Infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury

Abstract

The invention relates to the technical field of infrared radiation therapy parameter control, in particular to an infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury. The system monitors stress events in real time based on spasm onset impact and a tip perfusion index. And quantizing the rosemary-sympathetic response delay of the patient, further analyzing the most unstable moment of the center rate dynamics in the event time period to be analyzed, and quantizing the response-instability time difference. And according to the response-instability time difference and the rosemary-sympathetic response delay identification, accurately identifying the event category. And an accurate control instruction is fed back to the event type. According to the method, two cause identification problems with unclear macroscopic physiological data representation are converted into deterministic problems with clear sequence in time sequence, and the excitation event is accurately and effectively identified by combining the obtained time domain features, so that the control accuracy in the infrared light treatment process is ensured, and the medical accident risk is reduced.

Description

Technical Field

[0001] This invention relates to the field of infrared radiotherapy parameter control technology, specifically to an infrared irradiation parameter optimization system for neuropathic pain after spinal cord injury. Background Technology

[0002] Neuropathic pain following spinal cord injury is a common clinical complication that severely impacts patients' quality of life. Infrared light therapy, as a non-invasive treatment, has shown potential in alleviating this type of pain. However, current treatment parameters (such as power density and irradiation time) largely rely on general guidelines or the operator's clinical experience, lacking real-time, dynamic adjustment tailored to the individual patient's condition.

[0003] This problem is particularly prominent in patients with high-level spinal cord injuries at or above the 6th thoracic segment. These patients often suffer from a life-threatening complication—autonomic dysreflexia (AD). Any stimulus below the injury level can trigger AD, leading to a sudden increase in blood pressure, heart rate disturbances, and other consequences. During infrared light therapy, there are at least two potential triggers for AD: first, the photothermal effect of the infrared light itself; if the parameters are not set correctly, the accumulated heat may constitute a harmful stimulus; second, the unpredictable, sudden muscle spasms that often coexist in these patients are themselves a strong trigger for AD.

[0004] In current clinical practice, when abnormalities are detected in early Alzheimer's disease (AD) physiological indicators such as heart rate and blood pressure, operators often struggle to accurately distinguish the root cause of risk events using only non-invasive and simple monitoring methods. Specifically, they cannot determine whether the risk is caused by unsafe infrared light parameters or by a single, independent spastic event unrelated to treatment parameters. This causal confusion leads to difficulties in intervention decisions: misinterpreting physiological fluctuations caused by spasticity as a treatment parameter problem and interrupting treatment reduces efficacy; conversely, failing to identify and adjust unsafe treatment parameters in a timely manner may harm the patient. Existing technologies rely on the operator's subjective experience or introduce complex equipment to assist in spasticity monitoring, increasing system complexity and operating costs. Summary of the Invention

[0005] To address the technical problem of failing to promptly and effectively identify the cause of a stress event during infrared light irradiation therapy for spinal cord injury patients, thus hindering effective control measures, this invention aims to provide an infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury. The specific technical solution adopted is as follows: This invention proposes an infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury, the system comprising: The instantaneous state characterization module is used to extract the patient's peripheral perfusion index, heart rate variability, and spasm initiation impact at various moments during infrared light irradiation therapy, based on the patient's heart rate signal and trunk muscle movement signal. The physiological stress event detection module is used to detect stress events by prioritizing the spastic initiation impact as the first priority and the peripheral perfusion index as the second priority. Based on the time period, heart rate variability, and chaotic distribution of the peripheral perfusion index included in the stress event, unreliable events and events to be analyzed are classified. A multi-domain feature decoupling module is used to divide the sympathetic nerve response time period and the vagal nerve response time period in the event to be analyzed, and obtain the vagal-sympathetic response delay; a sliding window is set in the time period of the event to be analyzed, the variability fluctuation index of the heart rate in the sliding window is calculated, and the moment of the most unstable heart rate dynamics in the event to be analyzed is calculated based on the variability fluctuation index; the time interval between the start time of the sympathetic nerve response time period and the moment of the most unstable heart rate dynamics is used as the response-instability time difference; the event category of the event to be analyzed is identified based on the response-instability time difference and the vagal-sympathetic response delay; the event category includes muscle spasm, improper infrared parameters, and unreliable events; The infrared parameter control module is used to provide corresponding control commands based on the event category of the stress event.

