Artificial intelligence and machine learning for neuromodulation of neuronal targets in treatment of a medical condition

Abstract

The present disclosure is directed to a bio-feedback sensor in combination with AI and ML that can be implemented to optimize neuromodulation parameters for the management of a medical condition such a type II diabetes. The bio-feedback obtained by the sensors is optimized by the ML tools to predict and recommend the upregulation or downregulating the neuroregulators to modulate treatment. The algorithms could also learn to sense a life-threating pathological state, or bio-chemical imbalance, and in the case of diabetes a life-threating low blood sugar level the system overrides to stimulate the target fibers that control organs to regulate blood glucose levels.

Description

Attorney Docket No. 40759.0090-RESH-022-01WGARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN COMBINATION WITH NEUROMODULATION OF TWO OR MORE NEURONAL TARGETS IN TREATMENT OF A MEDICAL CONDITIONFIELD OF THE DISCLOSURE

[0001] The present disclosure is directed to the use of artificial intelligence and machine learning in combination with neuromodulation and, more particularly, the neuromodulation of two or more neuronal targets in the treatment of a medical condition such as type 2 diabetes.BACKGROUND

[0002] The peripheral nervous system is composed of two major branches, the somatosensory which traffics, for example, touch and pain information to the central nervous system (CNS). The other branch consists of the autonomic nervous system which carries information of the state of internal organs to the CNS and also information from the CNS to control the internal organs. The autonomic nervous system can further be broken up into two divisions, the sympathetic and parasympathetic nervous systems. These act in an antagonistic manner. The sympathetic puts the body in a “flight or fight” mode. For example activation of the sympathetic nervous system takes blood away from the stomach and guides to muscles, slows digestion and increases release of adrenalin. The parasympathetic nervous system put the body into a “rest and digest” mode. For example its activation would guide blood to the stomach and increases digestive processes. Most, if not all internal organs are innervated by both the sympathetic and parasympathetic nervous systems.Attorney Docket No. 40759.0090-RESH-022-01WG

[0003] In terms of neuromodulation (stimulation or conduction block) with the intent to control the function of an internal organ, modulation of only one ami of the autonomic nervous system may not have maximal, or very little, efficacy. This is due to the fact that when one arm is modulated the other arm may compensate for the exogenous up or down regulation of the nerve innervating the organ. Accordingly, the neuromodulation of a single neuronal target has often failed to produce an effect sufficient to treat a number of medical conditions.

[0004] An example of multisite neuromodulation for the treatment of a medical condition would be modulating multiple points of the vagus nerve for blood glucose regulation for type 2 diabetes. The vagus nerve innervates organ systems involved with blood glucose regulation. These include the pancreas which secrets hormones, glucagon and insulin, and the liver which is involved with glucose storage (and release) as well as controlling insulin resistance. By blocking (ligation or application of high frequency alternating current (HFAC)) of the hepatic branch (or the anterior vagus nerve central to the hepatic branch) with concurrent stimulation of the celiac branch (or the posterior vagus nerve central to the celiac branch) an increase in glycemic control can be achieved.

[0005] Furthermore, stimulation alone of the celiac branch causes an increase in blood glucose concentration. When block is added there is a decrease in blood glucose concentration. Thus, blood glucose levels can be held at a desired level by changing the block and stimulation output. To streamline and optimize changing block and stimulation output it would be desirable for a neuroregulating system to have frequently updated information regarding blood glucose concentrations as well as Artificial Intelligence (Al) and Machine Learning (ML) to navigate the neuroregulators on how to optimize therapy output.Attorney Docket No. 40759.0090-RESH-022-01WGSUMMARY

[0006] The present disclosure is directed to the neuromodulation (stimulation or conduction block) of two or more neuronal targets in combination with bio-feedback, Al and ML in treatment of a medical condition. The neuronal targets can be sympathetic nerves, parasympathetic nerves or a combination of sympathetic and parasympathetic nerves. The neuromodulation of the neuronal targets can comprise stimulation of the neuronal targets, a conduction block of the neuronal targets or a combination of stimulation and conduction blocks of the neuronal targets. The neuromodulation is performed by one or more neuroregulators which can comprise electrical or non-electrical neuroregulators. The neuromodulation of each of the neuronal targets can include the same or different start / end times, the same or different durations, and / or the same or different neuromodulation patterns. The neuromodulation of the first neuronal target works in cooperation with the neuromodulation of the second neuronal target to treat a medical condition.

[0007] In some embodiments it may be desirable to modulate two or more targets in the central nervous system or the combination of modulation of one or more target(s) in the central nervous system with modulation of one or more target(s) in the peripheral nervous system. For example, the neural circuitry that influences depression is complex and diffuse in the limbic system of the brain. However, it has been shown that electrical modulation of distinct points, or inputs, in this circuitry will treat refractory depression to various degrees; these include standalone stimulation vagus nerve afferents or standalone stimulation of Brodmann area 25. However, these standalone treatments may still be ineffective or not reach maximal efficacy. Dual electrical modulation ofAttorney Docket No. 40759.0090-RESH-022-01WGpoints (e.x. Brodmann area 25), or inputs (e.x. vagus nerve afferents), influencing multiplex depression neuronal circuitry would be a novel method of treatment for refractory depression.

[0008] What is meant by neuromodulation is the targeted placement of an external agent to the central or peripheral nervous system. This agent may deliver electrical impulses, secrete drug(s), ablate, or cause a localized immune response. Multiple target neuromodulation may consist of any combination of targeted placement of an agent.

[0009] In some embodiments a second or multiple target(s) for electrical stimulation would be smooth muscle. Targets would include, but not limited to, the stomach, duodenum, small intestine, large intestine, colon and bladder for the treatment of a medical condition. For example the antrum of the stomach and / or the duodenum could be stimulated to increase gastric emptying and decrease absorption of mono- and poly-saccharides for the treatment of type 2 diabetes. Antrum and / or duodenal stimulation could be combined with sub-diaphragmatic vagus nerve stimulation to further increased gastric emptying. Also, blocking of fibers innervating the liver would decrease insulin resistance and be used with antrum and / or duodenal stimulation.

[0010] Stimulation of the intestinal tract could be used to treat diseases, such as, but not limited to, Crohn's disease, which can produce chronic intestinal obstruction. Stimulation of nerves innervating the enteric nervous system as well as intestinal stimulation may be used to treat chronic intestinal obstruction.

[0011] Neuromodulation in combination of smooth muscle stimulation could also be used to treat diseases involved with voiding. Stimulation of the bladder in combination of blocking the pudendal nerve, which innervates the external ureteral sphincter, would induce voiding in patientsAttorney Docket No. 40759.0090-RESH-022-01WGwith spinal cord injuries. Stimulation of the colon in combination of blocking inferior rectal nerves innervating the external anal sphincter could be used to increase elimination of feces in patients with spinal cord injuries. Stimulation of the pudendal nerve could be used for the treatment of urinary incontinence and stimulation of inferior rectal nerves could be used to treat fecal incontinence

[0012] One aspect of the system is the ability to change the neuromodulation parameters based on information regarding the state of a medical condition being treated. Knowing what parameters to change would be done through Al and ML outside the body. An initial set of instructions would be given to the implantable system. A sensor would then be used to determine how the initial instructions impacted the medical condition. The sensing system would have a temporal resolution of about 1 second to 10 min. Small changes in the output of the implantable device would then be implemented and instruct the outside Al and ML engine if the change was positive or negative. This closed loop system would keep learning (on a time scale of 1 second to 10 min) until an optimal set of parameters are met to treat the medical condition. Once the optimal set of parameters are met, minor changes in parameters would still take place to learn if the body has adapted to the modulation. Thus, optimal parameters may not be static but change over time. In some embodiments a clinician would review and approve the change of the parameters that are outside the nominal operating range in therapy.

[0013] In some embodiments the system would be used to treat type 2 diabetes. During the day blood glucose rises and falls due to meals containing mono- and polysaccharides. This rise and fall of blood glucose is exacerbated in patients with T2DM which causes organ damage.Attorney Docket No. 40759.0090-RESH-022-01WGClamping blood glucose to a desired level in patients with T2DM would be desirable. To achieve a glucose clamp a high frequency blocking signal (from 200 Hz to 100K Hz) would be applied to the hepatic branch of the vagus nerve, or the anterior branch of the vagus nerve cranial to the branching point of the hepatic nerve, and simultaneous delivery of a low frequency signal (0.01 to 199 Hz) to the celiac branch of the vagus nerve, or the posterior branch of the vagus nerve cranial to the branching point of the celiac nerve. A glucose sensor would also be implanted to monitor (at a rate on the order of about 1 second to 10 min) blood glucose levels before and during the delivery of the electrical signals. The sensor would then relay information on blood glucose levels to a programming device (such as a smart phone). The Al and ML algorithm in the programming device has data processing and analytics to generate a predictive recommendation for enhancing the parameters sent to the nerves based on the bio-feedback.

[0014] Different parameters of the signals would be to turn block or stimulation on or off, to change the current amplitude (0.25 to 20 mA or 0.25 to 20 Volts) of HF AC block and stimulation, the frequencies of block (200 Hz-10k Hz) and stimulation (0.01 to 199 Hz), changing signal patterns of the block and stimulation (such as bursting), changing to only block or only stimulation, reversing block to stimulation and / or stimulation to block, the time of day to apply signals, or turning off both block and stimulation entirely. With time if the body adapts to a particular paradigm the Al and ML component would adjust the system to maximize therapeutic effect.

[0015] In addition to ML for an individual patient, data would be uploaded to a central computer system / repository / on a cloud / server using a HIPAA compliant system and using block chain protocol / hyper-ledger system. The central computer system / repository / or a cloud / server will utilize Al and ML tools to track population trends in glucose levels over time and how therapyAttorney Docket No. 40759.0090-RESH-022-01WGwas recommended. This continuous loop would gain further knowledge and enhance the Al and ML system for type 2 diabetes patients.

[0016] The communication between the neuroregulator, glucose sensor and programming device / app / handheld device can be through, but not limited to, blue tooth technology, radio frequency, WIFI or sound. Manual input can be entered into the smart device such as blood glucose measured by an external manual glucose monitor, food intake, insulin injections (by selfadministration or an insulin pump) and physical activity (which may include the use of accelerometers in the neuroregulators).DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is an exemplary schematic of a neuromodulation system for neuromodulating two or more neuronal targets according to the present disclosure.

[0018] Figure 2 is an exemplary schematic of another neuromodulation system for neuromodulating two or more neuronal targets according to the present disclosure.

[0019] Figure 3 is an exemplary schematic of another neuromodulation system for neuromodulation of two or more neuronal targets in combination with a bio-feedback sensor and an outside programming device with Al and ML tools. This system can communicate with a clinic which in turn communicates with a central computer system / repository / on a cloud / server using a HIPAA compliant system which contains an Al and ML algorithm. This in turn navigates the neuroregulator system for the optimal therapy output according to the present disclosure.Attorney Docket No. 40759.0090-RESH-022-01WG

[0020] Figure 4 is a time course, in relation to a Type II Diabetes Rat Model Study described herein, of various conditions in experimental group 2.

[0021] Figure 5 is graph, in relation to the Type II Diabetes Rat Model Study described herein, illustrating percent change in blood glucose concentration in sham or celiac branch stimulation and hepatic nerve vagotomy (experimental group 1) prior to a glucose challenge.

[0022] Figure 6 is a graph, in relation to the Type II Diabetes Rat Model Study described herein, illustrating time course of percent change of blood glucose concentration following a glucose challenge for the sham, celiac branch stimulation and hepatic nerve vagotomy (experimental group 1) and celiac branch stimulation with concurrent delivery of 5000 Hz to the hepatic nerve (experimental group 2) arms.

[0023] Figure 7 is a bar chart, in relation to the Type II Diabetes Rat Model Study described herein, illustrating peak glucose concentration (as measured as percentage change from baseline) following a glucose challenge for each experimental condition,

[0024] Figure 8 is a graph, in relation to the Type II Diabetes Rat Model Study described herein, of the percentage change in blood glucose versus time for the celiac branch stimulation and simultaneous delivery of 5000 Hz to the hepatic nerve (experimental group 2) arm before a glucose tolerance test.

[0025] Figure 9 is a graph, in relation to the Type II Diabetes Rat Model Study described herein, of the percentage change in blood glucose concentration over time with stimulation of theAttorney Docket No. 40759.0090-RESH-022-01WGceliac branch and concurrent delivery of 5000 Hz to the hepatic nerve (dashed line) with two glucose challenges.DETAILED DESCRIPTION

[0026] The present disclosure is directed to the neuromodulation (stimulation or conduction block) of two or more neuronal targets in treatment of a medical condition. The neuromodulation system also contains a bio-feedback sensor which gives information to an external artificial intelligence (Al) and machine learning (ML) engine to optimize the neuromodulation signals. The neuronal targets generally comprise sympathetic and / or parasympathetic nerves, which typically innervate an internal organ. In various embodiments, the neuronal targets may also include central nervous system tissue, smooth muscle tissue, or combinations thereof. One or more neuroregulators provide the desired stimulation or conduction block to the neuronal targets, which work in cooperation to treat a medical condition. The cooperative nature of the neuromodulation allows for enhanced therapeutic efficacy compared to single-target approaches, as the system can address multiple pathways simultaneously while avoiding compensatory responses that may limit treatment effectiveness. This multi-target approach recognizes that the autonomic nervous system operates through complex interconnected pathways, where modulation of a single target may trigger compensatory mechanisms that reduce therapeutic benefit. By simultaneously modulating multiple complementary or antagonistic pathways, the system can achieve more comprehensive physiological control and sustained therapeutic effects. The cooperative interaction between neuronal targets may manifest through various mechanisms including synergistic amplification where the combined therapeutic effect exceeds the sum of individual target effects, complementaryAttorney Docket No. 40759.0090-RESH-022-01WGpathway modulation where different targets address distinct aspects of the same pathological process, or coordinated physiological responses that enhance overall homeostatic regulation while preventing maladaptive compensatory mechanisms that could undermine treatment goals.

[0027] In various embodiments, the neuromodulation system comprises a first neuroregulator configured to apply a stimulation or a conduction block to a first neuronal target, and a second neuroregulator configured to apply a stimulation or a conduction block to a second neuronal target that is different from the first neuronal target. The system may further include additional neuroregulators configured to modulate third, fourth, or more neuronal targets as clinically appropriate for complex medical conditions requiring comprehensive neural pathway intervention. The system further includes a bio-feedback sensor configured to monitor a physiological parameter related to a medical condition, and a programming device comprising communication circuitry and an artificial intelligence and machine learning algorithm. In some implementations, multiple bio-feedback sensors may be employed to monitor different physiological parameters simultaneously, providing a more comprehensive assessment of the patient's condition. These sensors may include glucose sensors for metabolic monitoring, blood pressure monitors for cardiovascular assessment, heart rate sensors for autonomic function evaluation, oxygen saturation monitors for respiratory status, temperature sensors for metabolic activity assessment, pH sensors for acid-base balance monitoring, neurotransmitter sensors for neurochemical analysis, or other specialized biosensors depending on the medical condition being treated. The programming device is configured to receive data from the bio-feedback sensor and generate optimized neuromodulation parameters through real-time analysis and predictive modeling. The Al and ML algorithms may employ various computational approaches including deep neural networks forAttorney Docket No. 40759.0090-RESH-022-01WGcomplex pattern recognition and data analysis, reinforcement learning for adaptive optimization based on patient responses, Bayesian optimization for parameter tuning under uncertainty, genetic algorithms for evolutionary optimization of treatment protocols, support vector machines for classification of physiological states, decision trees for rule-based therapeutic decisions, or ensemble methods that combine multiple algorithms for enhanced performance and robustness in diverse clinical scenarios. In some embodiments, the Al system may implement convolutional neural networks (CNNs) specifically designed for processing time- series physiological data, with specialized architectures such as 1D CNNs for temporal pattern recognition in biosignals. 2D CNNs for analyzing multi-dimensional physiological data matrices, or 3D CNNs for processing spatiotemporal physiological patterns across multiple sensor locations. Recurrent neural networks (RNNs) may be employed for sequential data processing, including long short-term memory (LSTM) networks for capturing long-term dependencies in physiological patterns, gated recurrent units (GRUs) for efficient processing of sequential biosignal data, or bidirectional RNNs for analyzing physiological patterns that benefit from both forward and backward temporal context. Advanced transformer architectures may be utilized for attention-based analysis of physiological data, including self-attention mechanisms for identifying critical time periods in physiological responses, multi-head attention for simultaneously analyzing different aspects of physiological data, or temporal transformers specifically adapted for processing continuous physiological monitoring data. The ML system may incorporate federated learning approaches that enable knowledge sharing across multiple patient populations while preserving individual privacy, distributed learning algorithms that can operate across multiple clinical sites, or transfer learning techniques that leverage knowledge from related medical conditions to improve treatmentAttorney Docket No. 40759.0090-RESH-022-01WGoptimization for rare diseases or conditions with limited training data. Reinforcement learning implementations may include Q-learning algorithms for discrete therapy parameter optimization, policy gradient methods for continuous parameter adjustment, actor-critic architectures for balancing exploration and exploitation in therapy optimization, or multi-agent reinforcement learning for coordinating multiple neuroregulators in complex treatment scenarios. The Al system may employ Bayesian neural networks for uncertainty quantification in treatment recommendations, variational autoencoders for learning compact representations of physiological states, generative adversarial networks for data augmentation in scenarios with limited patient data, or graph neural networks for modeling complex physiological interactions between different organ systems and neural pathways. Advanced optimization techniques may include particle swarm optimization for global parameter optimization, simulated annealing for avoiding local optima in therapy parameter selection, evolutionary strategies for robust optimization under physiological variability, or multi-objective optimization algorithms for balancing competing therapeutic goals such as efficacy versus side effect minimization. Communication circuitry is configured to transmit the optimized neuromodulation parameters to the first neuroregulator and the second neuroregulator, wherein the stimulation or conduction block of the first neuronal target works in cooperation with the stimulation or conduction block of the second neuronal target to treat the medical condition based on the optimized neuromodulation parameters generated by the artificial intelligence and machine learning algorithm. The communication may occur through various protocols including wireless transmission using radio frequency or Bluetooth technology, wired connections for secure clinical programming, or hybrid approaches depending on the specific implementation requirements. Advanced communication systems may incorporate meshAttorney Docket No. 40759.0090-RESH-022-01WGnetworking for redundant communication paths, adaptive protocols that automatically adjust transmission parameters based on signal quality or interference conditions, error correction algorithms to ensure data integrity, or encryption protocols to protect patient data and prevent unauthorized access to the therapeutic system.

[0028] The first neuronal target and the second neuronal target may each comprise one or more sympathetic nerves, one or more parasympathetic nerves, or a combination thereof. In specific embodiments, the first neuronal target comprises one or more sympathetic nerves and the second neuronal target comprises one or more parasympathetic nerves, allowing for balanced modulation of the autonomic nervous system's opposing branches. Alternative configurations may include both targets comprising sympathetic nerves for coordinated sympathetic modulation, both targets comprising parasympathetic nerves for enhanced parasympathetic effects, or mixed configurations where one or both targets include combinations of nerve types to achieve complex physiological responses. The neuronal targets may also include afferent nerve fibers that carry sensory information from peripheral organs to the central nervous system, efferent nerve fibers that transmit motor commands from the central nervous system to target organs, or both, allowing for modulation of sensory input, motor output, or bidirectional neural communication. Afferent fibers play crucial roles in physiological feedback loops and may be modulated to alter sensory perception or reflex responses, while efferent fibers control organ function and may be targeted to directly influence therapeutic outcomes. By selectively modulating these different fiber types, the system can achieve precise control over specific physiological functions while maintaining the integrity of other neural pathways. In some embodiments, the targets may include specific nerve branches such as individual branches of the vagus nerve including hepatic, celiac, gastric, orAttorney Docket No. 40759.0090-RESH-022-01WGpulmonary branches, nerve plexuses such as the celiac plexus or hypogastric plexus that innervate multiple organs, or individual nerve fascicles within larger nerve trunks to achieve highly precise therapeutic effects. For example, the system may target specific branches of the vagus nerve such as the hepatic branch for liver function modulation, celiac branch for pancreatic and gastrointestinal effects, or gastric branches for stomach function control, or may focus on particular components of sympathetic ganglia or nerve plexuses to achieve targeted organ- specific effects. The neuroregulators may apply various combinations of stimulation and conduction blocks, including stimulation to both targets for coordinated activation, conduction blocks to both targets for comprehensive inhibition, or a combination where one target receives stimulation while the other receives a conduction block to achieve balanced physiological modulation. Additional variations may include alternating stimulation and blocking patterns that cycle between different therapeutic approaches, sequential activation of different targets in predetermined temporal sequences, or complex multi-phase protocols that adapt based on real-time feedback from physiological monitoring systems. These sophisticated modulation patterns may be designed to mimic natural physiological rhythms such as circadian cycles, respond to specific patient activities such as meals or exercise, or adapt to changing pathological conditions to maintain optimal therapeutic outcomes throughout varying physiological states.

