signals at a predetermined detection location in the body at a plurality of time intervals; processing, with at least one controller, detection information pertaining to thedetected physiological signals; receiving, at the at least one controller, a plurality of input parameters related to the patient or the physiological signals; generating, with the at least one controller, a multivariate predictive model based on the plurality of input parameters; determining a stimulation threshold for a current time interval based on an output of the predictive model; generating a stimulation instruction if the processed detection information for the current time interval exceeds the determined stimulation threshold; and applying, with a stimulus circuit unit of the implanted device, an electrical stimulus to the predetermined stimulation location based on the stimulation instruction.

29. The method of claim 28, wherein the plurality of input parameters includes at least one of a physiological signal temporal characteristic, a physiological signal morphology parameter, a patient-reported subjective parameter, a patient demographic parameter, or a patient behavioral parameter.

Images