[0006] Furthermore, the motion signal is a triaxial acceleration signal. For each moment, the L2 norm of the triaxial acceleration is calculated, and the first derivative of the L2 norm signal at each moment is used as the spasm initiation impact.

[0007] Furthermore, the method for detecting the stress event includes: For each moment, the initial spasm impact is compared with a preset impact threshold. If the initial spasm impact is greater than the preset impact threshold, the corresponding moment is directly used as the reference moment for the stress event. Otherwise, the peripheral perfusion index decay between the current moment and the previous moment is obtained. If the peripheral perfusion index decay is greater than a preset difference threshold, the current moment is used as the reference moment for the stress event. Otherwise, the current moment is ignored. Centered on the reference time of the stress event, the time period corresponding to the stress event is divided according to the preset size division rules.

[0008] Furthermore, the method for distinguishing between unreliable events and events to be analyzed includes: An initial time period is defined within the time frame of the stress event. The standard deviations of the peripheral perfusion index and heart rate variability within the initial time period are statistically analyzed. If either the standard deviation of the peripheral perfusion index or the standard deviation of heart rate variability exceeds the preset standard deviation threshold for the corresponding dimension, the corresponding stress event is determined to be an unreliable event. Emergency events other than unreliable events are considered events to be analyzed.

[0009] Furthermore, the process of dividing the event to be analyzed into sympathetic nerve response time periods and vagus nerve response time periods includes: After the reference time of the stress event, the moment when the peripheral perfusion index first falls below the preset perfusion baseline value for the first consecutive preset number of times is taken as the starting point of the sympathetic nerve response; the moment when the peripheral perfusion index first recovers to the preset perfusion baseline value after the starting point of the sympathetic nerve response is taken as the ending point of the sympathetic nerve response; the time period between the starting point and the ending point of the sympathetic nerve response is the sympathetic nerve response time period. After the reference time of the emergency event, the moment when the heart rate variability is first continuously lower than the preset heart rate variability benchmark value for the second number of consecutive times is taken as the vagal nerve response start point; the moment when the heart rate variability first recovers to the preset perfusion benchmark value after the vagal nerve response start point is taken as the vagal nerve response end point; the time period between the vagal nerve response start point and the vagal nerve response end point is the sympathetic nerve response time period.

[0010] Furthermore, the method for obtaining the variability volatility index includes: Within a sliding window, a heart rate interval sequence is constructed. The Poincaré diagram analysis algorithm is used to obtain the vertical standard deviation and the projected standard deviation along the diagonal of the heart rate interval sequence. The ratio obtained by using the vertical standard deviation as the numerator and the projected standard deviation as the denominator is used as the heart rate variability short-to-long ratio. The heart rate variability short-to-long ratio is used as the variability fluctuation index.

[0011] Furthermore, the method for obtaining the most unstable moment in heart rate dynamics includes: Traverse all sliding windows and select the center point of the sliding window corresponding to the minimum value of the variability fluctuation index as the moment when the heart rate dynamics are most unstable.

[0012] Furthermore, the step of identifying the event category of the event to be analyzed based on the response-instability time difference and the vagal-sympathetic response delay includes: If the response-instability time difference is greater than a preset first time difference threshold, it is determined to be muscle spasm; if the response-instability time difference is not greater than the preset first time difference threshold, and the vagal-sympathetic response delay is greater than a preset second time difference threshold, it is determined to be improper infrared parameters; other events to be analyzed are unreliable events.

[0013] Furthermore, the step of feeding back the corresponding control command based on the event category of the stress event includes: If the event category is muscle spasm, a pause irradiation command is sent to the infrared light therapy device, and the patient's physiological information is fed back in real time; If the event category is an inappropriate infrared light parameter event, a parameter downgrade instruction will be sent to the infrared light therapy device. If the event category is uncertain, a stop irradiation command will be sent to the infrared light therapy device, and an alarm will be triggered.

[0014] Furthermore, the system also includes an event statistics comparison and analysis module. This module is used to create two independent parameter sets for each patient in local storage. The independent parameter sets include a patient-specific insecure parameter set and a patient-specific safe parameter set. The individualized insecure parameter set stores infrared information corresponding to events where infrared parameters are inappropriate. The individualized safe parameter set stores infrared information used during the complete treatment process when no stress events were reported. Before each new treatment process begins, the set infrared information is compared with the infrared information in the independent parameter sets, and the optimal infrared information is recommended based on the comparison results.