[0029] In certain embodiments the two or more neuronal targets are: (a) neuromodulated over a common time period for synchronized therapeutic effects; (b) neuromodulated during overlapping time periods to achieve coordinated but not identical timing; (c) neuromodulated in distinct non-overlapping time periods for sequential therapeutic interventions; and / or (d) neuromodulated in any combination of (a), (b), and (c) to create complex temporal therapeuticAttorney Docket No. 40759.0090-RESH-022-01WGpatterns. The neuromodulation of each of the neuronal targets can include the same or different start / end times, the same or different durations, and / or the same or different neuromodulation patterns to optimize therapeutic outcomes for specific medical conditions and individual patient responses. This temporal flexibility allows for sophisticated treatment protocols that can be optimized for specific medical conditions and individual patient responses, taking into account factors such as disease progression, symptom patterns, or physiological variations. In some implementations, the timing may be synchronized with circadian rhythms to optimize treatment during specific phases of the sleep-wake cycle, meal times when metabolic demands change significantly, sleep cycles to avoid interference with natural rest patterns, or other physiological patterns such as hormonal fluctuations or activity levels. For example, in diabetes treatment, the system may increase neuromodulation intensity during meal times when glucose levels typically rise due to carbohydrate absorption, or may adjust parameters based on the patient's sleep-wake cycle to optimize metabolic control throughout the day while accounting for natural variations in insulin sensitivity and glucose metabolism. The system may employ adaptive timing algorithms that learn from patient responses and automatically adjust temporal parameters to maximize therapeutic benefit while minimizing side effects or patient discomfort. These algorithms may analyze patterns in physiological data to identify optimal treatment windows, identify optimal treatment windows based on individual patient characteristics, or predict when intervention is most needed based on historical data and current conditions such as activity levels, stress, or environmental factors. Complex temporal patterns may include burst stimulation involving delivery of high-frequency pulses in periodic bursts, ramping protocols that gradually increase or decrease stimulation intensity over time to achieve smooth physiological transitions, intermittentAttorney Docket No. 40759.0090-RESH-022-01WGactivation that provides therapeutic breaks to prevent tolerance or adaptation, or patient-triggered interventions based on specific physiological conditions or symptoms that allow for responsive treatment delivery. Burst stimulation may involve delivering high-frequency pulses in periodic bursts to achieve enhanced neural activation while minimizing energy consumption, while ramping protocols may gradually increase or decrease stimulation intensity over time to achieve smooth physiological transitions and avoid abrupt changes that could cause adverse effects or patient discomfort.

[0030] Stimulation neuroregulators can comprise, for example, electrical stimulators that deliver controlled electrical pulses to neural tissue or chemical agent stimulators that release therapeutic compounds directly to target sites. Additional types may include magnetic stimulators that use focused magnetic fields to induce neural activity through electromagnetic induction, ultrasonic stimulators that employ high-frequency sound waves to achieve precise neural modulation through mechanical energy transfer, optical stimulators that utilize specific wavelengths of light to activate photosensitive neural tissue or optogenetic constructs, or thermal stimulators that use controlled heating or cooling to modulate neural activity through temperaturesensitive ion channels, depending on the specific application requirements and target tissue characteristics. Magnetic stimulators may use focused magnetic fields to induce neural activity through transcranial or implanted coil systems, while ultrasonic stimulators may employ high-frequency sound waves focused through tissue to achieve precise neural modulation without requiring direct tissue contact. Optical stimulators may utilize specific wavelengths of light to activate photosensitive neural tissue or optogenetic constructs that have been genetically modified to respond to light stimulation, and thermal stimulators may use controlled heating or cooling toAttorney Docket No. 40759.0090-RESH-022-01WGmodulate neural activity through temperature-sensitive mechanisms. Electrical stimulators may be configured to deliver electrical pulses with current amplitudes ranging from 0.25 mA to 20 mA, or alternatively from 0.25 to 20 Volts, depending on the impedance characteristics of the target tissue and the desired therapeutic effect. In some embodiments, higher current amplitudes up to 50 mA or voltage ranges up to 50 Volts may be employed for specific applications requiring greater neural activation or for targeting deeper or less accessible neural structures such as deep brain targets or large peripheral nerves with high activation thresholds. The stimulation may comprise low frequency signals with frequencies ranging from 0.01 Hz to 199 Hz to achieve various therapeutic effects depending on the target tissue and desired physiological response. Alternative frequency ranges may include ultra-low frequencies below 0.01 Hz for certain applications such as modulating very slow physiological processes like hormonal cycles or metabolic regulation, or frequencies up to 500 Hz for specialized protocols targeting specific neural fiber types such as motor neurons or sensory fibers with different activation characteristics. In specific embodiments, the stimulation frequency may be approximately 1 Hz for basic neural activation, though other commonly used frequencies include 10 Hz for motor nerve stimulation and muscle activation, 20 Hz for certain autonomic applications such as vagus nerve stimulation, 50 Hz for muscle stimulation and motor function enhancement, or 100 Hz for pain management applications and sensory modulation depending on the target tissue and desired physiological response. The electrical stimulators may be configured to deliver square wave pulses with various pulse widths, such as approximately 4 milliseconds for standard neural activation. Alternative pulse shapes may include sinusoidal waves for smoother neural activation and reduced tissue damage, triangular waves for gradual onset stimulation that minimizes activation thresholds, exponential decay pulsesAttorney Docket No. 40759.0090-RESH-022-01WGfor mimicking natural neural signals and improving biocompatibility, or complex multi-phase waveforms that can selectively activate different neural fiber types based on their electrical properties and activation characteristics. Pulse widths may range from microseconds to several seconds depending on the application and target tissue characteristics, with common ranges including 0.1 to 10 milliseconds for neural stimulation, with shorter pulses typically used for myelinated fibers that have lower activation thresholds and longer pulses for unmyelinated fibers that require more charge for activation.

[0031] Conduction block neuroregulators can comprise, for example, physical section (e.g. ablation) conduction blocks that create permanent interruption of neural pathways, electrical conduction blocks using various waveforms and frequencies, electrical high frequency conduction blocks that reversibly interrupt neural conduction, chemical conduction blocks using local anesthetics or neurotoxins, or optogenetic conduction blocks obtained by delivery of an inhibitory opsin or excitatory opsin that is activated with high frequency light to achieve precise temporal control of neural inhibition. Additional blocking methods may include cryogenic blocks that use controlled cooling to temporarily disable neural conduction through reversible tissue effects, thermal ablation that uses heat to create permanent conduction blocks through controlled tissue destruction, focused ultrasound ablation that uses high-intensity focused ultrasound to create precise lesions without requiring surgical access, or magnetic field-induced blocks that use strong magnetic fields to disrupt neural signaling through electromagnetic interference with neural electrical activity. Electrical conduction blocks may utilize high frequency alternating current (HFAC) with frequencies ranging from 200 Hz to 100 kHz to achieve reversible neural conduction block without permanent tissue damage. Extended frequency ranges may include frequencies upAttorney Docket No. 40759.0090-RESH-022-01WGto 1 MHz for certain applications requiring very precise blocking of specific neural pathways or for targeting neural structures with unique electrical properties, or specialized low-frequency blocking protocols in the range of 50-200 Hz for applications where traditional high-frequency blocking may not be suitable due to tissue characteristics or anatomical constraints. In specific embodiments, the high frequency alternating current conduction block may have frequencies of approximately 5000 Hz for effective neural blocking, though other effective frequencies include 1000 Hz for blocking smaller nerve fibers with lower blocking thresholds, 10,000 Hz for comprehensive nerve blocking across multiple fiber types, 20,000 Hz for deep nerve structures that require higher frequencies for effective blocking, or 50,000 Hz for very precise blocking applications that require minimal current spread to surrounding tissues depending on the nerve type and blocking requirements. The conduction block may be applied with current amplitudes ranging from 0.25 mA to 20 mA, with specific embodiments utilizing approximately 12 mA for effective blocking of medium-sized peripheral nerves. Higher current amplitudes up to 100 mA may be required for certain nerve types such as large motor nerves with high blocking thresholds or for blocking applications in challenging anatomical locations where current spread is limited, while lower amplitudes down to 0.1 mA may be sufficient for sensitive neural tissues or when precise, minimal blocking is desired to avoid effects on adjacent neural structures.

[0032] Neuroregulators, electrical stimulators, electrical conduction blocks and electrical high frequency conduction blocks can be deemed electrical neuroregulators while all other noted neuroregulators can be deemed non-electrical neuroregulators. Hybrid neuroregulators that combine electrical and non-electrical modalities may also be employed to achieve enhanced therapeutic effects by leveraging the advantages of multiple modulation approachesAttorney Docket No. 40759.0090-RESH-022-01WGsimultaneously, such as combining the rapid onset and precise control of electrical modulation with the specificity and potentially longer-lasting effects of chemical or optogenetic approaches. Non-electrical neuroregulators may include chemical agent stimulators that deliver neurotransmitters such as acetylcholine or norepinephrine, neuromodulators such as GABA or glutamate, or pharmaceutical compounds such as local anesthetics or anti-inflammatory agents directly to neural tissue through implanted pumps or reservoir systems, chemical conduction blocks that use local anesthetics such as lidocaine or bupivacaine or neurotoxins such as botulinum toxin to temporarily or permanently disable neural conduction through biochemical mechanisms, physical ablation devices that create controlled lesions using radiofrequency energy, laser energy, or cryogenic techniques to interrupt neural pathways, and optogenetic modulators that use genetically modified neurons responsive to specific wavelengths of light for precise temporal and spatial control of neural activity. The optogenetic modulators may be configured to deliver light-activated neural modulation using inhibitory or excitatory opsins activated with high frequency light to achieve millisecond-precision control of neural activity. Light delivery systems may include fiber optic implants for precise light delivery to specific neural targets, LED arrays for broader illumination of neural tissue regions, laser diodes for high-intensity focused light delivery with precise spatial control, or wireless optogenetic devices that can be activated remotely through external light sources or magnetic field activation of implanted light sources. The light wavelengths may be specifically tuned to activate particular opsins, with common wavelengths including blue light (450-490 nm) for channelrhodopsin activation to achieve neural excitation, green light (520-570 nm) for certain inhibitory opsins that provide neural silencing, yellow light (570-590 nm) for halorhodopsin activation to achieve neural inhibition, or red light (620-750 nm)Attorney Docket No. 40759.0090-RESH-022-01WGfor deeper tissue penetration and activation of red-shifted opsins depending on the opsin type and desired neural response. Advanced optogenetic systems may employ multiple wavelengths simultaneously to achieve complex modulation patterns such as simultaneous excitation and inhibition of different neural populations, or may use pulsed light delivery to create temporal patterns of neural activation or inhibition that mimic natural neural firing patterns or achieve specific therapeutic effects.

[0033] The bio-feedback sensor is configured to monitor physiological parameters related to the medical condition being treated, providing real-time information about the patient's physiological state and treatment response. Multiple sensor types may be employed simultaneously to provide comprehensive physiological monitoring, including biochemical sensors for measuring glucose, lactate, neurotransmitters, or hormones that reflect metabolic or neurochemical status, mechanical sensors for detecting pressure, movement, or tissue deformation that indicate physical function or pathological changes, electrical sensors for monitoring neural activity or cardiac function through bioelectrical signal detection, optical sensors for measuring oxygen saturation or blood flow using light absorption or scattering techniques, or thermal sensors for detecting temperature changes or metabolic activity through heat production or dissipation. In some embodiments, the bio-feedback sensor comprises an implantable glucose sensor configured to monitor blood glucose levels continuously for diabetes management applications. Alternative glucose monitoring approaches may include continuous glucose monitors that provide real-time glucose readings through subcutaneous sensors, flash glucose monitors that require periodic scanning to obtain glucose data, or non-invasive glucose sensors using optical techniques such as near-infrared spectroscopy or electromagnetic techniques such as impedance measurement thatAttorney Docket No. 40759.0090-RESH-022-01WGcan measure glucose levels through the skin without requiring implantation or tissue penetration. The glucose sensor may have a temporal resolution ranging from about 1 second to about 10 minutes, allowing for real-time or near real-time monitoring of glucose fluctuations and rapid detection of dangerous glucose excursions. Higher temporal resolution sensors with sub-second sampling rates may be employed for critical applications such as detecting rapid glucose changes during hypoglycemic episodes that require immediate intervention, while lower resolution sensors with sampling intervals up to 30 minutes may be sufficient for certain monitoring scenarios such as long-term trend analysis or routine diabetes management where rapid changes are less critical. The sensor may be configured to communicate directly with the neuroregulators in response to detecting a medical emergency condition, such as a hypoglycemic state, enabling immediate therapeutic intervention without requiring external device communication or patient awareness. Emergency response protocols may include automatic activation of specific neuromodulation patterns designed to rapidly increase glucose levels through hepatic glucose release or reduced glucose uptake, alert notifications to healthcare providers or caregivers through wireless communication systems, or patient warning systems such as audible alarms or vibrating alerts to prompt immediate action such as consuming glucose or seeking medical attention to prevent serious complications.

[0034] The programming device may comprise various forms including a smart phone with specialized medical applications, handheld device designed specifically for medical device programming, tablet computer with enhanced processing capabilities, laptop computer for clinical use. desktop computer for comprehensive data analysis, or other portable or stationary computing device depending on the specific application requirements and user preferences. SpecializedAttorney Docket No. 40759.0090-RESH-022-01WGmedical programming devices may also be employed for clinical settings or enhanced security requirements, featuring dedicated hardware for medical device communication with enhanced reliability, enhanced encryption capabilities for protecting sensitive patient data, or specialized user interfaces designed for healthcare professionals with clinical workflow integration. The device includes an artificial intelligence and machine learning engine configured to process sensor data and generate therapy recommendations through sophisticated data analysis and pattern recognition. The Al and ML algorithms may employ various approaches including neural networks for pattern recognition and complex data analysis, deep learning for processing large datasets and identifying subtle patterns, reinforcement learning for adaptive optimization based on patient responses and outcomes, support vector machines for classification tasks such as identifying physiological states, decision trees for rule-based decisions that can be easily interpreted by clinicians, or ensemble methods that combine multiple algorithms for enhanced performance and robustness across diverse patient populations and clinical scenarios. The Al system may implement advanced deep learning architectures including residual neural networks (ResNets) for processing complex physiological signals with skip connections that enable training of very deep networks, attention mechanisms for focusing on critical features in physiological data streams, capsule networks for hierarchical feature learning in multi-dimensional physiological data, or neural architecture search (NAS) techniques for automatically discovering optimal network architectures for specific physiological monitoring tasks. Specialized Al approaches may include time-series forecasting models such as ARIMA (AutoRegressive Integrated Moving Average) for predicting physiological trends, Prophet algorithms for handling seasonal patterns in physiological data, or state-space models for tracking dynamic physiological states over time. The ML system mayAttorney Docket No. 40759.0090-RESH-022-01WGemploy clustering algorithms such as k-means clustering for patient stratification, hierarchical clustering for identifying physiological pattern hierarchies, or density-based clustering (DBSCAN) for detecting anomalous physiological states. Dimensionality reduction techniques may include principal component analysis (PCA) for identifying key physiological variables, independent component analysis (ICA) for separating mixed physiological signals, or t-distributed stochastic neighbor embedding (t-SNE) for visualizing complex physiological relationships. Advanced optimization algorithms may include gradient boosting methods such as XGBoost or LightGBM for ensemble learning with physiological data, random forest algorithms for robust classification of physiological states, or adaptive boosting (AdaBoost) for iteratively improving weak learners in physiological pattern recognition. The Al system may incorporate meta-leaming approaches that enable rapid adaptation to new patients or conditions, few-shot learning techniques for handling rare physiological conditions with limited training data, or continual learning methods that prevent catastrophic forgetting when adapting to new physiological patterns. Probabilistic machine learning approaches may include Gaussian processes for uncertainty quantification in physiological predictions, Bayesian optimization for hyperparameter tuning in physiological models, or variational inference for approximate Bayesian learning in complex physiological systems. The Al and ML algorithm operates on a time scale ranging from about 1 second to about 10 minutes to continuously learn and optimize the neuromodulation parameters based on real-time physiological feedback. Alternative time scales may include real-time processing with millisecond response times for emergency situations such as severe hypoglycemia or cardiac arrhythmias that require immediate intervention, or longer-term learning cycles spanning hours, days, or weeks for adaptive therapy optimization based on long-term patient response patterns and treatmentAttorney Docket No. 40759.0090-RESH-022-01WGoutcomes. The algorithm is configured to adjust therapy parameters based on physiological changes over time and can adapt to body adaptation to the neuromodulation, preventing tolerance and maintaining therapeutic effectiveness. Advanced algorithms may incorporate predictive modeling to anticipate physiological changes before they occur based on historical patterns and current trends, pattern recognition to identify optimal treatment timing based on circadian rhythms or patient activities, anomaly detection to identify unusual physiological states that may require modified treatment approaches, or personalized medicine approaches that customize treatment protocols based on individual patient characteristics, genetic factors, or response history to enhance treatment effectiveness while minimizing side effects. The Al system may employ online learning algorithms that continuously update models based on streaming physiological data, active learning approaches that intelligently select the most informative data points for model improvement, or multi-task learning frameworks that simultaneously optimize multiple therapeutic objectives such as efficacy, safety, and patient comfort.

[0035] Communication between system components may be achieved through various wireless technologies including Bluetooth technology for short-range, low-power communication between nearby devices, radio frequency communication for longer-range or higher-power applications that require greater transmission distances, WiFi communication for high-bandwidth data transfer and integration with internet-based systems, and sound communication using ultrasonic or audible frequencies for specialized applications where electromagnetic interference must be avoided. Additional communication methods may include near-field communication (NFC) for secure, close-proximity data exchange during clinical programming, infrared communication for line-of-sight applications in controlled environments, cellular communicationAttorney Docket No. 40759.0090-RESH-022-01WGfor remote monitoring and data transmission to healthcare providers, satellite communication for areas with limited terrestrial coverage such as rural or remote locations, or proprietary wireless protocols specifically designed for medical devices with enhanced security and reliability features optimized for healthcare applications. Wired communication options may include USB connections for direct device programming and data transfer, serial interfaces for clinical equipment integration with existing hospital systems, or specialized medical device connectors for secure, reliable connections during clinical programming or data download procedures that ensure data integrity and prevent accidental disconnection. The programming device may be configured to receive manual input data comprising food intake information including meal timing, carbohydrate content, and portion sizes that affect glucose metabolism, insulin injection information including dosage, timing, and injection site that influences glucose control, and physical activity information including exercise type, duration, and intensity that affects metabolic demands, which can be incorporated into the Al and ML decision-making process to provide more comprehensive treatment optimization. Additional input data may include medication schedules for other diabetes medications or supplements that may interact with neuromodulation therapy, sleep patterns that may affect glucose metabolism and treatment effectiveness, stress levels that can impact physiological responses and treatment outcomes, environmental factors such as temperature or altitude that may influence treatment effectiveness, or other lifestyle parameters that may influence treatment outcomes such as work schedules, travel, or illness. Voice recognition systems may allow patients to verbally input information for convenient data entry, gesture control interfaces may provide hands-free operation for patients with mobility limitations, or automated data import from other health monitoring devices such as fitness trackers, smart scales, or bloodAttorney Docket No. 40759.0090-RESH-022-01WGpressure monitors may streamline data entry and improve patient compliance with monitoring protocols while reducing the burden of manual data entry.

