# US 11,998,743 B2 · Systems and methods for using electrospinogram signals for closed loop control in Spinal Cord Stimulation therapy > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,998,743 B2 - Document ID: US11998743B2 - Status: Granted - Kind: B2 - Application number: 17/137,133 - Filed: 2020-12-29 - Published: 2021-08-19 - Granted: 2024-06-04 - Earliest claimed: 2020-02-14 - Assignees: Boston Scientific Neuromodulation Corporation - Applicants: Boston Scientific Neuromodulation Corporation - Inventors: Rosana Esteller; Tianhe Zhang; Qi An; Gezheng Wen - Web page: https://patentgrep.com/patent/US11998743B2 ## CPC - A61N 1/36062: {Spinal stimulation} - A61B 5/383: Somatosensory stimuli, e.g. electric stimulation - A61B 5/686: {Permanently implanted devices, e.g. pacemakers, other stimulators, biochips (A61B5/6861 takes precedence)} - A61N 1/0551: {Spinal or peripheral nerve electrodes} - A61N 1/36132: {using patient feedback} - A61N 1/36139: {with automatic adjustment} - G16H 20/40: relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture - A61B 5/29: for permanent or long-term implantation - A61B 5/294: for nerve conduction study [NCS] - A61B 5/407: {Evaluating the spinal cord (for locating the epidural space A61B5/4896)} - A61B 5/7267: {involving training the classification device} - A61B 2562/0219: Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches - A61N 1/36071: {Pain} - A61N 1/36164: {Sub-threshold or non-excitatory signals (non-excitatory signals to the heart A61N1/3628)} - A61N: ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY (measurement of bioelectric currents A61B; surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body A61B18/00; anaesthetic apparatus in general A61M; incandescent lamps H01K; infrared radiators for heating H05B) - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract Methods and systems for providing closed loop control of stimulation provided by an implantable stimulator device are disclosed herein. The disclosed methods and systems use a neural feature prediction model to predict a neural feature, which is used as a feedback control variable for adjusting stimulation. The predicted neural feature is determined based on one or more stimulation artifact features. The disclosed methods and systems can be used to provide closed loop feedback in situations, such as sub-perception therapy, when neural features cannot be readily directly measured. ## Claims ### Claim 1 A method of operating an implantable medical device (IMD) comprising an implantable pulse generator (IPG) and a plurality of electrodes configured to contact a patient's tissue, the method comprising: providing sub-threshold stimulation at at least one of the electrodes, wherein the stimulation is configured to not evoke paresthesia in the patient; sensing a stimulation artifact at one or more sensing electrodes, wherein the stimulation artifact comprises a signal formed by an electric field induced in the tissue by the stimulation; determining at least one stimulation artifact feature of the sensed stimulation artifact; determining at least one predicted neural feature from the determined at least one stimulation artifact feature using a prediction model that is configured to predict one or more neural features based on one or more stimulation artifact features, wherein the one or more neural features are indicative of a neural response to the stimulation; and using the determined at least one predicted neural feature to adjust the stimulation. ### Claim 2 (depends on 1) The method of claim 1, wherein using the at least one predicted neural feature to adjust the stimulation comprises adjusting the stimulation to maintain the at least one predicted neural feature within a set-range of values. ### Claim 3 (depends on 1) The method of claim 1, wherein using the at least one predicted neural feature to adjust the stimulation comprises adjusting the stimulation to maintain the at least one predicted neural feature relative to a set-point. ### Claim 4 (depends on 1) The method of claim 1, wherein using the at least one predicted neural feature to adjust the stimulation comprises using a control model to adjust the stimulation to maintain the at least one predicted neural feature with respect to a set point or within a set range. ### Claim 5 (depends on 4) The method of claim 4, wherein the control model is selected from the group consisting of Kalman filtering algorithms, heuristic algorithms, simple threshold model, proportional-integral-derivative (PID) controller models, and hybrids of the same. ### Claim 6 (depends on 1) The method of claim 1, further comprising determining the prediction model. ### Claim 7 (depends on 6) The method of claim 6, wherein determining the prediction model comprises: providing supra-threshold stimulation to the patient; recording an electrospinogram (ESG) of the patient; processing the ESG to extract a stimulation artifact signal and an evoked compound action potential (ECAP) signal from the ESG; determining one or more features of the stimulation artifact signal; determining one or more features of the ECAP signal; and determining a mathematical expression that expresses the one or more features of the ECAP signal as a function of the one or more features of the stimulation artifact signal. ### Claim 8 (depends on 7) The method of claim 7, wherein the mathematical expression comprises a linear regression. ### Claim 9 (depends on 7) The method of claim 7, wherein the mathematical expression comprises a fitting algorithm selected from the group consisting of regression models, Bayesian networks, genetic algorithms, support vector machines (SVM), decision trees, neural networks, and hybrids of the same. ### Claim 10 An implantable medical device (IMD) comprising an implantable pulse generator (IPG) and a plurality of electrodes configured to contact a patient's tissue, wherein the IPG is configured to: provide sub-threshold stimulation at at least one of the electrodes, wherein the stimulation is configured not to evoke paresthesia in the patient; sense a stimulation artifact at one or more sensing electrodes, wherein the stimulation artifact comprises a signal formed by an electric field induced in the tissue by the stimulation; determine at least one stimulation artifact feature of the sensed stimulation artifact; determine at least one predicted neural feature from the determined at least one stimulation artifact feature using a prediction model that is configured to predict one or more neural features based on one or more stimulation artifact features, wherein the one or more neural features are indicative of a neural response to the stimulation; and use the determined at least one predicted neural feature to adjust the stimulation. ### Claim 11 (depends on 10) The IMD of claim 10, wherein using the at least one predicted neural feature to adjust the stimulation comprises adjusting the stimulation to maintain the at least one predicted neural feature within a set-range of values. ### Claim 12 (depends on 11) The IMD of claim 11, wherein using the at least one predicted neural feature to adjust the stimulation comprises adjusting the stimulation to maintain the at least one predicted neural feature relative to a set-point. ### Claim 13 (depends on 12) The IMD of claim 12, wherein using the at least one predicted neural feature to adjust the stimulation comprises using a control model to adjust the stimulation to maintain the at least one predicted neural feature with respect to a set point or within a set range. ### Claim 14 (depends on 13) The IMD of claim 13, wherein the control model is selected from the group consisting of Kalman filtering algorithms, heuristic algorithms, simple threshold model, and proportional-integral-derivative (PID) controller models. ### Claim 15 (depends on 10) The IMD of claim 10, wherein the prediction model determines the at least one predicted neural feature as a weighted linear combination of a plurality of stimulation artifact features. ### Claim 16 (depends on 10) The IMD of claim 10, wherein the stimulation artifact is sensed at a sense amplifier in the IPG. ### Claim 17 (depends on 10) The IMD of claim 10, wherein the prediction model is programmed into a control circuitry of the IPG. ### Claim 18 (depends on 10) The IMD of claim 10, wherein the IMD comprises a Spinal Cord Stimulator device.