[0015] The present invention has the following beneficial effects: This invention acquires patients' heart rate signals and trunk muscle movement signals in real time. It utilizes peripheral perfusion index information to characterize macroscopic autonomic nervous system responses, spasticity initiation impacts to characterize intuitive muscle movement manifestations, and heart rate variability to characterize the stability of the myodynamic system. Therefore, this invention first monitors stress events in real time based on spasticity initiation impacts and peripheral perfusion index, prioritizing muscle movement manifestations as the first priority and neural responses as the second priority to ensure no stress event is missed. To avoid misidentification of stress events and the resulting treatment risks, this invention does not blindly perform coupling analysis on all stress events. It first divides stress events into unreliable events and events to be analyzed, and then performs coupling analysis on the driving sources of events with clear characteristics. This invention first quantifies the patient's vagal-sympathetic response delay, then analyzes the most unstable moment of myodynamics within the time period of the event to be analyzed, quantifying the response-instability time difference. This response-instability time difference transforms a difficult-to-distinguish physiological pattern recognition problem into a deterministic attribution problem based on a single, physiologically significant time feature. This feature characterizes the time offset between the starting point of the macroscopic response and the point where the internal dynamics reach their most unstable state. Because muscle spasms and improper infrared parameters elicit different responses from the patient's physiological structure, exhibiting significant temporal differences in both neural and physiological responses, this invention allows for accurate identification of event categories based on the response-instability time difference and vagal-sympathetic response delay. This enables the generation of precise control commands tailored to the event category. This invention transforms two causal identification problems, which are unclear in macroscopic physiological data representation, into deterministic problems with a clear temporal sequence. By combining the obtained temporal domain features, it accurately and effectively identifies stress events, ensuring accurate control during infrared light therapy and reducing the risk of medical accidents. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a block diagram of an infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury, provided as an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an infrared irradiation parameter optimization system for neuropathic pain after spinal cord injury proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0020] The following description, in conjunction with the accompanying drawings, details the specific scheme of the infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury provided by the present invention.

[0021] Please see Figure 1 The diagram illustrates a block diagram of an infrared irradiation parameter optimization system for neuropathic pain after spinal cord injury according to an embodiment of the present invention. The system includes: an instantaneous state characterization module 101, a physiological stress event detection module 102, a multi-domain feature decoupling module 103, and an infrared parameter control module 104.

[0022] In this embodiment of the invention, the instantaneous state characterization module 101 can directly call data from existing medical parameter acquisition devices. The data included in this embodiment includes heart rate signals and trunk muscle motion signals. The heart rate signal is acquired by a finger clip-on photoplethysmography (PPG) sensor worn on the patient's finger; the trunk muscle motion signals are acquired by a triaxial accelerometer fixed to the patient's trunk or wrist. Thus, a real-time heart rate signal and a real-time three-dimensional triaxial acceleration signal can be obtained. It should be noted that because the sampling frequencies of the heart rate signal and the motion signal may be different, in this embodiment, the sampling frequency of the heart rate signal is 100 Hz, and the sampling frequency of the motion signal is 50 Hz. Therefore, after obtaining the signal data, the motion signal needs to be upsampled using linear interpolation to obtain a heart rate data and a three-dimensional triaxial acceleration vector at each moment. It should be noted that linear interpolation algorithms are well-known to those skilled in the art. In other implementations of this embodiment, other interpolation algorithms with better shape preservation, such as cubic spline interpolation, can also be used, which will not be elaborated or limited here. The final instantaneous state characterization module can retrieve the patient's real-time and time-domain synchronized heart rate signal and trunk muscle motion signal for analysis.

[0023] The instantaneous state characterization module 101 is used to quantify the instantaneous state of the patient during treatment based on real-time physiological data. This embodiment of the invention employs three features to quantify the patient's instantaneous state, including the peripheral perfusion index, which is mainly represented by the ratio of the AC to DC components of the heart rate signal. This reflects the pulsatile blood flow perfusion level of the capillary bed at the fingertips, and its numerical changes are directly related to the sympathetic nerve-driven vasoconstriction or vasodilation state, which can be used for subsequent event identification and assessment. It also includes heart rate variability, which reflects the intensity of heart rate information changes within a local time period corresponding to a real-time moment. This feature quantifies the fluctuation of the heart rhythm in a very short time and can characterize the level of fine-tuning activity of the parasympathetic nervous system, i.e., the vagus nerve, on heart rate. Finally, it includes spastic initiation impact, an intuitive motion feature quantified by motion signals, which can quantify the intensity of sudden, basic impact movements generated in real time.