[0036] The system may further include a clinic communication device configured to communicate with the programming device for clinical oversight and treatment optimization, and a central computer system configured to receive data from the clinic communication device and generate population-based treatment recommendations using artificial intelligence and machine learning algorithms that leverage data from multiple patients to improve treatment protocols. The clinic communication device may include specialized medical workstations with enhanced security features for protecting patient data, secure communication terminals that comply with healthcare data protection requirements such as HIPAA, or integrated electronic health record systems that can seamlessly incorporate neuromodulation data into comprehensive patient records for holistic patient care. The central computer system may comprise a HIPAA compliant system with blockchain protocol to ensure data security and privacy while enabling population-level learning and optimization through secure, distributed data sharing. The central Al system may implement advanced distributed learning architectures including federated learning frameworks that enable collaborative model training across multiple clinical sites without sharing raw patient data, differential privacy mechanisms that add carefully calibrated noise to protect individual patient privacy while preserving population-level insights, or secure multi-party computation protocols that allow joint analysis of patient data from multiple institutions without revealing individual patient information. The population-based Al system may employ large-scale machine learning techniques including distributed gradient descent algorithms for training models across massive patient datasets, MapReduce frameworks for parallel processing of population-levelAttorney Docket No. 40759.0090-RESH-022-01WGphysiological data, or Apache Spark clusters for real-time processing of streaming physiological data from multiple patients simultaneously. Advanced Al architectures for population learning may include hierarchical Bayesian models that capture both individual patient characteristics and population-level trends, mixture models for identifying distinct patient subpopulations with different treatment responses, or multi-level modeling approaches that account for variability at patient, clinic, and population levels. The central system may implement knowledge distillation techniques for transferring insights from large population models to smaller patient-specific models, meta-learning algorithms that enable rapid adaptation to new patient populations or rare conditions, or continual learning frameworks that prevent catastrophic forgetting when incorporating new patient data or treatment protocols. Population-based optimization may employ genetic programming for evolving treatment protocols based on population outcomes, swarm intelligence algorithms for distributed optimization across patient populations, or evolutionary strategies for robust treatment protocol development that accounts for population variability. Additional security measures may include end-to-end encryption to protect data during transmission between all system components, multi-factor authentication to ensure authorized access by healthcare providers and patients, secure key management systems to protect encryption keys and prevent unauthorized access, or advanced cybersecurity protocols specifically designed for medical data protection including intrusion detection, threat monitoring, and automated security updates to protect against evolving cyber threats. The population-based learning may incorporate federated learning approaches that allow knowledge sharing without exposing individual patient data to privacy risks, differential privacy techniques that add statistical noise to protect individual privacy while preserving population trends for research and treatmentAttorney Docket No. 40759.0090-RESH-022-01WGoptimization, or other advanced methods to enable knowledge sharing while maintaining individual patient privacy and complying with healthcare data protection regulations. Cloud computing platforms may provide scalable processing power for large datasets and complex Al algorithms, edge computing may enable real-time processing at local sites with minimal latency for time-critical applications, or hybrid computing architectures may optimize system performance by balancing centralized and distributed processing capabilities based on specific computational requirements and data sensitivity considerations.

[0037] Power management for the implantable components may be achieved through rechargeable batteries within the neuroregulators that provide reliable, long-term power for therapeutic delivery. Alternative power sources may include primary batteries for long-term, maintenance-free operation in applications where recharging is impractical, supercapacitors for rapid charging and high power delivery during intensive therapy periods, energy harvesting systems that capture energy from body movement through kinetic energy conversion, temperature differences through thermoelectric generation, or electromagnetic fields through inductive coupling, or hybrid power solutions combining multiple energy sources to optimize reliability and longevity while ensuring continuous therapeutic delivery. An external charger may be configured to wirelessly recharge the rechargeable batteries through RF power transmission using electromagnetic coupling between external and internal components. Alternative charging methods may include inductive charging using magnetic coupling between external and internal coils for efficient power transfer, magnetic resonance charging that allows greater spatial freedom during charging and reduces alignment requirements, ultrasonic power transfer that uses focused sound waves to transmit energy through tissue without electromagnetic interference, or opticalAttorney Docket No. 40759.0090-RESH-022-01WGpower transmission using focused light beams for specialized applications where electromagnetic methods are not suitable. The external charger may comprise a transmit coil while the neuroregulators each comprise a receiving antenna to facilitate wireless power transfer with optimal coupling efficiency. Advanced charging systems may include multiple transmit coils for improved coupling efficiency and reduced sensitivity to coil alignment during patient movement, adaptive charging algorithms that optimize power transfer based on implant position and orientation to maximize charging efficiency, or smart charging protocols that monitor battery health and adjust charging parameters accordingly to maximize battery life and prevent overcharging or thermal damage. Power management algorithms may optimize energy consumption by adjusting neuromodulation parameters based on remaining battery capacity to extend operation time, implementing sleep modes during inactive periods to conserve power when therapy is not needed, or prioritizing critical functions during low battery conditions to ensure essential therapeutic functions continue even when power is limited, such as maintaining emergency response capabilities or basic monitoring functions.

[0038] Medical Conditions that can be treated through the cooperative neuromodulation of two or more neuronal targets include, but are not limited to: (a) type II diabetes involving dysregulated glucose homeostasis; (b) obesity involving complex metabolic and appetite regulation pathways; (c) refractory hypertension involving multiple cardiovascular control mechanisms; (d) pancreatitis involving inflammatory and digestive processes; (e) refractory asthma involving complex respiratory and inflammatory pathways; (f) urinary incontinence involving bladder and sphincter control mechanisms; (g) erectile dysfunction involving vascular and neural control of sexual function; (h) chronic heart failure involving cardiac and circulatoryAttorney Docket No. 40759.0090-RESH-022-01WGregulation; (i) chronic inflammatory disease involving immune system modulation; j) Parkinson's disease involving motor control and neurotransmitter pathways; (k) refractory depression involving complex mood regulation circuits; (1) refractory epilepsy involving seizure control mechanisms; and (m) migraine and cluster headaches involving pain and vascular control pathways. Additional medical conditions may include chronic pain syndromes such as neuropathic pain or complex regional pain syndrome that involve multiple pain processing pathways, gastroparesis involving delayed gastric emptying and complex gut-brain interactions, irritable bowel syndrome with its complex gut-brain interactions and visceral hypersensitivity, sleep disorders including sleep apnea or insomnia that involve multiple physiological control systems, anxiety disorders that may benefit from autonomic modulation and stress response regulation, post-traumatic stress disorder with its associated autonomic dysfunction and hyperarousal symptoms, addiction disorders where neural reward pathways may be targeted to reduce craving and relapse, eating disorders involving dysregulated appetite control and body weight regulation, chronic fatigue syndrome with its complex neurological components and autonomic dysfunction, fibromyalgia with widespread pain and autonomic dysfunction affecting multiple body systems, or other neurological, psychiatric, or metabolic conditions that may benefit from multi-target neuromodulation approaches that address the complex, interconnected nature of these disorders. The treatment protocols may be customized for specific disease subtypes based on pathophysiology and symptom patterns, patient populations based on age, gender, or genetic factors that influence treatment response, or comorbid conditions to optimize therapeutic outcomes while minimizing potential side effects or interactions with other treatments such as medications or other medical devices.Attorney Docket No. 40759.0090-RESH-022-01WG

[0039] For the treatment of type 2 diabetes, a specific embodiment involves a first electrical neuroregulator configured to apply a high frequency alternating current conduction block to a hepatic nerve or an anterior vagus nerve cranial to a branching point of the hepatic nerve to reduce hepatic glucose production and improve insulin sensitivity, and a second electrical neuroregulator configured to apply a low frequency electrical stimulation to a celiac nerve or a posterior vagus nerve cranial to a branching point of the celiac nerve to enhance insulin secretion and improve glucose metabolism. Alternative diabetes treatment configurations may include modulation of pancreatic nerves to directly influence insulin and glucagon secretion from pancreatic islets, gastric nerves to control gastric emptying and nutrient absorption rates that affect postprandial glucose excursions, intestinal nerves to modulate incretin hormone release such as GLP-1 and GIP that regulate glucose homeostasis, or other components of the enteric nervous system that play roles in glucose homeostasis and metabolic regulation. Additional targets may include sympathetic nerves innervating the liver to control hepatic glucose production and glycogen metabolism, pancreas to influence islet cell function and hormone secretion, or adipose tissue to modulate insulin sensitivity and glucose uptake, or parasympathetic nerves controlling insulin secretion or glucose metabolism in various target organs such as muscle, liver, or adipose tissue. This configuration, combined with glucose monitoring and AI / ML optimization, provides a comprehensive approach to glycemic control that addresses multiple aspects of glucose homeostasis simultaneously. The Al system for diabetes management may implement specialized algorithms including glucose prediction models using recurrent neural networks trained on continuous glucose monitoring data, meal detection algorithms using convolutional neural networks to analyze glucose patterns and automatically identify food intake events, insulin sensitivity estimation using Kalman filtering or particleAttorney Docket No. 40759.0090-RESH-022-01WGfiltering approaches, or circadian rhythm modeling using harmonic regression or Fourier analysis to optimize treatment timing based on natural metabolic cycles. Advanced diabetes Al may employ reinforcement learning algorithms specifically designed for glucose control, including model predictive control (MPC) frameworks for anticipating glucose changes and proactively adjusting neuromodulation, deep Q-networks (DQN) for learning optimal neuromodulation policies from glucose response data, or actor-critic methods for balancing glucose control objectives with safety constraints. The ML system may incorporate ensemble methods that combine multiple glucose prediction models for enhanced accuracy, anomaly detection algorithms for identifying unusual glucose patterns that may indicate illness or medication changes, or transfer learning approaches that leverage population data to improve individual patient models. Advanced diabetes management may incorporate continuous glucose monitoring with real-time data streaming for immediate feedback, insulin pump integration for coordinated therapy delivery that combines neuromodulation with pharmacological intervention, meal detection algorithms that can automatically adjust neuromodulation based on food intake patterns and carbohydrate content, or predictive glucose modeling that anticipates glucose changes and proactively adjusts treatment to achieve optimal glycemic outcomes while minimizing hypoglycemic episodes that can be dangerous or life-threatening, particularly in patients with hypoglycemia unawareness or other diabetes complications.

[0040] The neuromodulation system may be configured as a closed-loop system comprising a plurality of neuroregulators, each configured to neuromodulate a different neuronal target selected from sympathetic nerves, parasympathetic nerves, central nervous system tissue, and smooth muscle to provide comprehensive physiological control. The closed-loop configuration enablesAttorney Docket No. 40759.0090-RESH-022-01WGreal-time adaptation of therapy parameters based on physiological feedback, creating a dynamic treatment system that responds to changing patient conditions throughout the day and over longer time periods such as disease progression or treatment adaptation. The system includes at least one bio-feedback sensor configured to monitor a physiological state, and a programming device comprising an artificial intelligence and machine learning algorithm configured to analyze data from the bio-feedback sensor and generate predictive recommendations for neuromodulation parameters based on current physiological status and anticipated needs. The closed-loop Al system may implement advanced control algorithms including model predictive control (MPC) for anticipating physiological changes and proactively adjusting therapy, adaptive control methods that automatically tune control parameters based on patient response, or robust control techniques that maintain performance despite physiological variability and uncertainty. Machine learning approaches for closed-loop control may include reinforcement learning algorithms such as temporal difference learning for optimizing long-term therapeutic outcomes, policy gradient methods for continuous parameter optimization, or deep reinforcement learning approaches that can handle high-dimensional physiological state spaces. The Al system may employ Bayesian filtering techniques such as extended Kalman filters for state estimation in physiological systems, particle filters for nonlinear physiological state tracking, or unscented Kalman filters for handling uncertainty in physiological models. Advanced closed-loop systems may incorporate multiple feedback loops operating at different time scales such as rapid responses for emergency situations and slower adaptations for long-term optimization, hierarchical control structures that prioritize different therapeutic objectives such as safety over efficacy or immediate needs over long-term goals, or adaptive control algorithms that learn from patient responses and automatically optimizeAttorney Docket No. 40759.0090-RESH-022-01WGtreatment protocols based on individual patient characteristics and response patterns over time. Fail-safe mechanisms may include redundant sensors to ensure continued monitoring if one sensor fails or becomes inaccurate, backup control systems that can maintain basic therapeutic functions during system malfunctions or communication failures, or emergency shutdown protocols that safely disable the system if dangerous conditions are detected to ensure patient safety and system reliability under all operating conditions including device failures or unexpected physiological responses.

[0041] In alternative embodiments, the system may function as a multi-target neuromodulation apparatus comprising first and second neuromodulation components selected from electrical stimulators, electrical conduction blocks, chemical agent stimulators, chemical conduction blocks, physical ablation devices, and optogenetic modulators to provide diverse therapeutic modalities. Each component is configured to modulate different targets selected from sympathetic nerves, parasympathetic nerves, central nervous system tissue, and smooth muscle to address multiple aspects of complex medical conditions. The multi-target approach may be expanded to include three, four, or more neuromodulation components for complex medical conditions requiring comprehensive neural pathway modulation, such as conditions involving multiple organ systems or complex neurological disorders that cannot be adequately addressed through single or dualtarget approaches. A sensor system provides bio-feedback regarding the medical condition, while an artificial intelligence and machine learning processing unit analyzes the bio-feedback and determines optimal modulation parameters through sophisticated data analysis and predictive modeling. The sensor system may include multiple sensor types such as biochemical sensors for measuring metabolites or neurotransmitters, mechanical sensors for detecting physical changes orAttorney Docket No. 40759.0090-RESH-022-01WGmovement, electrical sensors for monitoring bioelectrical activity, and optical sensors for measuring blood flow or oxygenation, sensor arrays that provide spatial information about physiological processes across different body regions, or distributed sensing networks that monitor multiple body locations simultaneously to provide comprehensive physiological monitoring and early detection of adverse events or treatment responses. The Al processing unit for multi-target systems may implement specialized coordination algorithms including multi-objective optimization for balancing competing therapeutic goals, game-theoretic approaches for optimizing interactions between different neuromodulation components, or distributed consensus algorithms for coordinating multiple autonomous neuromodulation units. Machine learning approaches may include supervised learning using labeled training data from successful treatments to identify optimal treatment patterns, unsupervised learning to identify hidden patterns in physiological data that may not be apparent to clinicians, reinforcement learning that learns optimal actions through trial and reward mechanisms that adapt to individual patient responses, or hybrid approaches that combine multiple Al techniques for enhanced performance and robustness across diverse patient populations and clinical scenarios. The Al system may employ multi-agent learning frameworks where different Al agents control different neuromodulation components and learn to cooperate for optimal therapeutic outcomes, hierarchical reinforcement learning for managing complex multi-target treatment protocols, or curriculum learning approaches that gradually increase treatment complexity as the system learns optimal coordination strategies.

[0042] The control system coordinates the neuromodulation components based on the optimal modulation parameters, ensuring that the modulation of multiple targets works cooperatively to treat the medical condition while avoiding interference or counterproductive interactions.Attorney Docket No. 40759.0090-RESH-022-01WGAdvanced coordination algorithms may include synchronization protocols that ensure precise timing between different neuromodulation components to maximize therapeutic synergy, interference mitigation strategies that prevent unwanted interactions between different modulation signals that could reduce efficacy or cause adverse effects, or optimization algorithms that balance competing therapeutic objectives such as maximizing efficacy while minimizing side effects or energy consumption. The Al-driven control system may implement advanced coordination strategies including distributed optimization algorithms that allow multiple neuromodulation components to optimize their parameters collaboratively, consensus-based control methods that ensure all components work toward common therapeutic objectives, or hierarchical control architectures that manage coordination at multiple levels from individual component control to system-wide optimization. Machine learning approaches for coordination may include multi-task learning frameworks that simultaneously optimize multiple therapeutic objectives, meta-learning algorithms that quickly adapt coordination strategies to new patients or conditions, or continual learning methods that prevent interference between different coordination strategies as the system adapts to changing patient needs. The system may be configured to coordinate the neuromodulation components with different start times, different end times, and different durations, or during overlapping time periods as clinically appropriate for the specific medical condition and patient characteristics. Complex coordination patterns may include sequential activation where different targets are activated in a specific order to achieve cascading physiological effects, alternating patterns that switch between different targets to prevent adaptation or tolerance, synchronized bursts that coordinate multiple targets simultaneously for enhanced therapeutic impact, or adaptive timing based on real-time physiological feedback thatAttorney Docket No. 40759.0090-RESH-022-01WGadjusts coordination patterns based on patient response and changing physiological conditions. The control system may also incorporate safety monitoring that continuously checks for adverse effects or unexpected physiological responses, fault detection that identifies system malfunctions or component failures, or emergency shutdown capabilities that can safely disable the system if dangerous conditions are detected to ensure patient safety during all operating conditions including device malfunctions, unexpected physiological responses, or emergency medical situations.

[0043] The disclosure herein is directed to the neuromodulation (stimulation or conduction block) of two or more neuronal targets in treatment of a medical condition. The neuromodulation system also contains a bio-feedback sensor which gives information to an external Al and ML engine to optimize the neuromodulation signals. In some embodiments the bio-feedback sensor would be an implantable glucose sensor. The neuronal targets generally comprise sympathetic and / or parasympathetic nerves, which typically innervate an internal organ. One or more neuroregulators provide the desired stimulation or conduction block to the neuronal targets, which work in cooperation to treat a medical condition. In certain embodiments the two or more neuronal targets are: (a) neuromodulated over a common time period; (b) neuromodulated during overlapping time periods; (c) neuromodulated in distinct non-overlapping time periods; and / or (d) neuromodulated in any combination of (a), (b), and (c). The neuromodulation of each of the neuronal targets can include the same or different start / end times, the same or different durations, and / or the same or different neuromodulation patterns.

[0044] Stimulation neuroregulators can comprise, for example, electrical stimulators or chemical agent stimulators. Conduction block neuroregulators can comprise, for example,Attorney Docket No. 40759.0090-RESH-022-01WGphysical section (e.g. ablation) conduction blocks, electrical conduction blocks, electrical high frequency conduction blocks, chemical conduction blocks or an optogenetic conduction blocks obtained by delivery of an inhibitory opsin or excitatory opsin that is flashed with high frequency light. Of the noted neuroregulators, electrical stimulators, electrical conduction blocks and electrical high frequency conduction blocks can be deemed electrical neuroregulators while all other noted neuroregulators can be deemed non-electrical neuroregulators.

[0045] Medical Conditions that can be treated the cooperative neuromodulation of two or more neuronal targets include, but are not limited to: (a) type II diabetes;; (b) obesity; (c) refractory hypertension; (d) pancreatitis; (e) refractory asthma; (f) urinary incontinence; (g) erectile dysfunction; (h)chronic heart failure; (i)chronic inflammatory disease; (j) Parkinson’s disease; (k) refractory depression; (1) refectory epilepsy; and (m) migraine and cluster headaches.

[0046] Figure 1 illustrates a simplified schematic of a neuromodulation system wherein a single neuroregulator is used to neuromodulate a plurality of neuronal targets through multiple output channels or electrode arrays. This configuration may be advantageous for applications requiring synchronized modulation of multiple targets or when space constraints limit the number of implantable components, such as in pediatric applications or when targeting closely spaced neural structures. Figure 1 illustrates a simplified schematic of a neuromodulation system 10 wherein a single neuroregulator 12 is used to neuromodulate a plurality of neuronal targets 14a, 14b...14(n).

[0047] Figure 2 illustrates a simplified schematic of a neuromodulation system wherein each neuronal target is neuromodulated by a corresponding neuroregulator, providing independentAttorney Docket No. 40759.0090-RESH-022-01WGcontrol and optimization for each target. This distributed approach allows for independent control of each target and may provide enhanced flexibility in treatment protocols, enabling customized therapy parameters for each target based on its specific characteristics and therapeutic requirements. Other combinations of neuroregulators and neuronal targets within a neuromodulation system are also possible, including hybrid configurations that combine single and multiple neuroregulator approaches for optimal therapeutic flexibility, modular systems that can be expanded as needed by adding additional neuroregulators or sensors based on changing patient needs or disease progression, or adaptive systems that can reconfigure based on treatment requirements or changing patient conditions such as disease progression or treatment response. The neuromodulation of a neuronal target operates in cooperation with the neuromodulation of the other of the two or more neuronal targets to treat a medical condition through coordinated physiological effects. The cooperative interaction may involve synergistic effects where the combined effect is greater than the sum of individual effects due to physiological amplification or pathway convergence, complementary mechanisms that address different aspects of the same condition such as different symptoms or pathophysiological processes, or coordinated physiological responses that enhance overall therapeutic efficacy while minimizing compensatory responses that might reduce treatment effectiveness or cause adverse effects.

[0048] Still referring to Figure 2 which illustrates a simplified schematic of a neuromodulation system 10 wherein each neuronal target 14a, 14b...14(n) is neuromodulated by a corresponding neuroregulator 12a, 12b...12(n). Other combinations of neuroregulators and neuronal targets within a neuromodulation system are also possible. The neuromodulation of a neuronal targetAttorney Docket No. 40759.0090-RESH-022-01WGoperates in cooperation with the neuromodulation of the other of the two or more neuronal targets to treat a medical condition.

[0049] Figure 3 provides a schematic of an exemplary neuromodulation system having a first neuroregulator to neuromodulate a first neuronal target and a second neuroregulator to neuromodulate a second neuronal target in a coordinated therapeutic approach. In this example, each of the first and second neuroregulators comprise electrical neuroregulators that can provide electrical stimulation and / or electrical conduction blocking with programmable parameters for optimal therapeutic delivery. The electrical neuroregulators may include programmable pulse generators that can create complex waveforms for specialized therapeutic applications, constant current sources that maintain stable current delivery regardless of tissue impedance changes over time, constant voltage sources for applications requiring stable voltage delivery such as certain blocking protocols, or adaptive stimulation devices that automatically adjust parameters based on tissue impedance or physiological feedback to maintain optimal therapeutic delivery despite changing tissue conditions. In other examples, the first neuroregulator of the neuromodulation system may comprise an electrical neuroregulator while the second neuroregulator of the neuromodulation system may comprise a non-electrical neuroregulator such as a chemical delivery system or optogenetic modulator. This hybrid approach may combine the precision and controllability of electrical modulation with the specificity and potentially longer-lasting effects of chemical or optogenetic approaches, providing enhanced therapeutic options for complex medical conditions that may benefit from different types of neural modulation simultaneously, such as combining rapid electrical effects with sustained chemical modulation.Attorney Docket No. 40759.0090-RESH-022-01WG

[0050] In addition to an output therapy system there is also an input sensor system (such as a glucose sensor) which has communication circuitry which can communicate with communication circuitry of one or more rechargeable neuroregulators for real-time therapy optimization. The sensor system may include multiple sensor types for comprehensive monitoring of different physiological parameters, redundant sensors for reliability and fail-safe operation to ensure continued monitoring if individual sensors fail, or sensor arrays for comprehensive monitoring of physiological parameters across different body locations or organ systems to provide spatial information about treatment effects. The sensor can also communicate with an external programming device's communication circuitry for data analysis and therapy optimization. The programming device can also receive input from a handheld monitor (such as a glucose monitor) for additional data sources and patient involvement in treatment monitoring. Additional input sources may include wearable devices such as fitness trackers or smartwatches that monitor activity and physiological parameters, smartphone apps that allow patient input of symptoms or lifestyle factors such as meals or stress levels, electronic health records that provide comprehensive medical history and medication information, or other medical monitoring equipment such as blood pressure monitors or heart rate monitors to provide comprehensive patient data for more informed treatment decisions and holistic patient care. The programming device has an Al and ML engine and its communication circuitry can then communicate with the neuroregulator(s) communication circuitry to optimize therapy based on comprehensive data analysis and predictive modeling. The Al and ML engine may employ cloud-based processing for access to powerful computational resources and large datasets for population-based learning, edge computing for real-time processing with minimal latency for time-critical applications, or hybrid approaches that balanceAttorney Docket No. 40759.0090-RESH-022-01WGcomputational power with real-time response requirements by performing time-critical processing locally while using cloud resources for more complex analysis and learning from population data. The Al engine may implement distributed computing architectures including microservices for modular Al functionality, containerized machine learning models for scalable deployment, or serverless computing frameworks for efficient resource utilization. Advanced Al processing may employ specialized hardware accelerators including graphics processing units (GPUs) for parallel machine learning computations, tensor processing units (TPUs) for optimized neural network inference, or field-programmable gate arrays (FPGAs) for custom Al acceleration tailored to specific physiological signal processing requirements.

[0051] The programming device can also communicate with a clinic's communication device for clinical oversight and professional medical management. Data from the clinic can then be uploaded to an Al and ML algorithm engine for population-based learning and treatment optimization. The Al and ML algorithm engine can, for example, be on a central computer system / repository / on a cloud or server based information system that provides scalable processing power and data storage. The central system may employ distributed computing across multiple servers for scalability and redundancy, parallel processing to handle large datasets efficiently and reduce processing time, or specialized Al hardware such as graphics processing units or tensor processing units to handle large-scale data analysis and population-level learning with enhanced speed and efficiency for complex machine learning algorithms. The central Al system may implement advanced distributed learning frameworks including federated learning architectures that enable collaborative model training across multiple clinical sites without sharing raw patient data, parameter server architectures for distributed machine learning across multiple computingAttorney Docket No. 40759.0090-RESH-022-01WGnodes, or blockchain-based learning systems that ensure data integrity and traceability in population-based Al training. Advanced population-based Al may employ large-scale optimization techniques including distributed stochastic gradient descent for training models across massive patient datasets, asynchronous parallel algorithms for efficient distributed learning, or consensus-based optimization methods that ensure convergence in distributed learning environments. The central system may implement advanced data processing pipelines including Apache Kafka for real-time streaming of physiological data, Apache Spark for large-scale batch processing of population data, or Hadoop ecosystems for distributed storage and processing of massive healthcare datasets. In turn the Al and ML algorithm engine on the central computer system / repository / on a cloud / server can then send recommendations, based on data from a population of patients, of optimal therapy to the clinic which can then communicate with the programming device to instruct the neuroregulator (s) to deliver different therapy parameters for improved treatment outcomes. The population-based recommendations may include treatment protocols for specific patient subgroups based on demographics, disease characteristics, or response patterns that have been identified through population analysis, predictive models for treatment outcomes that can help clinicians make informed decisions about therapy modifications, or adaptive algorithms that improve over time as more patient data becomes available, leading to continuously improving treatment effectiveness across the entire patient population and advancing the field of neuromodulation therapy. The population Al system may employ advanced analytics including survival analysis for predicting long-term treatment outcomes, causal inference methods for identifying true therapeutic effects versus confounding factors, or longitudinal data analysisAttorney Docket No. 40759.0090-RESH-022-01WGtechniques for tracking treatment effectiveness over extended time periods across diverse patient populations.

[0052] In the case of an immediate medical emergency the sensor can communicate directly with the neuroregulator(s) to change the therapy parameters without requiring external device communication or patient intervention. An example of this would be a patient going into a severe hypoglycemic state where immediate intervention is critical. The neuroregulators could then change block or stimulation parameters to quickly increase blood glucose levels through hepatic glucose release or other physiological mechanisms. The emergency response Al system may implement specialized algorithms including anomaly detection models trained to recognize emergency physiological patterns, decision trees for rapid emergency response classification, or rule-based expert systems that encode clinical emergency protocols for immediate automated response. Emergency Al may employ real-time pattern recognition using streaming machine learning algorithms, threshold-based alerting systems with adaptive thresholds that learn from individual patient baselines, or predictive models that can anticipate emergency conditions before they become critical. The emergency response system may implement hierarchical decisionmaking algorithms that escalate response intensity based on severity assessment, multi-criteria decision analysis for balancing competing emergency response objectives, or fuzzy logic systems for handling uncertainty in emergency condition assessment. Emergency response protocols may include predefined parameter sets for different emergency conditions such as severe hypoglycemia, hyperglycemia, or cardiac arrhythmias that have been validated for safety and efficacy, escalating intervention strategies that increase intervention intensity if initial responses are insufficient to resolve the emergency, or automatic notification of healthcare providers orAttorney Docket No. 40759.0090-RESH-022-01WGemergency services to ensure comprehensive emergency response and professional medical evaluation. The system may also include patient alert mechanisms such as audible alarms or vibrating alerts to notify the patient of the emergency situation, caregiver notifications sent to family members or healthcare providers through wireless communication, or integration with emergency medical systems to ensure comprehensive emergency response capabilities that can save lives in critical situations and provide peace of mind for patients and their families.

[0053] The neuromodulation system generally includes an external component that resides outside the body and an internal component that is implanted within the body below the dermis for long-term therapeutic delivery. The external component includes a charger that is coupled, via connector, to a battery and a transmit coil for wireless power and data transmission. The external component may also include user interface elements such as displays for showing system status and therapy information, buttons or touchscreens for user interaction and control, status indicators such as LED lights or displays showing system status and battery levels, communication modules for connecting to smartphones or other devices for enhanced functionality, or additional sensors for comprehensive system monitoring such as temperature sensors or motion detectors that can provide information about system operation and patient activity. The internal component includes the first and second neuroregulators, each of which is coupled by one or more leads to one or more electrodes, which are placed on the neuronal targets for precise therapeutic delivery. The leads may include specialized designs for different anatomical locations such as flexible leads for mobile areas that experience frequent movement, rigid leads for stable placement in fixed anatomical locations, steerable leads for precise placement during surgery that allow surgeons to navigate complex anatomy, or multi-contact leads for selective nerve stimulation that can target specificAttorney Docket No. 40759.0090-RESH-022-01WGnerve fascicles or provide multiple stimulation sites along a single nerve for enhanced therapeutic flexibility. Each of the neuroregulators includes a rechargeable battery and a receiving antenna for wireless power and communication. In certain examples, the neuroregulators share a common battery and a common receiving antenna to reduce system complexity and implant volume. Alternative configurations may include distributed power systems that provide redundancy and enhanced reliability, backup batteries that ensure continued operation during charging or primary battery failure, or energy harvesting capabilities such as kinetic energy harvesting from body movement or thermal energy harvesting from body heat to enhance system reliability and longevity while reducing dependence on external charging.

[0054] The leads may comprise bipolar leads each of which is connected to first and second electrodes for differential stimulation or recording. However, other lead and electrode configurations may be used as appropriate to a specific application, including monopolar leads that use a single electrode with a distant return electrode for broader stimulation fields, tripolar leads that provide additional stimulation flexibility and current steering capabilities, multipolar leads that allow complex stimulation patterns and precise current control, or specialized electrode arrays for complex stimulation patterns such as steering current in specific directions or creating focused stimulation fields that minimize effects on surrounding tissues. Electrode materials may include platinum for excellent biocompatibility and corrosion resistance in the biological environment, platinum-iridium alloys for enhanced mechanical properties and durability, titanium for lightweight applications and excellent biocompatibility, or other biocompatible conductive materials optimized for long-term implantation such as iridium oxide for enhanced charge injection capacity or conductive polymers for improved tissue interface properties. In certainAttorney Docket No. 40759.0090-RESH-022-01WGexamples, each of the neuroregulators comprises an implantable component that is independent from the other while in other examples the first and second neuroregulators are combined within a single implantable component to reduce surgical complexity and implant volume. Modular designs may allow for system expansion by adding additional neuroregulators as treatment needs evolve, component replacement if individual components fail without requiring complete system replacement, or customization based on specific patient needs or changing medical conditions such as disease progression or treatment response changes. In certain examples, each of the neuroregulators is configured to interface with a common external component while in other examples each of the neuroregulators interfaces with their own corresponding external component for independent control and optimization. The interface design may optimize for power efficiency to maximize battery life and reduce charging frequency, communication reliability to ensure consistent data transmission and therapy delivery, or user convenience to simplify patient interaction with the system and improve treatment compliance depending on the specific application requirements and patient characteristics.

[0055] In operation, the first and second neuroregulators of the neuromodulation system produce electrical pulses that are delivered to their respective first and second neuronal targets through the electrically conductive leads and electrodes with precise control over therapeutic parameters. The pulse generation may employ sophisticated waveform synthesis to create complex stimulation patterns optimized for specific therapeutic applications, adaptive impedance matching to maintain consistent stimulation despite tissue changes over time, or real-time parameter adjustment based on tissue response or physiological feedback to optimize therapeutic delivery and maintain efficacy. In addition to delivering the electrical pulses, each of the neuroregulatorsAttorney Docket No. 40759.0090-RESH-022-01WGalso receives wireless command signals from the programming device for therapy optimization and system control. The command signals may include therapy parameters such as stimulation amplitude, frequency, and duration that define the therapeutic intervention, scheduling information for when stimulation should be delivered based on patient needs or circadian rhythms, safety limits to prevent harmful stimulation levels that could cause tissue damage or adverse effects, or diagnostic commands for system monitoring and troubleshooting to ensure proper system operation. Each of the neuroregulators is powered by their internal battery; in certain examples, the neuroregulators share a common internal battery for simplified system design. Power management strategies may include dynamic power allocation that distributes available power based on current therapeutic needs and system demands, sleep modes during inactive periods to conserve battery life when therapy is not required, or adaptive power consumption based on therapy requirements that adjusts power usage to extend battery life while maintaining therapeutic effectiveness and ensuring critical functions remain available. The internal battery is periodically recharged by RF power that is radiated by the transmit coil and picked up by the receiving antenna through electromagnetic coupling. Advanced charging protocols may optimize charging efficiency to minimize charging time and maximize power transfer, monitor battery health to predict when replacement may be needed and prevent unexpected failures, or provide predictive maintenance alerts to ensure the system continues operating reliably and patients can plan for necessary maintenance. The charger provides the electrical excitation of the transmit coil needed to deliver RF power to the neuroregulators with optimal efficiency. In certain examples, a rechargeable battery powers the charger for portable operation and patient convenience. The transmit coil serves to transmit RF power transdermally from the charger to the neuroregulators throughAttorney Docket No. 40759.0090-RESH-022-01WGelectromagnetic coupling. The transmit coil further facilitates bi-directional RF communications between the neuroregulators and the charger for system monitoring and control. Communication protocols may include error correction to ensure accurate data transmission despite electromagnetic interference, encryption to protect patient data and prevent unauthorized access to the therapeutic system, or adaptive transmission power to optimize reliability and security while minimizing power consumption and electromagnetic interference with other devices. The programming device enables a clinician to program each of the neuroregulators with a treatment schedule and with therapy parameters for delivering stimulation or conduction blocking to their respective neuronal targets based on clinical assessment and treatment goals. Advanced programming interfaces may include graphical user interfaces that make programming intuitive and reduce the likelihood of programming errors, automated parameter optimization that suggests optimal settings based on patient data and clinical evidence, or integration with electronic health record systems for comprehensive patient management that incorporates neuromodulation data into the broader medical record for holistic patient care.

[0056] An example of a neuromodulation system similar to the one described above, albeit with only one neuroregulator and a different communication system, is the MAESTRO® Rechargeable System available from ReShape Lifesciences, Inc., 1001 Calle Amanecer, San Clemente, CA 92673, USA. The MAESTRO® Rechargeable System is described in detail in various US Patent Publications which are incorporated by reference in their entirety for their teachings on neuromodulation system design and implementation. Other commercial neuromodulation systems may provide additional examples of implementation approaches, including systems from Medtronic such as their deep brain stimulation systems for neurologicalAttorney Docket No. 40759.0090-RESH-022-01WGdisorders, Boston Scientific such as their spinal cord stimulation systems for pain management, Abbott such as their peripheral nerve stimulation systems for various therapeutic applications, or other medical device manufacturers that have developed various approaches to neural stimulation and monitoring for different medical conditions. These systems may employ different communication protocols such as proprietary wireless protocols optimized for medical applications or standard communication methods adapted for medical use, power management strategies such as primary batteries for long-term operation or wireless charging for patient convenience, or electrode configurations such as paddle leads for broad stimulation areas or cylindrical leads for precise targeting that could be adapted for multi-target neuromodulation applications depending on the specific therapeutic requirements and anatomical considerations.

[0057] Numerous medical conditions can be treated through the neuromodulation of two or more neuronal targets using coordinated therapeutic approaches that address the complex, interconnected nature of physiological systems. Various neuromodulation systems, including those described herein can be used to achieve the desired neuromodulation with appropriate customization for specific applications. The selection of specific neuronal targets, neuromodulation parameters, and system configurations may be optimized based on the particular medical condition and its underlying pathophysiology, patient characteristics such as age, weight, or disease severity that may influence treatment response, or treatment objectives such as symptom relief, disease modification, or quality of life improvement that define therapeutic success. Table 1 below provides examples of at least first and second neuronal targets that can be neuromodulated (stimulated (Stim) or conduction blocked (Block)) for treatment of a medical condition, demonstrating the versatility and broad applicability of multi-target neuromodulation approaches.Attorney Docket No. 40759.0090-RESH-022-01WGAdditional target combinations may be developed based on advancing understanding of neural pathways and their roles in disease processes, alternative neuromodulation approaches such as new stimulation waveforms or blocking techniques that may offer enhanced therapeutic benefits, or novel therapeutic applications as new medical conditions are identified that may benefit from multi-target neuromodulation as the field advances and new clinical evidence becomes available through ongoing research and clinical trials that continue to expand the therapeutic potential of these innovative approaches.

[0058] The neuromodulation system may be used in conjunction with conventional treatments to provide comprehensive therapeutic management that addresses multiple aspects of complex medical conditions through integrated care approaches that leverage the strengths of both neuromodulation technology and established medical interventions. This combination approach may enhance therapeutic outcomes by providing synergistic effects where the neuromodulation system works cooperatively with established treatment modalities to achieve superior clinical results compared to either approach alone, creating additive or multiplicative therapeutic benefits that exceed the sum of individual treatment effects. The integration of neuromodulation with conventional therapies may enable personalized treatment protocols that can be tailored to individual patient characteristics, disease severity, comorbid conditions, and treatment response patterns to optimize therapeutic outcomes while minimizing adverse effects and treatment burden.

[0059] For instance, in the treatment of type 2 diabetes, the neuromodulation system may be employed alongside pharmaceutical interventions such as metformin for improving insulin sensitivity and reducing hepatic glucose production, insulin therapy for direct glucose control,Attorney Docket No. 40759.0090-RESH-022-01WGsulfonylureas for stimulating pancreatic insulin secretion, GLP- 1 receptor agonists for enhancing incretin effects and promoting satiety, SGLT-2 inhibitors for promoting glucose excretion through the kidneys, or other antidiabetic medications to reduce required medication dosages while maintaining or improving glycemic control, thereby minimizing medication-related side effects such as hypoglycemia, gastrointestinal disturbances, weight gain, or cardiovascular complications while enhancing overall therapeutic effectiveness through complementary mechanisms of action. The neuromodulation system may work synergistically with these medications by addressing the underlying neural dysregulation that contributes to diabetes pathophysiology, such as autonomic imbalance affecting pancreatic function, hepatic glucose regulation, or peripheral insulin sensitivity, while the medications provide direct biochemical intervention to control glucose levels and improve metabolic function.

[0060] The system may complement dietary management and lifestyle interventions by providing physiological support that makes dietary compliance more effective and sustainable, such as modulating appetite control mechanisms through hypothalamic pathway regulation, enhancing insulin sensitivity to improve the body's response to nutritional interventions, regulating gastric emptying to optimize nutrient absorption and postprandial glucose responses, or influencing satiety signaling to support portion control and meal timing strategies that are critical for diabetes management. The neuromodulation system may enhance the effectiveness of medical nutrition therapy by supporting the physiological changes needed to maintain dietary modifications long-term, addressing the neural mechanisms that often undermine dietary adherence such as food cravings, emotional eating, or metabolic adaptation that can sabotage weight management efforts.Attorney Docket No. 40759.0090-RESH-022-01WG

[0061] Additionally, the neuromodulation system may be integrated with continuous glucose monitoring systems that provide real-time glucose data for treatment optimization, insulin pump therapy for automated insulin delivery based on glucose trends, and closed-loop artificial pancreas systems to create a comprehensive artificial pancreas approach that combines the precision of automated insulin delivery with the physiological benefits of neural pathway modulation. This integration may enable more sophisticated control algorithms that incorporate both glucose feedback and neural modulation parameters to achieve tighter glycemic control with reduced risk of hypoglycemia, improved time-in-range metrics, and enhanced quality of life for patients with diabetes. The combined system may also incorporate predictive algorithms that anticipate glucose excursions based on meal intake, physical activity, stress levels, or circadian patterns and proactively adjust both insulin delivery and neural modulation parameters to prevent dangerous glucose fluctuations.