[0024] In this embodiment of the invention, the analysis frequency for each real-time moment is once every 200 milliseconds, meaning the time interval between adjacent moments is 200 milliseconds. It should be noted that because the peripheral perfusion index acquisition method requires a segment of heart rate signal for analysis, this invention sets a forward 2-second time window for each real-time moment. The endpoint of this time window is the real-time moment, and the peripheral perfusion index is analyzed within this time window. Similarly, heart rate variability mainly reflects the instability of heart rate, and its analysis is also less effective than analyzing information distribution within this time window.

[0025] It should be noted that, because heart rate variability reflects the irregularity between heartbeat intervals, the heart rate signal needs to be further converted into a heartbeat interval signal for heart rate variability analysis. In this embodiment of the invention, the heart rate signal acquired by the sensor is first filtered to eliminate baseline drift and noise. Then, a peak identification algorithm based on the detection of the zero-crossing point of the first derivative of the signal is used to identify the systolic peak point of each heartbeat cycle. The heartbeat interval is the time difference between these peak points, determining the heartbeat interval sequence within the time window corresponding to the real-time moment. The standard deviation of the heartbeat intervals in the heartbeat interval sequence is used as the heart rate variability.

[0026] It should be noted that the method for obtaining the peripheral perfusion index is a well-known technique in the art. It involves further normalizing the ratio of the AC to DC components of the heart rate signal acquired by a photoplethysmography (PPG) sensor within a time window. The AC component represents the difference between the peak and trough values, and the DC component represents the trough value. The specific formula is as follows: ;in Let be the peripheral perfusion index at time t, where AC represents the alternating current component and DC represents the direct current component. The specific methods for obtaining this index and the meaning of the formula are well-known to those skilled in the art and will not be elaborated or limited here.

[0027] Preferably, in this embodiment of the invention, since the motion signal is a triaxial acceleration signal, three-dimensional information is obtained at each moment. Therefore, the L2 norm of the triaxial acceleration is first calculated to convert the three-dimensional information into one-dimensional information, thereby obtaining the L2 norm signal under the time window. The first derivative of the L2 norm signal at the real time is taken as the spasm initiation impact. That is, the spasm initiation impact characterizes the rate of change of the L2 norm, which reflects acceleration information; that is, the greater the acceleration, the stronger the impact.

[0028] After obtaining the instantaneous state at real time, the system of this embodiment needs to detect and identify stress events in real time for event attribution. Therefore, the physiological stress event detection module 102 detects the state feature information at each moment in real time. Considering that the spastic initiation impact represents an intuitive state impact feature and the peripheral perfusion index represents a macroscopic neural response, this embodiment prioritizes the spastic initiation impact at each moment as the first priority detection and the peripheral perfusion index as the second priority detection. Intuitive information is given the highest priority, followed by neural response, which can identify events with significant stress responses while preventing the omission of neural stress response events.

[0029] Furthermore, regarding stress events, some events may not reflect the stress state through data alone. In this embodiment of the invention, these are defined as unreliable events. Because the information is unreliable, the system output cannot be used to directly intervene in the patient's condition. Therefore, such unreliable events should be specifically judged by a physician. To initially screen out these unreliable events, the physiological stress event detection module 102 mainly classifies them based on the disordered distribution of heart rate variability and peripheral perfusion index within the time period of the stress event. The more disordered the distribution, the more likely it is to be considered. Since these two types of data are intuitive and predictable data, if they exhibit obvious disordered characteristics within the time period, it indicates that the stress event is unreliable; otherwise, it is an event to be analyzed for subsequent causal analysis in other modules.

[0030] Preferably, in this embodiment of the invention, the method for detecting stress events includes: For each moment, the initial spasm impact is first compared with a preset impact threshold. If the initial spasm impact is greater than the preset impact threshold, it indicates that a direct stress response has been observed, and the corresponding moment should be directly used as the reference moment for the stress event. Otherwise, a second priority judgment is made, obtaining the peripheral perfusion index decay between the current moment and the previous moment. If the peripheral perfusion index decay is greater than a preset difference threshold, the current moment is used as the reference moment for the stress event; otherwise, the current moment is ignored. Ignoring the current moment means that the current moment is not the moment when the event occurs or develops.