[0062] In obesity treatment, the system may work synergistically with bariatric surgery procedures such as gastric bypass, sleeve gastrectomy, or adjustable gastric banding by providing ongoing neural modulation that supports the anatomical changes created by surgical intervention, potentially improving long-term weight maintenance and reducing the risk of weight regain that commonly occurs after surgical procedures due to metabolic adaptation, hormonal changes, or behavioral factors. The neuromodulation system may address the neural pathways involved in appetite regulation, energy expenditure, and metabolic rate that are often disrupted following bariatric surgery, helping to maintain the beneficial effects of surgical intervention while supporting the lifestyle changes necessary for sustained weight loss. The system may also complement non-surgical weight management approaches such as very low-calorie diets, mealAttorney Docket No. 40759.0090-RESH-022-01WGreplacement programs, or behavioral weight loss interventions by providing physiological support that enhances adherence to these challenging interventions and improves their long-term effectiveness.

[0063] For cardiovascular conditions such as hypertension, the neuromodulation system may be combined with antihypertensive medications including ACE inhibitors, angiotensin receptor blockers, calcium channel blockers, beta-blockers, or diuretics to achieve better blood pressure control with lower medication doses, reducing the risk of medication-related side effects such as electrolyte imbalances, sexual dysfunction, fatigue, or cough while providing more stable and sustained blood pressure management through complementary mechanisms that address both the neural and humoral factors contributing to hypertension. The system may target sympathetic nervous system overactivity that is a key driver of hypertension while medications address other pathophysiological mechanisms such as renin- angiotensin system activation, sodium retention, or vascular resistance.

[0064] The system may also complement cardiac rehabilitation programs by facilitating autonomic nervous system rebalancing that supports the cardiovascular benefits of exercise training and lifestyle modifications, potentially enhancing exercise tolerance, improving heart rate variability, reducing arrhythmia risk, and supporting the long-term cardiovascular benefits of rehabilitation interventions. The neuromodulation system may help patients achieve better outcomes from cardiac rehabilitation by addressing autonomic dysfunction that often persists after cardiac events and can limit the effectiveness of exercise-based interventions.Attorney Docket No. 40759.0090-RESH-022-01WG

[0065] In neurological conditions such as Parkinson's disease, the neuromodulation system may be used alongside dopaminergic medications such as levodopa, dopamine agonists, or MAO-B inhibitors, as well as physical therapy, occupational therapy, and speech therapy to provide comprehensive symptom management that addresses both the motor symptoms such as tremor, rigidity, and bradykinesia, and non-motor aspects of the disease such as autonomic dysfunction, sleep disorders, mood disturbances, and cognitive changes, potentially allowing for reduced medication doses and improved quality of life while minimizing the motor fluctuations and dyskinesias that often develop with long-term dopaminergic therapy. The system may target non-dopaminergic pathways that contribute to Parkinson's symptoms, providing therapeutic benefits that complement the effects of dopamine replacement therapy.

[0066] The combination approach may also include integration with rehabilitation protocols to facilitate neural plasticity and accelerate recovery processes in conditions such as stroke or traumatic brain injury, where the neuromodulation system can provide physiological support that enhances the effectiveness of physical therapy for motor recovery, occupational therapy for functional independence, and speech therapy interventions for communication and swallowing function. The system may promote neuroplasticity through targeted stimulation of specific neural pathways while rehabilitation exercises provide the behavioral training necessary to strengthen and consolidate new neural connections, creating a synergistic approach that may improve functional outcomes and accelerate recovery timelines.

[0067] For chronic pain conditions, the system may be combined with pharmacological pain management including non-opioid analgesics, anticonvulsants, antidepressants, or whenAttorney Docket No. 40759.0090-RESH-022-01WGnecessary, carefully managed opioid therapy, along with physical therapy for functional restoration, psychological interventions such as cognitive-behavioral therapy or mindfulnessbased stress reduction, and complementary approaches such as acupuncture or massage therapy to provide multimodal pain relief that addresses the complex neurobiological mechanisms including central sensitization and descending pain modulation, physical factors such as muscle tension and joint dysfunction, and psychological aspects such as pain catastrophizing, depression, and anxiety that contribute to chronic pain syndromes. This comprehensive approach may reduce reliance on opioid medications while providing more effective and sustainable pain management through the integration of neural pathway modulation with conventional pain management strategies, potentially improving functional outcomes, quality of life, and long-term prognosis for patients with chronic pain conditions.

[0068] The integration of neuromodulation with conventional treatments may also enable adaptive treatment protocols that can be modified based on treatment response, disease progression, or changing patient needs, allowing for personalized medicine approaches that optimize therapeutic outcomes while minimizing treatment burden and healthcare costs. Advanced integration may include the use of artificial intelligence and machine learning algorithms that can analyze data from multiple treatment modalities simultaneously to identify optimal treatment combinations, predict treatment responses, and automatically adjust therapy parameters to maintain optimal therapeutic outcomes as patient conditions change over time.

[0069] Figure 3 provides a schematic of an exemplary neuromodulation system 100 having a first neuroregulator 102 to neuromodulate a first neuronal target and a second neuroregulator 104Attorney Docket No. 40759.0090-RESH-022-01WGto neuromodulate a second neuronal target. In this example, each of the first and second neuroregulators 102, 104 comprise electrical neuroregulators that can provide electrical stimulation and / or electrical conduction blocking. In other examples, the first neuroregulator of the neuromodulation system may comprise an electrical neuroregulator while the second neuroregulator of the neuromodulation system may comprise a non-electrical neuroregulator, e.g. the non-electrical stimulation and / or conduction block neuroregulators described in the paragraphs above. In addition to an output therapy system there would also be an input sensor system 126 (such as a glucose sensor) which has communication circuitry 125 which can communicate with communication circuitry 127 of one or more rechargeable neuroregulators 102 and 104. The sensor can also communicate with an external programming device’s 118 communication circuitry 129. The programing device can also receive input from a hand held monitor 134 (such as a glucose monitor). The programing device 118 has an Al and ML engine and its communication circuitry 129 can then communicate with the neuroregulator(s) 102 and 104 communication circuitry 127 to optimize therapy. The programming device 118 can also communicate with a clinic’s 131 communication device 130. Data from the clinic can then be uploaded to an Al and ML algorithm engine 132. The Al and ML algorithm engine can. for example, be on a central computer system / repository / on a cloud or server based information system. In turn the Al and ML algorithm engine on the central computer system / repository / on a cloud / server 132 can then send recommendations, based data from a population of patients, of optimal therapy to the clinic 131 which can then communicate with the programming device 118 to instruct the neuroregulator(s) 102 and 104 to deliver different therapy parameters. In the case of an immediate medical emergency the sensor 126 can communicate directly with the neuroregulator(s) 102 and 104 toAttorney Docket No. 40759.0090-RESH-022-01WGchange the therapy parameters. An example of this would be a patient going into a severe hypoglycemic state. The neuroregulators could then change block or stimulation parameters to quickly increase blood glucose levels. Continuing with the example of Figure 3, the neuromodulation system 100 generally includes an external component 106 that resides outside the body 108 and an internal component 110 that is implanted within the body 108 below the dermis. The external component 106 includes a charger 112 that is coupled, via connector 113, to a battery 114 and a transmit coil 116. The internal component 110 includes the first and second neuroregulators 102, 104. each of which is coupled by one or more leads 120 to one or more electrodes 122, which are placed on the neuronal targets. Each of the neuroregulators 102, 104 includes a rechargeable battery 124 and a receiving antenna 126. In certain examples, the neuroregulators 102, 104 share a common battery 124 and a common receiving antenna 126.

[0070] In the example of Figure 3, the leads 120 comprise bipolar leads each of which is connected to first and second electrodes 122a, 122b. However, other lead and electrode configurations may be used as appropriate to a specific application. In certain examples, each of the neuroregulators 102, 104 comprises an implantable component that is independent from the other while in other examples the first and second neuroregulators 102, 104 are combined within a single implantable component. In certain examples, each of the neuroregulators 102, 104 is configured to interface with a common external component 106 while in other examples each of the neuroregulators 102, 104 interfaces with their own corresponding external component 106.

[0071] In operation, the first and second neuroregulators 102, 104 of the neuromodulation system 100 produce electrical pulses that are delivered to their respective first and second neuronalAttorney Docket No. 40759.0090-RESH-022-01WGtargets through the electrically conductive leads 120 and electrodes 122. In addition to delivering the electrical pulses, each of the neuroregulators 102, 104 also receives wireless command signals from the programming device 118. Each of the neuroregulators 102, 104 is powered by their internal battery 124; in certain examples, the neuroregulators 102, 104 share a common internal battery’ 124. The internal battery 124 is periodically recharged by RF power that is radiated by the transmit coil 116 and picked up by the receiving antenna 126. The charger 112 provides the electrical excitation of the transmit coil 116 needed to deliver RF power to the neuroregulators 102, 104. In certain examples, a rechargeable battery (not shown) powers the charger 112. The transmit coil 116 serves to transmit RF power transdermally from the charger 112 to the neuroregulators 102, 104. The transmit coil 116 further facilitates bi-directional RF communications between the neuroregulators 102, 104 and the charger 112. The programming device 118 enables a clinician to program each of the neuroregulators 102, 104 with a treatment schedule and with therapy parameters for delivering stimulation or conduction blocking to their respective neuronal targets.

[0072] An example of a neuromodulation system similar to the one described above, albeit with only one neuroregulator and a different communication system, is the MAESTRO® Rechargeable System available from ReShape Lifesciences, Inc., 1001 Calle Amanecer, San Clemente, CA 92673, USA. The MAESTRO® Rechargeable System is described in detail in US Patent Publication Nos.: US 7,489,969; US 7,167,750; US 7,444,183; US 7,613,515; US 7,720,540; US 7,630,769; US 7,693,577; US 7,729,771; US 7,844,338; US 8,046,085; US 7,986,995; US 8,010,204; US 8,369,952; US 8,538,542; US 9,174,040; US 8,538,533; US 9,162,062, US 8,862,233; US 7,672,727; US 8,103,349; US 7,822,486; US 8,140,167; USAttorney Docket No. 40759.0090-RESH-022-01WG8,532.787; US 8,068,918; US 8,521,299; US 7,917,226; US 8,483,838; US 8,326,426; US 9,186,502; US 8,483,830; US 9,333,340; US 6,699,275: US 6,860.851; US 8.825,164; US 2014 / 0214129; US 9.393.420; US 8,101,204; US 8,768,469; US 9.095.711; and US 2013 / 0237948. E ach of the noted patent publications is hereby incorporated by reference in its entirety.DEFINITIONS

[0073] As used herein, the term "neuromodulation" refers to the targeted alteration of neural activity through the delivery of electrical, chemical, optical, magnetic, ultrasonic, or thermal stimuli to specific neural structures, or through the application of conduction blocks that reversibly or permanently interrupt neural signal transmission. Neuromodulation encompasses both stimulation techniques that enhance or trigger neural activity and blocking techniques that inhibit or prevent neural conduction.

[0074] As used herein, the term "neuronal target" refers to any neural structure selected for therapeutic intervention, including but not limited to sympathetic nerves, parasympathetic nerves, central nervous system tissue, peripheral nerve branches, nerve plexuses, individual nerve fascicles, afferent nerve fibers, efferent nerve fibers, or smooth muscle tissue that can be modulated to achieve a therapeutic effect.

[0075] As used herein, the term "neuroregulator" refers to any device or system capable of delivering neuromodulation to a neuronal target, including electrical stimulators, electrical conduction blocks, chemical agent stimulators, chemical conduction blocks, physical ablationAttorney Docket No. 40759.0090-RESH-022-01WGdevices, optogenetic modulators, magnetic stimulators, ultrasonic stimulators, optical stimulators, thermal stimulators, or hybrid devices that combine multiple modulation modalities.

[0076] As used herein, the term "cooperative neuromodulation" or "works in cooperation" describes the coordinated modulation of two or more neuronal targets wherein the therapeutic effects are achieved through synergistic amplification where the combined effect exceeds the sum of individual target effects, complementary pathway modulation where different targets address distinct aspects of the same pathological process, or coordinated physiological responses that enhance overall homeostatic regulation while preventing maladaptive compensatory mechanisms.

[0077] As used herein, the term "conduction block" refers to the reversible or permanent interruption of neural signal transmission through a nerve or neural pathway, achieved through electrical high frequency alternating current (HFAC), chemical agents, physical ablation, optogenetic inhibition, cryogenic cooling, thermal ablation, focused ultrasound, or magnetic field interference.

[0078] As used herein, the term "bio-feedback sensor" refers to any device capable of monitoring physiological parameters related to a medical condition being treated, including but not limited to biochemical sensors, mechanical sensors, electrical sensors, optical sensors, thermal sensors, glucose sensors, blood pressure monitors, heart rate sensors, oxygen saturation monitors, temperature sensors, pH sensors, or neurotransmitter sensors that provide real-time information about patient physiological state and treatment response.Attorney Docket No. 40759.0090-RESH-022-01WG

[0079] As used herein, the term "artificial intelligence and machine learning algorithm" refers to computational systems that employ neural networks, deep learning, reinforcement learning, Bayesian optimization, genetic algorithms, support vector machines, decision trees, ensemble methods, convolutional neural networks, recurrent neural networks, transformer architectures, federated learning, or other advanced computational approaches to analyze physiological data, generate predictive recommendations, and optimize neuromodulation parameters based on real-time feedback and patient responses.

[0080] As used herein, the term "optimized neuromodulation parameters" refers to therapy settings determined through artificial intelligence and machine learning analysis, including but not limited to turning stimulation or conduction block on or off, adjusting current amplitude, modifying frequency settings, changing signal patterns such as bursting, altering timing of application, or coordinating temporal sequences between multiple neuronal targets to maximize therapeutic efficacy while minimizing adverse effects.

[0081] As used herein, the term "temporal coordination" refers to the timing relationships between neuromodulation of different neuronal targets, including synchronized therapeutic effects over common time periods, coordinated but not identical timing during overlapping periods, sequential therapeutic interventions during distinct non- overlapping periods, or complex combinations thereof to optimize therapeutic outcomes for specific medical conditions and individual patient responses.

[0082] As used herein, the term "closed-loop system" refers to a neuromodulation configuration that enables real-time adaptation of therapy parameters based on physiologicalAttorney Docket No. 40759.0090-RESH-022-01WGfeedback from bio-feedback sensors, creating a dynamic treatment system that automatically responds to changing patient conditions through artificial intelligence and machine learning algorithms without requiring manual intervention.

[0083] Numerous medical conditions can be treated through the neuromodulation of two or more neuronal targets. Various neuromodulation systems, including those described herein can be used to achieve the desired neuromodulation. Table 1 below provides just some examples of at least first and second neuronal targets that can be neuromodulated (stimulated (Stirn) or conduction blocked (Block)) for treatment of a medical condition.Attorney Docket No. 40759.0090-RESH-022-01WGTable 1.First Neuronal Target Second Neuronal Target Medical ConditionCardio Splenic Nerve - Block Cervical V agus Nerve - Stim Heart FailureV agus Nerve - Stim Greater and / or Lesser Inflammation Splanchic Nerve - BlockBaroreceptors - Stim Renal Nerves - Block HypertensionVagus Nerve - Block Greater and / or Lesser Pancreatitis Splanchic Nerve - StimVagus Nerve - Block Sympathetic Nerves of Refractory Asthma Smooth Muscles of the LungsInnervation - StimSacral Nerves - Stim Lumbar Sympathetic Nerves Urinary Incontinence — BlockSacral Nerves - Stim Lumbar Sympathetic Nerves Erectile Dysfunction - BlockVago- Vagal Reflex (A Delta Satiety Afferents - Stim* ObesityFibers of the Sub-Attorney Docket No. 40759.0090-RESH-022-01WGDiaphragmatic Vagus Nerve)- BlockSubthalamic Nucleus (STN) - Cortex - Block Parkinson’s Disease StimReward Centers - Stim Vagus Nerve - Block ObesityBrodmann Area 25 - Stim or V agus Nerve - Stim Refractory Depression BlockVagus Nerve - Stim Cortex - Block Refractory EpilepsyOccipital Nerve - Stim Trigeminal Nerve - Block (or Cluster Headaches Ablation)Afferents of Peripheral Nerve Efferents of Peripheral Nerve Obesity and Other Various - Stim - Block Medical ConditionsAfferents of Peripheral Nerve Efferents of Peripheral Nerve Various Medical Conditions - Block - StimCeliac Nerve (or Posterior Hepatic Nerve (or Anterior Type II DiabetesV agus Nerve above the Vagus Nerve above theBranching Point of the Celiac Branching Point of theNerve) - Stim Hepatic Nerve)-BlockThird Neuronal Target - Gall Bladder (Stim) for Bile ProductionAttorney Docket No. 40759.0090-RESH-022-01WG

[0084] A specific example illustrating the neuromodulation of two neuronal targets, e.g., the celiac nerve and hepatic nerve, for treatment of a medical condition, e.g. Type II Diabetes, through use of a neuromodulation system is provided below.

[0085] Example: Type II Diabetes - Results of Rat Model Study

[0086] The incidence of type II diabetes (which accounts for 90% of the cases of diabetes, World Health Organization (WHO)) is on the rise with significant consequences. In 2012 1.5 million deaths were attributed to diabetes and 2.2 million deaths were related to high blood glucose (WHO). In the US diabetes is the 6thleading cause of death. People living with type II diabetes for years can develop neuropathic pain, nerve degeneration, blindness, and amputation. Treatments are limited and there is a large unmet need for novel therapies.

[0087] Type II diabetes takes a large toll on the body due to the toxic effect of sustained high blood glucose concentration. One such organ, the pancreas, plays a key role in stabilizing high blood glucose concentration. With prolonged high blood sugar levels, islet cells of the pancreas have a diminished ability to produce insulin which further decreases the body’s ability to cope with high glucose. Finding a method to stabilize blood glucose levels for type II diabetics would stop this harmful cycle.

[0088] With the above in mind, a study was performed to determine the possible effectiveness of controlling glucose levels through stimulation of the posterior vagus nerve at the level of the celiac, e.g. a first neuronal target, in combination with a block of the hepatic nerve, e.g. a secondAttorney Docket No. 40759.0090-RESH-022-01WGneuronal target. More specifically, a differential modulation procedure was tested by using an IV glucose challenge (injection of a high concentration of glucose into the circulatory system) following a hepatic nerve vagotomy, or high frequency alternating current conduction block, with concurrent stimulation of the posterior vagus nerve at the level of the celiac branch in a rat model.

[0089] Methods

[0090] Rats (250-300 grams) were divided into a control (n=5) and 2 experimental groups (n=5 for stimulation / vagotomy and 6 for stimulation / 5000Hz). In all groups, rats were fasted for at least 18 hours similar to other studies investigating changes in blood glucose concentrations (Lee and Miller, 1985). A sham surgery was performed in the control group. First, the rats were anesthetized with an intramuscular injection of a combination of ketamine, xylazine and acepromazine. Reflex testes were periodically preformed to determine that the rat was anesthetized and if the rat was lightly conscious an injection of ketamine was given. Next, the abdominal cavity was opened and the liver retracted. The hepatic nerve branch of the anterior vagal trunk (which will be referred to as the “hepatic branch”) and the celiac nerve branch of the posterior vagus nerve (which will be referred to as the “celiac branch”) were isolated and, using gentle dissection, were separated from the esophagus for the sham procedure. In the experimental groups, a similar surgical approach was performed as in the sham procedure except that the celiac branch was positioned on a bipolar platinum hook electrode for electrical stimulation and combined with vagotomy of the hepatic branch in the first experimental group and an electrode was positioned underneath the hepatic nerve to achieve conduction block combined with stimulation of the celiac branch in the second experimental group.Attorney Docket No. 40759.0090-RESH-022-01WG

[0091] Blood samples were taken by wrapping a warm cloth around the tail of the rat (to stimulate blood flow) followed by cutting the end of the tail of the rat. The tail was then “milked” by squeezing from the end connecting the tail to the body of the rat to the cut end to ensure fresh systemic blood was sampled. An AlphaTrak (Abbott Laboratories, North Chicago, IL, USA) blood glucose monitor was used to measure blood glucose concentrations (mg / dl) from the milked rat tail.