[0031] It should be noted that the impact threshold and difference threshold in the embodiments of the present invention are obtained by data unification in the early stage of patient treatment. Because the patient's condition is relatively stable in the early stage of treatment, the average value and standard deviation of the initial impact value of spasticity generated in the early stage of treatment are statistically analyzed. The sum of the average value and 5 times the standard deviation is used as the impact threshold, that is, the setting of the impact threshold satisfies the 5 sigma principle. As for the difference threshold, the attenuation of the peripheral perfusion index in the embodiments of the present invention is the absolute value of the difference between the peripheral perfusion index at the current moment and the peripheral perfusion index at the previous moment. The standard deviation of the peripheral perfusion index generated in the early stage of treatment is statistically analyzed, and three times the standard deviation is set as the difference threshold.

[0032] Centered on the reference time of the stress event, the time period corresponding to the stress event is divided according to a preset size division rule. In this embodiment of the invention, considering that the reference time of the stress event should be the time when the event occurs, but the partial time period before the time of occurrence still needs to be analyzed and can be used for subsequent event division, the time period of 5 seconds before and 25 seconds after the reference time of the stress event is taken as the time period corresponding to the stress event. That is, the first 4 seconds of the reference time of the stress event belongs to the initial time period under the corresponding time period.

[0033] Preferably, in this embodiment of the invention, the method for distinguishing between unreliable events and events to be analyzed includes: An initial time period is defined within the stress event's duration, which can be considered a stable period prior to the event. If unwanted data distributions occur during this stable period, it indicates that the event has been contaminated by potential pre-event fluctuations, making normal data analysis impossible. Therefore, the standard deviations of the peripheral perfusion index and heart rate variability are statistically analyzed within the initial time period. If either the standard deviation of the peripheral perfusion index or the standard deviation of heart rate variability exceeds a preset standard deviation threshold for the corresponding dimension, the corresponding stress event is determined to be an unreliable event. Emergency events other than unreliable events are considered events to be analyzed. In this embodiment of the invention, the standard deviation threshold for the peripheral perfusion index is set to 0.1%, and the standard deviation for heart rate variability is set to 5 ms.

[0034] The multi-domain feature decoupling module 103 is used to further analyze the data features in the event to be analyzed and determine the event category. First, considering a complete autonomic nervous system stress response, the observable physiological indicators are manifested as sympathetic-driven vasoconstriction and secondary changes in vagal nerve activity. The vagal-sympathetic response delay reflects the patient's autonomic nervous system's regulatory capacity; this feature quantifies the macroscopic response delay of the vagal system relative to the sympathetic system under observable physiological indicators. This feature can serve as one of the bases for subsequent event category analysis. Therefore, in the event to be analyzed, the sympathetic nerve response time period and the vagal nerve response time period are first divided according to the response distribution of each state feature, and the vagal-sympathetic response delay is obtained.

[0035] Furthermore, considering that heart rate is a nonlinear dynamic system jointly regulated by the sympathetic and parasympathetic nervous systems, different types of external stimuli disrupt its internal dynamic structure in different ways. For stress responses triggered by muscle spasms, the stimulus originates from the spinal cord segments within the patient and is a sudden, intense nerve impulse. This impulse rapidly disrupts the autonomic regulation of the heart, causing the heart rate dynamic structure to become unstable before or almost simultaneously with the macroscopic contraction response of peripheral blood vessels. In this case, the moment of heart rate disturbance will be close to the moment of sympathetic nerve response. On the other hand, for stress responses caused by improper infrared parameters, the stimulus originates from the external skin and is a gradual cumulative thermal effect. This stimulus first triggers a local nociceptor response, which is transmitted through neural pathways, leading to systemic sympathetic nerve excitation, manifested as peripheral vasoconstriction. Only when this sympathetic excitation is sustained or sufficiently strong will it further affect and disrupt the intrinsic stability of heart rate dynamics. Therefore, the starting point of the sympathetic nerve response will be significantly earlier than the moment of heart rate disturbance. Based on this principle, the redundant feature decoupling module sets a sliding window within the time period of the event to be analyzed for localized block analysis. By statistically analyzing the heart rate variability fluctuation index within the sliding window, it identifies the moment when the heart rate dynamics are most unstable due to the event. The time interval between the start of the sympathetic response period and the moment of most heart rate dynamic instability is then used as the response-instability time difference. Based on the response-instability time difference and the vagal-sympathetic response delay, the event category is identified as muscle spasm, inappropriate infrared parameters, and unreliable events. Similarly, unreliable events here also fall under the category of events that this embodiment of the invention considers to require human intervention for judgment.