[0092] A blood glucose measurement was taken at the start of the experiment and considered baseline. In experimental group 1, the hepatic branch was cut and the celiac branch was stimulated immediately following baseline sampling at a rate of 1 Hz. A piece of synthetic interstitial fluid (SIF) soaked gauze was placed between the hooks of the stimulation electrode to insure that no desiccation of the nerve occurred during the course of the experiment. A square wave with a pulse width of 4 millisecond was generated by a grass s44 stimulator (Grass Medical Instruments, Quincy, MA, USA) which drove a stimulus isolation unit (Model A360, World Precision Instruments, Sarasota. FL, USA). The pulse amplitude was 9 mA. In experimental group 1, the stimulation protocol was given for 1 hour with periodic sampling of blood glucose concentration. For the sham group, no stimulation was delivered to the celiac branch and the hepatic branch was not cut, but blood samples were taken with the same time course as the experimental group.

[0093] In experimental group 2, the celiac branch was stimulated with the same procedure as experimental group 1. However, instead of a hepatic nerve vagotomy, a 5000 Hz alternating current signal was applied to the hepatic branch with a bipolar platinum iridium hook electrode (the “blocking electrode”). The device delivering the 5000 Hz signal was a rechargeable neuralAttorney Docket No. 40759.0090-RESH-022-01WGregulator (RNR) similar to one used in the ReShape Lifesciences’ Recharge Clinical Study. A piece of SIF-soaked gauze was placed between the hooks of the blocking electrode to insure that no desiccation of the nerve occurred during the course of the experiment. The soaked gauze also decreased the impedances between the bipolar blocking electrodes to a level that was in the safety range of the RNR. The current amplitude was 12 mA with a typical impedance of 1000 ohms (measured at 1000 Hz at 3 mA). Voltages were measured with a portable fluke (Everett, WA, USA) oscilloscope across the blocking electrodes and were typically around 7 volts. This was smaller than the anticipated 12 volts which would be predicted by the impedance test. Previous experiments have determined that the differences between the anticipated and measured voltages are due to the differences in frequency and current amplitude between the impedance test and the blocking signal. Following the baseline blood sample, stimulation of the celiac branch and delivery of 5000 Hz to the hepatic nerve was concurrently initiated. A blood sample was then taken at 5 and 15 minutes following the start of the experiment.

[0094] One hour following the initiation of experimental group 1 conditions (and sham) and 15 min following initiation of experimental group 2 conditions, a blood glucose challenge was performed. The blood glucose challenge consisted of an IV injection into the tail vein of a 0.5 g / kg dose of glucose made up in 0.9% saline with a 20% weight / volume concentration. Blood glucose was then sampled for 30 minutes following injection. Stimulation / vagotomy (experimental group 1) or stimulation / 5000 Hz (experimental group 2) was continuously delivered during the glucose challenge and 30 minutes following. In 4 out of 6 rats in experimental group 2, a second glucose challenge was administered 15 minutes following the termination of the stimulation / 5000 Hz procedure (Figure 4).Attorney Docket No. 40759.0090-RESH-022-01WG

[0095] Statistics consisted of a 2 tailed student’s t-test assuming unequal variance performed with Microsoft Excel software (Redmond, WA, USA). All data are presented as mean ± SEM. Percent change in glucose concentration was calculated using the following equation:

[0096] % Change - ((Blood glucose concentration at time x - Baseline blood glucose concentration) / (Baseline blood glucose concentration))* 100

[0097] Area under the curve following the glucose challenge (% change in glucose concentration * time = area units) was calculated by assuming linearity between data points. The area between the line connecting two subsequent data points and the x-axis was calculated as one segment. The total number of segments following the glucose challenge was then summated.

[0098] Results

[0099] Experimental group 1

[0100] Effect of hepatic vagotomy and celiac branch stimulation before glucose challenge

[0101] Blood glucose concentrations for the sham group remained relatively constant for the hour before the glucose challenge (Figure 5). At 5 min following stimulation / vagotomy in experimental group 1 there was no apparent change in glucose concentration. However, starting at 15 min there was a slight decrease in blood glucose concentration which lasted for the 60 min prior to the glucose tolerance test (Figure 5). As stated in the methods, the rats were fasted and had a low initial blood glucose concentration (typically around 170 mg / dL) which may have caused a floor effect. Therefore, a glucose challenge was administered to test if theAttorney Docket No. 40759.0090-RESH-022-01WGstimulation / vagotomy combination would give the rat the ability to tolerate a bolus injection of glucose into its circulatory system.

[0102] Glucose tolerance test

[0103] Five minutes following the glucose injection the sham group experienced, on average, a large increase in blood glucose concentration (60 ± 22%, Figure 6). The time of peak glucose concentration was variable between sham rats. The average in peak glucose concentration in sham was an increase of 90 + 16% (Figure 7). The glucose concentration remained high for the 30 minutes following the IV glucose injection with an apparent decrease at 30 minutes.

[0104] In the experimental group 1, 5 minutes following the glucose injection there was only a 5 ± 14% increase in glucose concentration, on average (Figure 6). The glucose concentration remained relatively low for the 30 minutes following the glucose challenge with an average peak (21 ± 13% increase) at 15 minutes post injection. As with the sham group, the time of the peak in glucose concentration was variable between rats. The average peak (21 ± 14% increase) in experimental group 1 was significantly lower than sham (Figure 7, p = 0.01).

[0105] The area under the curve of the % change in glucose concentration versus time following the glucose challenge was calculated as a measure of the total effect of the glucose tolerance test. This demonstrated a noticeable difference between the two groups (1704±553 area units for sham vs 202+322 area units for stimulation / vagotomy). Thus, not only was there a difference in the peak glucose concentrations between groups, but the total ability for experimental group 1 to tolerate a glucose challenge over time was strikingly greater than sham.Attorney Docket No. 40759.0090-RESH-022-01WG

[0106] Experimental group 2

[0107] Five thousand Hz alternating current applied to the hepatic nerve with concurrent stimulation of the celiac branch significan tly improved the ability to cope with a glucose challenge

[0108] It has been shown that a 5000 Hz alternating current signal will reversibly block conduction through sub-diaphragmatic vagus nerve in rat. Here, the hypothesis that applying a 5000 Hz alternating current signal to the hepatic nerve while stimulating the celiac branch will have the same effect as cutting the hepatic nerve (with celiac branch stimulation) was tested. Similar to experimental group 1 there was a minor decrease in blood glucose concentration prior to the glucose challenge (Figure 8).

[0109] Following the glucose challenge blood glucose concentrations remained similar to experimental group 1 and noticeably lower than sham (Figure 6). The time of the peak in blood glucose concentration following the challenge varied in experimental group 2 similar to experimental group 1 and sham. The average peak was a 31 ± 6% change from baseline which was significantly lower than sham (p = 0.005, Figure 7). The ability of the rat to cope with the glucose challenge in experimental group 2 was also evident in its area under the curve value compared to sham (1704+553 area units for sham vs 418+140 area units for stimulation / 5000 Hz).

[0110] Reversibility of the stimulation / 5000 Hz procedure

[0111] The large advantage of using 5000 Hz to block conduction through the hepatic nerve versus a complete hepatic vagotomy is that 5000 Hz conduction block is reversible allowing for normal communication from the brain to the liver (and vice versa) to occur at controlled times. ToAttorney Docket No. 40759.0090-RESH-022-01WGtest if delivering 5000 Hz to the hepatic nerve along with stimulation of the celiac branch is reversible, the signals were turned off 30 minutes following the glucose challenge and 15 minutes later a subsequent glucose challenge was performed (timeline of procedure in Figure 4) in 4 out of the 6 rats in experimental group 2.

[0112] During the 15 minutes post stimulation / 5000 Hz block, blood glucose concentrations remained relatively stable (Figure 9). Following the 2ndblood glucose challenge at time 15 minutes post stim / block (with all electrical signals turned off), there was a large increase in blood glucose concentration. As with all other glucose challenges in this study, the time of the peak blood glucose concentration varied between rats. The average peak was a 130 + 42% increase which was not significantly different from sham, but had a slight apparent increase. Area under the curve was also not significantly different than sham (1704+553 area units for sham vs 2566+915 area units for the 2ndglucose challenge). It should be noted that 1 out of the 6 rats in experimental group 2 had a large increase in blood glucose concentration (similar to sham) following the first glucose challenge. Again, to prevent the nerves from desiccation, SIF soaked gauze was placed between the hooks of the stimulation and blocking electrodes. SIF is conductive and it was likely that the blocking electrode and / or the stimulation electrode had shorted out during this experiment. This rat was considered an outlier and not included in the data set.

[0113] This study demonstrated that by blocking neuronal information traveling to and from the liver while stimulating the celiac nerve (at the branching point from the posterior vagus nerve) significantly lowers blood glucose concentration following a glucose tolerance test compared to sham. Further, by using 5000 Hz to block neuronal information down the hepatic nerve (versus aAttorney Docket No. 40759.0090-RESH-022-01WGvagotomy), and stimulation of the celiac branch, makes this procedure reversible. The stimulation can always be turned off and it is well established that high frequency conduction block is reversible. This was demonstrated by the rat’s biological system behaving in a similar manner as the sham group (large increase in blood glucose following a glucose challenge) after cessation of the stimulation / 5000 Hz procedure.

[0114] The results of the study suggest that a device which uses high frequency conduction block of the anterior sub-diaphragmatic vagal nerve trunk above the level of the hepatic branch and low frequency stimulation of the posterior sub-diaphragmatic vagal nerve trunk above the level of the celiac branch may be an effective method of glycemic control and a therapy for type II diabetes.

[0115] It will be appreciated that aspects of the various embodiments disclosed herein may be combined in any way to provide numerous additional embodiments. These embodiments will not be described individually for the sake of brevity.

[0116] While the present invention has been described above primarily with reference to the accompanying drawings, it will be appreciated that the invention is not limited to the illustrated embodiments; rather, these embodiments are intended to disclose the invention to those skilled in this art. In the drawings, like numbers refer to like elements throughout.

[0117] It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed aAttorney Docket No. 40759.0090-RESH-022-01WGsecond element, and, similarly, a second element could be termed a first element, without departing from the scope of the present invention.

[0118] Well-known functions or constructions may not be described in detail for brevity and / or clarity. As used herein the expression “and / or” includes any and all combinations of one or more of the associated listed items.

[0119] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises”, “comprising”, “includes” and / or “including” when used in this specification, specify the presence of stated features, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, operations, elements, components, and / or groups thereof.

[0120] Herein, the terms “attached”, “coupled”, “connected”, “interconnected”, “contacting”, “mounted” and the like can mean either direct or indirect attachment or contact between elements, unless stated otherwise.

[0121] Aspects of the present disclosure are further described in the below clauses:

[0122] Clause 1: A system for regulating at least two neuronal targets using artificial intelligence and machine learning, comprising a first neuroregulator configured to apply a stimulation or a conduction block to a first neuronal target, a second neuroregulator configured to apply a stimulation or a conduction block to a second neuronal target that is different from the first neuronal target, a bio-feedback sensor configured to monitor a physiological parameter related toAttorney Docket No. 40759.0090-RESH-022-01WGa medical condition, a programming device comprising communication circuitry and an artificial intelligence and machine learning algorithm, wherein the programming device is configured to receive data from the bio-feedback sensor and generate optimized neuromodulation parameters, and communication circuitry configured to transmit the optimized neuromodulation parameters to the first neuroregulator and the second neuroregulator, wherein the stimulation or conduction block of the first neuronal target works in cooperation with the stimulation or conduction block of the second neuronal target to treat the medical condition based on the optimized neuromodulation parameters generated by the artificial intelligence and machine learning algorithm.

[0123] Clause 2: The system of Clause 1, wherein the first neuronal target and the second neuronal target each comprise one or more sympathetic nerves.

[0124] Clause 3: The system of Clause 1, wherein the first neuronal target and the second neuronal target each comprise one or more parasympathetic nerves.

[0125] Clause 4: The system of Clause 1, wherein the first neuronal target comprises one or more sympathetic nerves and the second neuronal target comprises one or more parasympathetic nerves.

[0126] Clause 5: The system of any of Clauses 2-4, wherein the first neuronal target comprises afferent nerve fibers and the second neuronal target comprises efferent nerve fibers.

[0127] Clause 6: The system of Clause 1, wherein the first neuroregulator applies a stimulation to the first neuronal target and the second neuroregulator applies a stimulation to the second neuronal target.Attorney Docket No. 40759.0090-RESH-022-01WG

[0128] Clause 7: The system of Clause 1, wherein the first neuroregulator applies a conduction block to the first neuronal target and the second neuroregulator applies a conduction block to the second neuronal target.

[0129] Clause 8: The system of Clause 1, wherein the first neuroregulator applies a conduction block to the first neuronal target and the second neuroregulator applies a stimulation to the second neuronal target.

[0130] Clause 9: The system of Clause 8, wherein the first neuronal target comprises a hepatic nerve and the second neuronal target comprises a celiac nerve.

[0131] Clause 10: The system of Clause 9, wherein the medical condition is type 2 diabetes.

[0132] Clause 11: The system of Clause 8, wherein the first neuronal target comprises an anterior vagus nerve cranial to a branching point of the hepatic nerve and the second neuronal target comprises a posterior vagus nerve cranial to a branching point of the celiac nerve.

[0133] Clause 12: The system of any of Clauses 6-8, wherein the stimulation or conduction block applied to the first neuronal target has a different duration than the stimulation or conduction block applied to the second neuronal target.

[0134] Clause 13: The system of any of Clauses 6-8, wherein the stimulation or conduction block applied to the first neuronal target has different start and end times than the stimulation or conduction block applied to the second neuronal target.

[0135] Clause 14: The system of Clause 1, wherein the first neuroregulator and the second neuroregulator each comprise electrical neuro regulators.

[0136] Clause 15: The system of Clause 14, wherein the electrical neuroregulators are configured to deliver electrical pulses with current amplitudes ranging from 0.25 mA to 20 mA.Attorney Docket No. 40759.0090-RESH-022-01WG

[0137] Clause 16: The system of Clause 15, wherein the electrical neuroregulators are configured to deliver electrical pulses with current amplitudes ranging from 0.25 mA to 50 mA.

[0138] Clause 17: The system of any of Clauses 15-16, wherein the conduction block comprises high frequency alternating current with frequencies ranging from 200 Hz to 100 kHz.

[0139] Clause 18: The system of Clause 17, wherein the conduction block comprises high frequency alternating current with frequencies ranging from 200 Hz to 1 MHz.

[0140] Clause 19: The system of any of Clauses 17-18, wherein the stimulation comprises low frequency signals with frequencies ranging from 0.01 Hz to 199 Hz.

[0141] Clause 20: The system of Clause 19. wherein the stimulation comprises low frequency signals with frequencies ranging from 0.01 Hz to 500 Hz.

[0142] Clause 21: The system of Clause 1, wherein at least one of the first neuroregulator and the second neuroregulator comprises a non-electrical neuroregulator.

[0143] Clause 22: The system of Clause 21, wherein the non-electrical neuroregulator comprises a chemical agent stimulator or a chemical conduction block.

[0144] Clause 23: The system of Clause 21, wherein the non-electrical neuroregulator comprises an optogenetic modulator configured to deliver light-activated neural modulation.

[0145] Clause 24: The system of any of Clauses 21-23, wherein the non-electrical neuroregulator comprises a magnetic stimulator, ultrasonic stimulator, or thermal stimulator.

[0146] Clause 25: The system of Clause 1, wherein the bio-feedback sensor comprises a glucose sensor configured to monitor blood glucose levels.

[0147] Clause 26: The system of Clause 25. wherein the glucose sensor has a temporal resolution ranging from about 1 second to about 10 minutes.Attorney Docket No. 40759.0090-RESH-022-01WG

[0148] Clause 27: The system of Clause 26, wherein the glucose sensor is implantable.

[0149] Clause 28: The system of Clause 25. wherein the glucose sensor comprises a continuous glucose monitor or flash glucose monitor.

[0150] Clause 29: The system of Clause 1, wherein the bio-feedback sensor comprises multiple sensor types selected from biochemical sensors, mechanical sensors, electrical sensors, optical sensors, and thermal sensors.

[0151] Clause 30: The system of any of Clauses 25-29, wherein the programming device is configured to receive manual input data comprising at least one of food intake information, insulin injection information, and physical activity information.

[0152] Clause 31: The system of Clause 30, wherein the manual input data further comprises medication schedules, sleep patterns, or stress levels.

[0153] Clause 32: The system of Clause 1, wherein the communication circuitry comprises at least one of Bluetooth technology, radio frequency communication, WiFi communication, and sound communication.

[0154] Clause 33: The system of Clause 32, wherein the communication circuitry further comprises near-field communication, infrared communication, or cellular communication.

[0155] Clause 34: The system of Clause 1, further comprising a clinic communication device configured to communicate with the programming device.

[0156] Clause 35: The system of Clause 34, further comprising a central computer system configured to receive data from the clinic communication device and generate population-based treatment recommendations using artificial intelligence and machine learning algorithms.Attorney Docket No. 40759.0090-RESH-022-01WG

[0157] Clause 36: The system of Clause 35, wherein the central computer system comprises a HIPAA compliant system with blockchain protocol.

[0158] Clause 37: The system of any of Clauses 35-36, wherein the central computer system implements federated learning approaches for knowledge sharing without exposing individual patient data.

[0159] Clause 38: The system of Clause 1, wherein the bio-feedback sensor is configured to communicate directly with the first neuroregulator and the second neuroregulator in response to detecting a medical emergency condition.

[0160] Clause 39: The system of Clause 38. wherein the medical emergency condition comprises a hypoglycemic state.

[0161] Clause 40: The system of any of Clauses 38-39, wherein the medical emergency condition comprises hyperglycemia or cardiac arrhythmias.

[0162] Clause 41: The system of Clause 1, wherein the medical condition comprises at least one of type 2 diabetes, obesity, refractory hypertension, pancreatitis, refractory asthma, urinary incontinence, erectile dysfunction, chronic heart failure, chronic inflammatory disease, Parkinson's disease, refractory depression, refractory epilepsy, and migraine headaches.

[0163] Clause 42: The system of Clause 41, wherein the medical condition comprises chronic pain syndromes, gastroparesis, irritable bowel syndrome, or sleep disorders.

[0164] Clause 43: The system of Clause 1, wherein the artificial intelligence and machine learning algorithm is configured to adjust the optimized neuromodulation parameters over time based on body adaptation to the neuromodulation.Attorney Docket No. 40759.0090-RESH-022-01WG

[0165] Clause 44: The system of Clause 43, wherein the artificial intelligence and machine learning algorithm operates on a time scale ranging from about 1 second to about 10 minutes to continuously learn and optimize the neuromodulation parameters.

[0166] Clause 45: The system of any of Clauses 43-44, wherein the artificial intelligence and machine learning algorithm employs neural networks, deep learning, reinforcement learning, or Bayesian optimization.

[0167] Clause 46: A method for treating a medical condition using artificial intelligence and machine learning guided neuromodulation, comprising neuromodulating, with a stimulation or a conduction block, a first neuronal target using a first neuroregulator, neuromodulating, with a stimulation or a conduction block, a second neuronal target using a second neuroregulator, wherein the second neuronal target is different from the first neuronal target, monitoring a physiological parameter related to the medical condition using a bio-feedback sensor, processing data from the bio-feedback sensor using an artificial intelligence and machine learning algorithm to generate optimized neuromodulation parameters, and adjusting the neuromodulation of the first neuronal target and the second neuronal target based on the optimized neuromodulation parameters, wherein the neuromodulation of the first neuronal target works in cooperation with the neuromodulation of the second neuronal target to treat the medical condition.

[0168] Clause 47: The method of Clause 46, wherein the first neuronal target and the second neuronal target each comprise one or more sympathetic nerves.

[0169] Clause 48: The method of Clause 46, wherein the first neuronal target and the second neuronal target each comprise one or more parasympathetic nerves.Attorney Docket No. 40759.0090-RESH-022-01WG

[0170] Clause 49: The method of Clause 46, wherein the first neuronal target comprises one or more sympathetic nerves and the second neuronal target comprises one or more parasympathetic nerves.

[0171] Clause 50: The method of any of Clauses 47-49, wherein the first neuronal target comprises afferent nerve fibers and the second neuronal target comprises efferent nerve fibers.

[0172] Clause 51: The method of Clause 46, wherein neuromodulating the first neuronal target comprises applying a stimulation and neuromodulating the second neuronal target comprises applying a stimulation.

[0173] Clause 52: The method of Clause 46, wherein neuromodulating the first neuronal target comprises applying a conduction block and neuromodulating the second neuronal target comprises applying a conduction block.

[0174] Clause 53: The method of Clause 46, wherein neuromodulating the first neuronal target comprises applying a conduction block and neuromodulating the second neuronal target comprises applying a stimulation.

[0175] Clause 54: The method of Clause 53, wherein the first neuronal target comprises a hepatic nerve and the second neuronal target comprises a celiac nerve.

[0176] Clause 55: The method of Clause 54, wherein the medical condition is type 2 diabetes.

[0177] Clause 56: The method of any of Clauses 51-53, wherein the neuromodulation of the first neuronal target has a different duration than the neuromodulation of the second neuronal target.Attorney Docket No. 40759.0090-RESH-022-01WG

[0178] Clause 57: The method of any of Clauses 51-53, wherein the neuromodulation of the first neuronal target has different start and end times than the neuromodulation of the second neuronal target.

[0179] Clause 58: The method of Clause 46, wherein the first neuroregulator and the second neuroregulator each comprise electrical neuroregulators.

[0180] Clause 59: The method of Clause 58, wherein the electrical neuroregulators deliver electrical pulses with current amplitudes ranging from 0.25 mA to 20 mA.

[0181] Clause 60: The method of Clause 59, wherein the conduction block comprises high frequency alternating current with frequencies ranging from 200 Hz to 100 kHz.

[0182] Clause 61: The method of Clause 60, wherein the stimulation comprises low frequency signals with frequencies ranging from 0.01 Hz to 199 Hz.

[0183] Clause 62: The method of Clause 46, wherein at least one of the first neuroregulator and the second neuroregulator comprises a non-electrical neuroregulator.

[0184] Clause 63: The method of Clause 62, wherein the non-electrical neuroregulator comprises a chemical agent stimulator or a chemical conduction block.

[0185] Clause 64: The method of any of Clauses 62-63, wherein the non-electrical neuroregulator comprises an optogenetic modulator, magnetic stimulator, ultrasonic stimulator, or thermal stimulator.

[0186] Clause 65: The method of Clause 46, wherein the bio-feedback sensor comprises a glucose sensor configured to monitor blood glucose levels.

[0187] Clause 66: The method of Clause 65, wherein the glucose sensor has a temporal resolution ranging from about 1 second to about 10 minutes.Attorney Docket No. 40759.0090-RESH-022-01WG

[0188] Clause 67: The method of Clause 66, wherein the glucose sensor is implantable.

[0189] Clause 68: The method of any of Clauses 65-67, further comprising a step of receiving manual input data comprising at least one of food intake information, insulin injection information, and physical activity information.

[0190] Clause 69: The method of Clause 46, wherein the optimized neuromodulation parameters comprise at least one of turning stimulation or conduction block on or off, changing current amplitude, changing frequency, changing signal patterns, and changing time of application.

[0191] Clause 70: The method of Clause 69, wherein changing current amplitude comprises adjusting current amplitude within a range of 0.25 mA to 20 mA.

[0192] Clause 71: The method of any of Clauses 69-70, wherein changing frequency comprises adjusting conduction block frequencies within a range of 200 Hz to 10 kHz and adjusting stimulation frequencies within a range of 0.01 Hz to 199 Hz.

[0193] Clause 72: The method of Clause 46, further comprising a step of communicating the optimized neuromodulation parameters using at least one of Bluetooth technology, radio frequency communication, WiFi communication, and sound communication.

[0194] Clause 73: The method of Clause 46, further comprising a step of transmitting data to a clinic communication device.

[0195] Clause 74: The method of Clause 73, further comprising a step of receiving population-based treatment recommendations from a central computer system.

[0196] Clause 75: The method of Clause 74, wherein the central computer system comprises a HIPAA compliant system with blockchain protocol.Attorney Docket No. 40759.0090-RESH-022-01WG

[0197] Clause 76: The method of Clause 46, further comprising a step of detecting a medical emergency condition and directly adjusting the neuromodulation in response to the medical emergency condition.

[0198] Clause 77: The method of Clause 76, wherein the medical emergency condition comprises a hypoglycemic state.

[0199] Clause 78: The method of any of Clauses 76-77, wherein the medical emergency condition comprises hyperglycemia, cardiac arrhythmias, or severe hypotension.

[0200] Clause 79: A neuromodulation system for treating type II diabetes, comprising a first electrical neuroregulator configured to apply a high frequency alternating current conduction block to a hepatic nerve or an anterior vagus nerve cranial to a branching point of the hepatic nerve, a second electrical neuroregulator configured to apply a low frequency electrical stimulation to a celiac nerve or a posterior vagus nerve cranial to a branching point of the celiac nerve, a glucose sensor configured to monitor blood glucose levels, an external programming device comprising an artificial intelligence and machine learning engine configured to process glucose data and generate therapy recommendations, and wireless communication circuitry configured to enable communication between the glucose sensor, the programming device, and the first and second electrical neuroregulators, wherein the high frequency alternating current conduction block and the low frequency electrical stimulation are applied cooperatively to regulate blood glucose levels.

[0201] Clause 80: The neuromodulation system of Clause 79, wherein the high frequency alternating current conduction block has frequencies ranging from 200 Hz to 100 kHz.Attorney Docket No. 40759.0090-RESH-022-01WG

[0202] Clause 81: The neuromodulation system of Clause 80, wherein the high frequency alternating current conduction block has frequencies of approximately 5000 Hz.

[0203] Clause 82: The neuromodulation system of any of Clauses 80-81, wherein the high frequency alternating current conduction block has frequencies ranging from 1000 Hz to 50000 Hz.

[0204] Clause 83: The neuromodulation system of Clause 79, wherein the low frequency electrical stimulation has frequencies ranging from 0.01 Hz to 199 Hz.

[0205] Clause 84: The neuromodulation system of Clause 83, wherein the low frequency electrical stimulation has a frequency of approximately 1 Hz.

[0206] Clause 85: The neuromodulation system of any of Clauses 83-84, wherein the low frequency electrical stimulation has frequencies of approximately 10 Hz, 20 Hz, 50 Hz, or 100 Hz.

[0207] Clause 86: The neuromodulation system of Clause 79, wherein the first electrical neuroregulator delivers current amplitudes ranging from 0.25 mA to 20 mA.

[0208] Clause 87: The neuromodulation system of Clause 86, wherein the first electrical neuroregulator delivers a current amplitude of approximately 12 mA.

[0209] Clause 88: The neuromodulation system of Clause 79, wherein the second electrical neuroregulator delivers current amplitudes ranging from 0.25 mA to 20 mA.

[0210] Clause 89: The neuromodulation system of Clause 88, wherein the second electrical neuroregulator delivers a current amplitude of approximately 9 mA.Attorney Docket No. 40759.0090-RESH-022-01WG

[0211] Clause 90: The neuromodulation system of any of Clauses 86-89, wherein the first electrical neuroregulator and the second electrical neuroregulator deliver current amplitudes ranging from 0.25 mA to 100 mA for specialized applications.

[0212] Clause 91: The neuromodulation system of Clause 79, wherein the glucose sensor is implantable.

[0213] Clause 92: The neuromodulation system of Clause 91, wherein the glucose sensor has a temporal resolution ranging from about 1 second to about 10 minutes.

[0214] Clause 93: The neuromodulation system of any of Clauses 91-92, wherein the glucose sensor has a temporal resolution with sub-second sampling rates for critical applications.

[0215] Clause 94: The neuromodulation system of Clause 79, wherein the glucose sensor is configured to communicate directly with the first and second electrical neuroregulators in response to detecting a hypoglycemic state.

[0216] Clause 95: The neuromodulation system of Clause 79, wherein the external programming device comprises a smart phone.

[0217] Clause 96: The neuromodulation system of Clause 95, wherein the external programming device comprises a handheld device, tablet computer, or laptop computer.

[0218] Clause 97: The neuromodulation system of Clause 79, wherein the artificial intelligence and machine learning engine is configured to adjust therapy parameters based on glucose level changes over time.

[0219] Clause 98: The neuromodulation system of Clause 97, wherein the therapy parameters comprise at least one of turning stimulation or conduction block on or off, changing current amplitude, changing frequency, and changing signal patterns.Attorney Docket No. 40759.0090-RESH-022-01WG

[0220] Clause 99: The neuromodulation system of any of Clauses 97-98, wherein the artificial intelligence and machine learning engine employs neural networks, reinforcement learning, or Bayesian optimization algorithms.

[0221] Clause 100: The neuromodulation system of Clause 79, wherein the wireless communication circuitry comprises at least one of Bluetooth technology, radio frequency communication, WiFi communication, and sound communication.

[0222] Clause 101: The neuromodulation system of Clause 100, wherein the wireless communication circuitry further comprises near-field communication, infrared communication, or cellular communication.

[0223] Clause 102: The neuromodulation system of Clause 79, further comprising a clinic communication device configured to communicate with the external programming device.

[0224] Clause 103: The neuromodulation system of Clause 102, further comprising a central computer system configured to receive data from the clinic communication device and generate population-based treatment recommendations.

[0225] Clause 104: The neuromodulation system of Clause 103, wherein the central computer system comprises a HIPAA compliant system with blockchain protocol.

[0226] Clause 105: The neuromodulation system of Clause 79, wherein the first electrical neuroregulator comprises a rechargeable battery.

[0227] Clause 106: The neuromodulation system of Clause 105, wherein the second electrical neuroregulator comprises a rechargeable battery.

[0228] Clause 107: The neuromodulation system of any of Clauses 105-106, further comprising an external charger configured to wirelessly recharge the rechargeable batteries.Attorney Docket No. 40759.0090-RESH-022-01WG

[0229] Clause 108: The neuromodulation system of Clause 107, wherein the external charger comprises a transmit coil and the first and second electrical neuroregulators each comprise a receiving antenna.

[0230] Clause 109: The neuromodulation system of any of Clauses 107-108, wherein the external charger employs inductive charging, magnetic resonance charging, or ultrasonic power transfer.

[0231] Clause 110: The neuromodulation system of Clause 79, wherein the first electrical neuroregulator is configured to deliver square wave pulses with a pulse width of approximately 4 milliseconds.

[0232] Clause 111: The neuromodulation system of Clause 110, wherein the first electrical neuroregulator is configured to deliver sinusoidal waves, triangular waves, or exponential decay pulses.

[0233] Clause 112: The neuromodulation system of Clause 79, wherein the external programming device is configured to receive manual input data comprising at least one of food intake information, insulin injection information, and physical activity information.

[0234] Clause 113: The neuromodulation system of Clause 112, wherein the manual input data further comprises medication schedules, sleep patterns, stress levels, or environmental factors.

[0235] Clause 114: The neuromodulation system of Clause 79, wherein the artificial intelligence and machine learning engine operates on a time scale ranging from about 1 second to about 10 minutes to continuously optimize therapy parameters.Attorney Docket No. 40759.0090-RESH-022-01WG

[0236] Clause 115: The neuromodulation system of Clause 114, wherein the artificial intelligence and machine learning engine operates with real-time processing with millisecond response times for emergency situations.

[0237] Clause 116: The neuromodulation system of Clause 79, wherein the first electrical neuroregulator is configured to apply the high frequency alternating current conduction block to the hepatic nerve.

[0238] Clause 117: The neuromodulation system of Clause 116, wherein the second electrical neuroregulator is configured to apply the low frequency electrical stimulation to the celiac nerve.

[0239] Clause 118: The neuromodulation system of Clause 79, wherein the first electrical neuroregulator is configured to apply the high frequency alternating current conduction block to the anterior vagus nerve cranial to the branching point of the hepatic nerve.

[0240] Clause 119: The neuromodulation system of Clause 118, wherein the second electrical neuroregulator is configured to apply the low frequency electrical stimulation to the posterior vagus nerve cranial to the branching point of the celiac nerve.

[0241] Clause 120: A closed-loop neuromodulation system comprising artificial intelligence and machine learning, comprising a plurality of neuroregulators, each configured to neuromodulate a different neuronal target selected from sympathetic nerves, parasympathetic nerves, central nervous system tissue, and smooth muscle, at least one bio-feedback sensor configured to monitor a physiological state, a programming device comprising an artificial intelligence and machine learning algorithm configured to analyze data from the at least one biofeedback sensor and generate predictive recommendations for neuromodulation parameters.Attorney Docket No. 40759.0090-RESH-022-01WGcommunication circuitry configured to enable data exchange between the at least one biofeedback sensor, the programming device, and the plurality of neuroregulators, and a central computer system configured to receive population data and enhance the artificial intelligence and machine learning algorithm using population trends, wherein the plurality of neuroregulators operate cooperatively based on the predictive recommendations to treat a medical condition.

[0242] Clause 121: The closed-loop neuromodulation system of Clause 120, wherein the plurality of neuroregulators comprises at least a first neuroregulator configured to neuromodulate a first neuronal target and a second neuroregulator configured to neuromodulate a second neuronal target.

[0243] Clause 122: The closed-loop neuromodulation system of Clause 121, wherein the first neuroregulator is configured to apply a stimulation to the first neuronal target and the second neuroregulator is configured to apply a conduction block to the second neuronal target.

[0244] Clause 123: The closed-loop neuromodulation system of Clause 122, wherein the first neuronal target comprises a celiac nerve and the second neuronal target comprises a hepatic nerve.

[0245] Clause 124: The closed-loop neuromodulation system of any of Clauses 121-123, wherein the plurality of neuroregulators comprises at least three neuroregulators, each configured to neuromodulate a different neuronal target.

[0246] Clause 125: The closed-loop neuromodulation system of Clause 120, wherein the plurality of neuroregulators comprises electrical neuroregulators configured to deliver electrical pulses with current amplitudes ranging from 0.25 mA to 20 mA.Attorney Docket No. 40759.0090-RESH-022-01WG

[0247] Clause 126: The closed-loop neuromodulation system of Clause 125, wherein the electrical neuroregulators are configured to deliver high frequency alternating current with frequencies ranging from 200 Hz to 100 kHz for conduction blocking.

[0248] Clause 127: The closed-loop neuromodulation system of Clause 126, wherein the electrical neuroregulators are configured to deliver low frequency signals with frequencies ranging from 0.01 Hz to 199 Hz for stimulation.

[0249] Clause 128: The closed-loop neuromodulation system of Clause 120, wherein at least one of the plurality of neuroregulators comprises a non-electrical neuroregulator selected from chemical agent stimulators, chemical conduction blocks, and optogenetic neuroregulators.

[0250] Clause 129: The closed-loop neuromodulation system of Clause 128, wherein the non-electrical neuroregulator comprises magnetic stimulators, ultrasonic stimulators, or thermal stimulators.

[0251] Clause 130: The closed-loop neuromodulation system of Clause 120, wherein the at least one bio-feedback sensor comprises a glucose sensor configured to monitor blood glucose levels.

[0252] Clause 131: The closed-loop neuromodulation system of Clause 130, wherein the glucose sensor is implantable and has a temporal resolution ranging from about 1 second to about 10 minutes.

[0253] Clause 132: The closed-loop neuromodulation system of Clause 131, wherein the glucose sensor is configured to communicate directly with the plurality of neuroregulators in response to detecting a hypoglycemic state.Attorney Docket No. 40759.0090-RESH-022-01WG

[0254] Clause 133: The closed-loop neuromodulation system of any of Clauses 130-132, wherein the at least one bio-feedback sensor comprises multiple sensor types selected from biochemical sensors, mechanical sensors, electrical sensors, optical sensors, and thermal sensors.

[0255] Clause 134: The closed-loop neuromodulation system of Clause 120, wherein the programming device comprises a smart phone or handheld device.

[0256] Clause 135: The closed-loop neuromodulation system of Clause 134, wherein the programming device comprises a tablet computer, laptop computer, or desktop computer.

[0257] Clause 136: The closed-loop neuromodulation system of Clause 120, wherein the artificial intelligence and machine learning algorithm is configured to generate predictive recommendations comprising at least one of turning stimulation or conduction block on or off, changing current amplitude, changing frequency, and changing signal patterns.

[0258] Clause 137: The closed-loop neuromodulation system of Clause 136, wherein the artificial intelligence and machine learning algorithm operates on a time scale ranging from about 1 second to about 10 minutes to continuously optimize the neuromodulation parameters.

[0259] Clause 138: The closed-loop neuromodulation system of any of Clauses 136-137, wherein the artificial intelligence and machine learning algorithm employs neural networks, deep learning, reinforcement learning, or ensemble methods.

[0260] Clause 139: The closed-loop neuromodulation system of Clause 120, wherein the communication circuitry comprises at least one of Bluetooth technology, radio frequency communication, WiFi communication, and sound communication.Attorney Docket No. 40759.0090-RESH-022-01WG

[0261] Clause 140: The closed-loop neuromodulation system of Clause 139, wherein the communication circuitry further comprises near-field communication, infrared communication, cellular communication, or satellite communication.

[0262] Clause 141: The closed-loop neuromodulation system of Clause 120, wherein the central computer system comprises a HIPAA compliant system with blockchain protocol.

[0263] Clause 142: The closed-loop neuromodulation system of Clause 141, wherein the central computer system implements federated learning approaches for knowledge sharing without exposing individual patient data.

[0264] Clause 143: The closed-loop neuromodulation system of any of Clauses 141-142, wherein the central computer system is configured to track population trends in physiological parameters over time and generate enhanced treatment recommendations.

[0265] Clause 144: The closed-loop neuromodulation system of Clause 120, further comprising a clinic communication device configured to communicate with the programming device and the central computer system.

[0266] Clause 145: The closed-loop neuromodulation system of Clause 144, wherein the clinic communication device is configured to receive population-based treatment recommendations from the central computer system and transmit therapy modifications to the programming device.

[0267] Clause 146: The closed-loop neuromodulation system of Clause 120, wherein the medical condition comprises at least one of type 2 diabetes, obesity, refractory hypertension, pancreatitis, refractory asthma, urinary incontinence, erectile dysfunction, chronic heart failure.Attorney Docket No. 40759.0090-RESH-022-01WGchronic inflammatory disease, Parkinson's disease, refractory depression, refractory epilepsy, and migraine headaches.

[0268] Clause 147: The closed-loop neuromodulation system of Clause 146, wherein the medical condition comprises chronic pain syndromes, gastroparesis, irritable bowel syndrome, sleep disorders, anxiety disorders, or addiction disorders.

[0269] Clause 148: The closed-loop neuromodulation system of Clause 120, wherein the plurality of neuroregulators are configured to neuromodulate the different neuronal targets with different start times, different end times, and different durations.

[0270] Clause 149: The closed-loop neuromodulation system of Clause 148, wherein the plurality of neuroregulators are configured to neuromodulate the different neuronal targets during overlapping time periods.

[0271] Clause 150: The closed-loop neuromodulation system of any of Clauses 148-149, wherein the plurality of neuroregulators are configured to neuromodulate the different neuronal targets during distinct non-overlapping time periods.

[0272] Clause 151: The closed-loop neuromodulation system of Clause 120, wherein the plurality of neuroregulators comprises rechargeable batteries.

[0273] Clause 152: The closed-loop neuromodulation system of Clause 151, further comprising an external charger configured to wirelessly recharge the rechargeable batteries through RF power transmission.

[0274] Clause 153: The closed-loop neuromodulation system of Clause 152, wherein the external charger comprises a transmit coil and the plurality of neuroregulators each comprise a receiving antenna.Attorney Docket No. 40759.0090-RESH-022-01WG

[0275] Clause 154: The closed-loop neuromodulation system of any of Clauses 151-153, wherein the plurality of neuroregulators comprises primary batteries, supercapacitors, or energy harvesting systems.

[0276] Clause 155: The closed-loop neuromodulation system of Clause 120, wherein the programming device is configured to receive manual input data comprising at least one of food intake information, insulin injection information, and physical activity information.

[0277] Clause 156: The closed-loop neuromodulation system of Clause 155, wherein the manual input data further comprises medication schedules, sleep patterns, stress levels, environmental factors, or other lifestyle parameters.

[0278] Clause 157: The closed-loop neuromodulation system of Clause 120, wherein the artificial intelligence and machine learning algorithm is configured to adjust the predictive recommendations over time based on body adaptation to the neuromodulation.

[0279] Clause 158: The closed-loop neuromodulation system of Clause 157, wherein the artificial intelligence and machine learning algorithm employs online learning algorithms, active learning approaches, or multi-task learning frameworks.

[0280] Clause 159: The closed-loop neuromodulation system of Clause 120, wherein the plurality of neuroregulators comprises at least three neuroregulators, each configured to neuromodulate a different neuronal target selected from sympathetic nerves, parasympathetic nerves, central nervous system tissue, and smooth muscle.

[0281] Clause 160: The closed-loop neuromodulation system of Clause 159, wherein the at least three neuroregulators are configured to operate cooperatively in a coordinated manner to provide enhanced therapeutic efficacy compared to individual neuroregulator operation.Attorney Docket No. 40759.0090-RESH-022-01WG

[0282] Clause 161: A multi-target neuromodulation apparatus, comprising a first neuromodulation component selected from the group consisting of electrical stimulators, electrical conduction blocks, chemical agent stimulators, chemical conduction blocks, physical ablation devices, and optogenetic modulators, wherein the first neuromodulation component is configured to modulate a first target selected from sympathetic nerves, parasympathetic nerves, central nervous system tissue, and smooth muscle, a second neuromodulation component selected from the group consisting of electrical stimulators, electrical conduction blocks, chemical agent stimulators, chemical conduction blocks, physical ablation devices, and optogenetic modulators, wherein the second neuromodulation component is configured to modulate a second target different from the first target, a sensor system configured to provide bio-feedback regarding a medical condition, an artificial intelligence and machine learning processing unit configured to analyze the bio-feedback and determine optimal modulation parameters, and a control system configured to coordinate the first neuromodulation component and the second neuromodulation component based on the optimal modulation parameters, wherein the modulation of the first target and the second target work cooperatively to treat the medical condition.

[0283] Clause 162: The multi-target neuromodulation apparatus of Clause 161, wherein the first neuromodulation component comprises an electrical stimulator and the second neuromodulation component comprises an electrical conduction block.

[0284] Clause 163: The multi-target neuromodulation apparatus of Clause 162, wherein the electrical stimulator is configured to deliver low frequency signals with frequencies ranging from 0.01 Hz to 199 Hz.Attorney Docket No. 40759.0090-RESH-022-01WG

[0285] Clause 164: The multi-target neuromodulation apparatus of Clause 163, wherein the electrical conduction block is configured to deliver high frequency alternating current with frequencies ranging from 200 Hz to 100 kHz.

[0286] Clause 165: The multi-target neuromodulation apparatus of any of Clauses 162-164, wherein the first neuromodulation component and the second neuromodulation component comprise hybrid neuroregulators that combine electrical and non-electrical modalities.

[0287] Clause 166: The multi-target neuromodulation apparatus of Clause 161, wherein the first target comprises a celiac nerve and the second target comprises a hepatic nerve.

[0288] Clause 167: The multi-target neuromodulation apparatus of Clause 166, wherein the medical condition is type 2 diabetes.

[0289] Clause 168: The multi-target neuromodulation apparatus of any of Clauses 166-167, wherein the first target comprises specific nerve branches, nerve plexuses, or individual nerve fascicles.

[0290] Clause 169: The multi-target neuromodulation apparatus of Clause 161, wherein the first neuromodulation component and the second neuromodulation component are configured to deliver current amplitudes ranging from 0.25 mA to 20 mA.

[0291] Clause 170: The multi-target neuromodulation apparatus of Clause 169, wherein the first neuromodulation component and the second neuromodulation component are configured to deliver current amplitudes ranging from 0.25 mA to 100 mA for specialized applications.

[0292] Clause 171: The multi-target neuromodulation apparatus of Clause 161, wherein the sensor system comprises a glucose sensor configured to monitor blood glucose levels.Attorney Docket No. 40759.0090-RESH-022-01WG

[0293] Clause 172: The multi-target neuromodulation apparatus of Clause 171, wherein the glucose sensor is implantable and has a temporal resolution ranging from about 1 second to about 10 minutes.

[0294] Clause 173: The multi-target neuromodulation apparatus of Clause 172, wherein the glucose sensor is configured to communicate directly with the control system in response to detecting a hypoglycemic state.

[0295] Clause 174: The multi-target neuromodulation apparatus of any of Clauses 171-173, wherein the sensor system comprises multiple sensor types, sensor arrays, or distributed sensing networks.

[0296] Clause 175: The multi-target neuromodulation apparatus of Clause 161, wherein the artificial intelligence and machine learning processing unit is configured to generate optimal modulation parameters comprising at least one of turning stimulation or conduction block on or off, changing current amplitude, changing frequency, and changing signal patterns.

[0297] Clause 176: The multi-target neuromodulation apparatus of Clause 175, wherein the artificial intelligence and machine learning processing unit operates on a time scale ranging from about 1 second to about 10 minutes to continuously optimize the modulation parameters.

[0298] Clause 177: The multi-target neuromodulation apparatus of any of Clauses 175-176, wherein the artificial intelligence and machine learning processing unit employs multi- objective optimization, game-theoretic approaches, or distributed consensus algorithms.

[0299] Clause 178: The multi-target neuromodulation apparatus of Clause 161, wherein the control system comprises wireless communication circuitry configured to enable data exchangeAttorney Docket No. 40759.0090-RESH-022-01WGbetween the sensor system, the artificial intelligence and machine learning processing unit, and the first and second neuromodulation components.

[0300] Clause 179: The multi-target neuromodulation apparatus of Clause 178, wherein the wireless communication circuitry comprises at least one of Bluetooth technology, radio frequency communication, WiFi communication, and sound communication.

[0301] Clause 180: The multi-target neuromodulation apparatus of any of Clauses 178-179, wherein the wireless communication circuitry employs mesh networking, adaptive protocols, or error correction algorithms.

[0302] Clause 181: The multi-target neuromodulation apparatus of Clause 161, wherein the first neuromodulation component comprises a chemical agent stimulator configured to deliver therapeutic agents to the first target.

[0303] Clause 182: The multi-target neuromodulation apparatus of Clause 181, wherein the second neuromodulation component comprises an electrical conduction block configured to block neural conduction in the second target.

[0304] Clause 183: The multi-target neuromodulation apparatus of any of Clauses 181-182, wherein the chemical agent stimulator delivers neurotransmitters, neuromodulators, or pharmaceutical compounds.

[0305] Clause 184: The multi-target neuromodulation apparatus of Clause 161, wherein the first neuromodulation component comprises an optogenetic modulator configured to deliver light-activated neural modulation to the first target.Attorney Docket No. 40759.0090-RESH-022-01WG

[0306] Clause 185: The multi-target neuromodulation apparatus of Clause 184, wherein the optogenetic modulator is configured to deliver an inhibitory opsin or excitatory opsin that is activated with high frequency light.

[0307] Clause 186: The multi-target neuromodulation apparatus of any of Clauses 184-185, wherein the optogenetic modulator employs fiber optic implants, LED arrays, laser diodes, or wireless optogenetic devices.

[0308] Clause 187: The multi-target neuromodulation apparatus of Clause 161, wherein the control system is configured to coordinate the first neuromodulation component and the second neuromodulation component with different start times, different end times, and different durations.

[0309] Clause 188: The multi-target neuromodulation apparatus of Clause 187, wherein the control system is configured to coordinate the first neuromodulation component and the second neuromodulation component during overlapping time periods.

[0310] Clause 189: The multi-target neuromodulation apparatus of any of Clauses 187-188, wherein the control system employs synchronization protocols, interference mitigation strategies, or optimization algorithms.

[0311] Clause 190: The multi-target neuromodulation apparatus of Clause 161, wherein the medical condition comprises at least one of type 2 diabetes, obesity, refractory hypertension, pancreatitis, refractory asthma, urinary incontinence, erectile dysfunction, chronic heart failure, chronic inflammatory disease, Parkinson's disease, refractory depression, refractory epilepsy, and migraine headaches.Attorney Docket No. 40759.0090-RESH-022-01WG

[0312] Clause 191: The multi-target neuromodulation apparatus of Clause 190, wherein the medical condition comprises chronic pain syndromes, gastroparesis, irritable bowel syndrome, sleep disorders, anxiety disorders, post-traumatic stress disorder, addiction disorders, eating disorders, chronic fatigue syndrome, or fibromyalgia.

[0313] Clause 192: The multi-target neuromodulation apparatus of Clause 161, further comprising a programming device configured to interface with the artificial intelligence and machine learning processing unit.

[0314] Clause 193: The multi-target neuromodulation apparatus of Clause 192, wherein the programming device comprises a smart phone or handheld device configured to receive manual input data comprising at least one of food intake information, insulin injection information, and physical activity information.

[0315] Clause 194: The multi-target neuromodulation apparatus of Clause 193, wherein the programming device is configured to communicate with a clinic communication device.

[0316] Clause 195: The multi-target neuromodulation apparatus of Clause 194, further comprising a central computer system configured to receive population data from the clinic communication device and enhance the artificial intelligence and machine learning processing unit using population trends.

[0317] Clause 196: The multi-target neuromodulation apparatus of Clause 195, wherein the central computer system comprises a HIPAA compliant system with blockchain protocol.

[0318] Clause 197: The multi-target neuromodulation apparatus of any of Clauses 195-196, wherein the central computer system employs distributed computing, parallel processing, or specialized Al hardware.Attorney Docket No. 40759.0090-RESH-022-01WG

[0319] Clause 198: The multi-target neuromodulation apparatus of Clause 161, wherein the first neuromodulation component and the second neuromodulation component each comprise rechargeable batteries.

[0320] Clause 199: The multi-target neuromodulation apparatus of Clause 198, further comprising an external charger configured to wirelessly recharge the rechargeable batteries through RF power transmission.

[0321] Clause 200: The multi-target neuromodulation apparatus of Clause 199, wherein the external charger comprises a transmit coil and the first and second neuromodulation components each comprise a receiving antenna.

[0322] Clause 201: The multi-target neuromodulation apparatus of any of Clauses 198-200, wherein the first neuromodulation component and the second neuromodulation component comprise distributed power systems, backup batteries, or energy harvesting capabilities.

[0323] Clause 202: The multi-target neuromodulation apparatus of Clause 161, wherein the artificial intelligence and machine learning processing unit is configured to adjust the optimal modulation parameters over time based on body adaptation to the modulation of the first target and the second target.

[0324] Clause 203: The multi-target neuromodulation apparatus of Clause 202, wherein the artificial intelligence and machine learning processing unit employs multi-agent learning frameworks, hierarchical reinforcement learning, or curriculum learning approaches.

[0325] Clause 204: The multi-target neuromodulation apparatus of any of Clauses 202-203, wherein the artificial intelligence and machine learning processing unit implements cloud-based processing, edge computing, or hybrid approaches.Attorney Docket No. 40759.0090-RESH-022-01WG

[0326] Clause 205: The multi-target neuromodulation apparatus of Clause 161, further comprising third and fourth neuromodulation components configured to modulate third and fourth targets different from the first and second targets.

[0327] Clause 206: The multi-target neuromodulation apparatus of Clause 205, wherein the control system is configured to coordinate all neuromodulation components in a synchronized, sequential, or adaptive manner.

[0328] Clause 207: The multi-target neuromodulation apparatus of any of Clauses 205-206, wherein the artificial intelligence and machine learning processing unit employs advanced coordination algorithms for managing multiple neuromodulation components.

[0329] Clause 208: The multi-target neuromodulation apparatus of Clause 161, wherein the control system incorporates safety monitoring, fault detection, or emergency shutdown capabilities.

[0330] Clause 209: The multi-target neuromodulation apparatus of Clause 208, wherein the safety monitoring continuously checks for adverse effects, unexpected physiological responses, or system malfunctions.

[0331] Clause 210: The multi-target neuromodulation apparatus of any of Clauses 208-209, wherein the emergency shutdown capabilities can safely disable the system during dangerous conditions, device malfunctions, or emergency medical situations.

[0332] Although exemplary embodiments of this invention have been described, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments and methods without materially departing from the novel teachings and advantages of this invention. Accordingly, all such modifications are intended to be included within the scopeAttorney Docket No. 40759.0090-RESH-022-01WGof this invention as defined in the claims. The invention is defined by the following claims, with equivalents of the claims to be included therein.

Claims

Attorney Docket No. 40759.0090-RESH-022-01WGCLAIMS1. A system for regulating at least two neuronal targets using Artificial Intelligence and Machine Learning, comprising:a first pulse generator that applies a stimulation or a conduction block to a first neuronal target; anda second pulse generator that applies a stimulation or a conduction block to a second neuronal target that is different from the first neuronal target, wherein the stimulation or conduction block of the first neuronal target works in cooperation with the stimulation or conduction block of the second neuronal target to treat a medical condition based on the bio¬ feedback generated by a ML and Al algorithm.

2. The system of claim 1, further comprising a programing device that commutates with a glucose sensor (inside or outside the body) in which the programming device contains a ML and Al learning system / algorithm to communicate with a neuroregulator device to optimize therapy.

3. The system of claim 1, wherein the programming device communicates with a healthcare provider to receive instructions to change therapy.

4. The system of claim 1, wherein patient population information is processed through an Al and ML platform on a central computer system / repository / on a cloud / server to predict recommendations for treatment modulations.

5. The system of claim 1, wherein the first neuronal target and the second neuronal target each comprise one or more sympathetic nerves.Attorney Docket No. 40759.0090-RESH-022-01WG6. The system of claim 1, wherein the first neuronal target and the second neuronal target each comprise one or more parasympathetic nerves.

7. The system of claim 1, wherein the first neuronal target comprises one or more sympathetic nerves and wherein the second neuronal target comprises one or more parasympathetic nerves.

8. The system of claim 1, wherein the stimulation or conduction block applied to the first neuronal target is of a different duration than that stimulation or conduction block applied to the second neuronal target.

9. The system of claim 1, wherein the first pulse generator applies the stimulation to the first neuronal target and the second pulse generator applies the stimulation to the second neuronal target.

10. The system of claim 1, wherein the first pulse generator applies the conduction block to the first neuronal target and wherein the second pulse generator applies the conduction block to the second neuronal target.

11. The system of claim 1, wherein the first pulse generator applies the conduction block to the first neuronal target and wherein the second pulse generator applies the stimulation to the second neuronal target.

12. The system of claim 1, wherein the medical condition can consist of, heart failure, inflammation, hypertension, pancreatitis, refractory asthma, urinary incontinence, erectileAttorney Docket No. 40759.0090-RESH-022-01WGdysfunction, type II diabetes, obesity. Parkinson’s disease, refractory depression, refectory epilepsy and cluster headaches.

13. The system of claim 1, in which there is an addiction of a sensor (with a sample rate in a range of 1 second to 10 min) unit that measures the state of a medical condition14. The system of claim 13, in which the sensor communicates with the neuroregulator for an immediate change in therapy for a medical emergency such as a hypoglycemic state15. The system of claim 14, in which the sensor communicates with an external programmable unit, such as a smart device16. The system of claim 15, in which the communication occurs though Bluetooth technology, radio frequency, WIFI or sound17. The system of claim 15, wherein the external programmable unit communicates with ML and Al to predict the optimal parameters for a neuroregulator to deliver to a target nerve and in which the external programmable unit communicates with the neuroregulator and / or external programmable unit for the treatment of a medical condition 18. The system of claim 2, wherein the modulated therapy can turn block or stimulation on or off, to change the current amplitude of HFAC block and stimulation (0.25 to 20 mA or 0.25 to 20 V), the frequencies of block (200 Hz-10k Hz) and stimulation (0.1-199 Hz), changing signal patterns of the block and stimulation (such as bursting), changing to only block or only stimulation, reversing block to stimulation and / or stimulation to block, the time of day to apply signals, or turning off both block and stimulation entirely.Attorney Docket No. 40759.0090-RESH-022-01WG19. The system of claim 11, wherein the first neuronal target comprises a hepatic nerve, or the anterior vagus nerve cranial to the branching point of the hepatic nerve, and the second neuronal target comprises the celiac nerve, or the posterior vagus nerve cranial to the branching point of the celiac nerve.

20. The system of claim 1, wherein the first pulse generator is an electrical neuroregulator.

21. The system of claim 20, wherein the second pulse generator is an electrical neuroregulator.

22. The system wherein neuromodulation is achieved by a non-electrical neuroregulator such as chemical or Optogenetic neuro-modulation.

23. A method for treating a medical condition, comprising Artificial Intelligence and Machine Learning, comprising:neuromodulating, with a stimulation or a conduction block, a first neuronal target; andneuromodulating, with a stimulation or a conduction block, a second neuronal target, wherein the second neuronal target is different from the first neuronal target, and wherein the neuromodulation of the first neuronal target works in cooperation with the neuromodulation of the second neuronal target to treat a medical condition based on the bio-feedback generated by a ML and Al algorithm.Attorney Docket No. 40759.0090-RESH-022-01WG24. The method of claim 23, in which there is a programing device that commutates with a glucose sensor (inside or outside the body) in which the programming device contains a ML and Al learning system / algorithm to communicate with a neuroregulator device to optimize therapy.

25. The method of claim 23, in which the programming device communicates with a healthcare provider to receive instructions to change therapy.

26. The method of claim 23, in which patient population information is processed through an Al and ML platform on a central computer system / repository / on a cloud / server to predict recommendations for treatment modulations.

27. The method of claim 23, wherein the first neuronal target and the second neuronal target each comprise one or more sympathetic nerves.

28. The method of claim 23, wherein the first neuronal target and the second neuronal target each comprise one or more parasympathetic nerves.

29. The method of claim 23, wherein the first neuronal target comprises one or more sympathetic nerves and wherein the second neuronal target comprises one or more parasympathetic nerves.

30. The method of claim 23, wherein neuromodulating the first neuronal target comprises neuromodulating with the stimulation and wherein neuromodulating the second neuronal target comprises neuromodulating with the stimulation.Attorney Docket No. 40759.0090-RESH-022-01WG.

31. The method of claim 23, wherein neuromodulating the first neuronal target comprises neuromodulating with the stimulation and wherein neuromodulating the second neuronal target comprises neuromodulating with the conduction block.

32. The method of claim 23, wherein neuromodulating the first neuronal target comprises neuromodulating with the conduction block and wherein neuromodulating the second neuronal target comprises neuromodulating with the conduction block.

33. The method of claim 23, wherein the medical condition can consist of, heart failure, inflammation, hypertension, pancreatitis, refractory asthma, urinary incontinence, erectile dysfunction, type II diabetes, obesity, Parkinson’s disease, refractory depression, refectory epilepsy and cluster headaches.

34. The method of claim 23, in which there is an addition of a sensor unit that measures the state of a medical condition35. The method of claim 23, in which there is an addition of a sensor unit that measures blood glucose36. The method of claim 34, in which the sensor (with a sample rate in a range of 1 second to 10 min) communicates with the neuroregulator for an immediate change in therapy for a medical emergency such as a hypoglycemic state36. The method of claim 35, in which the sensor communicates with an external programmable unit such as a smart phoneAttorney Docket No. 40759.0090-RESH-022-01WG37. The method of 36, in which the communication occurs though Bluetooth technology, radio frequency, WIFI or sound38. The method of claim 36, in which the external programmable unit communicates with ML and Al to predict the optimal parameters for a neuroregulator to deliver to a target nerve and in which the external programmable unit communicates with the neuroregulator and / or external programmable unit for the treatment of a medical condition39. The method of claim 24, in which the modulated therapy causes to turn block or stimulation on or off, to change the current amplitude of HFAC block and stimulation (0.25 mA-20 mA or 0.25-20 V), the frequencies of block (200 Hz-10 K Hz) and stimulation (0.01-199 Hz), changing signal patterns of the block and stimulation (such as bursting), changing to only block or only stimulation, reversing block to stimulation and / or stimulation to block, the time of day to apply signals, or turning off both block and stimulation entirel.

Images