[0036] It should be noted that the response-instability time difference is the time when the sympathetic response begins minus the time when the heart rate is most unstable; the vagal-sympathetic response delay is the time when the vagal response begins minus the time when the sympathetic response begins.

[0037] Preferably, in this embodiment of the invention, the sympathetic nerve response time period and the vagus nerve response time period are divided in the event to be analyzed, including: Because sympathetic nerve-driven vasoconstriction can be reflected by a decrease in the peripheral perfusion index, and the subsequent changes in vagal nerve activity can be reflected by a decrease in heart rate variability, the characteristics of the two types of neural responses can be determined based on these two characteristics within the time period corresponding to the event being analyzed.

[0038] After the reference time of the stress event, the moment when the peripheral perfusion index first falls below the preset perfusion baseline value for the first consecutive preset number of times is taken as the starting point of the sympathetic nerve response; the moment when the peripheral perfusion index first recovers to the preset perfusion baseline value after the starting point of the sympathetic nerve response is taken as the ending point of the sympathetic nerve response; the time period between the starting point and the ending point of the sympathetic nerve response is the sympathetic nerve response time period. After the reference time of the emergency event, the moment when the heart rate variability is first continuously lower than the preset heart rate variability benchmark value for the second number of consecutive times is taken as the vagal nerve response start point; the moment when the heart rate variability first recovers to the preset perfusion benchmark value after the vagal nerve response start point is taken as the vagal nerve response end point; the time period between the vagal nerve response start point and the vagal nerve response end point is the sympathetic nerve response time period.

[0039] It should be noted that, in this embodiment of the invention, both the first quantity and the second quantity are set to 3. Furthermore, the perfusion baseline value is set to the average peripheral perfusion index during the initial time period of the event, and similarly, the heart rate variability baseline value is set to the average heart rate variability during the initial time period.

[0040] Preferably, in this embodiment of the invention, the method for obtaining the variability fluctuation index includes: Within a sliding window, a heart rate interval sequence is constructed. The Poincaré diagram analysis algorithm is used to obtain the vertical standard deviation and the projected standard deviation along the diagonal of the heart rate interval sequence. The ratio obtained by using the vertical standard deviation as the numerator and the projected standard deviation as the denominator is used as the heart rate variability short-to-long ratio. The heart rate variability short-to-long ratio is used as the variability fluctuation index.

[0041] It should be noted that in this embodiment of the invention, the sliding window size is set to 15 heartbeats, and the sliding step size is 1 heartbeat, meaning the length of the heart rate interval sequence is 15. Under the event segment to be analyzed, each slide of the sliding window generates a variability fluctuation index. The Poincaré plot analysis algorithm is a technique well-known to those skilled in the art. It involves first constructing a local Poincaré scatter plot of the heart rate interval sequence, and then using a fixed algorithm from existing techniques to obtain the vertical standard deviation and projected standard deviation. The specific algorithm is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0042] Furthermore, in this embodiment of the invention, all sliding windows are traversed, and the center point of the sliding window corresponding to the minimum value of the variability fluctuation index is selected as the moment when the heart rate dynamics are most unstable.

[0043] Preferably, in this embodiment of the invention, considering that the response-instability time difference is the time from the onset of the sympathetic nerve response to the time of the most unstable heart rate dynamics, if the event to be analyzed is a muscle spasm event, the calculated response-instability time difference will be close to 0, or even positive; if the event to be analyzed is an event with improper infrared parameters, the calculated response-instability time difference will be a significantly negative value. Therefore, this embodiment of the invention sets a threshold condition: if the response-instability time difference is greater than a preset first time difference threshold, it is determined to be a muscle spasm.

[0044] If the response-instability time difference is not greater than a preset first time difference threshold, and at the same time a vagal-sympathetic response delay occurs, i.e. a vagal-sympathetic response delay greater than a preset second time difference threshold, then it indicates that the stress at this time is from an external source, showing a clear progressive response state, and is judged to be due to improper infrared parameters.

[0045] All other events to be analyzed are unreliable.

[0046] In this embodiment of the invention, the first time difference threshold is set to -0.5, and the second time difference threshold is set to 0. The specific setting of these thresholds can be determined by the implementer according to their desired recognition accuracy, and will not be elaborated upon here.

[0047] Finally, the infrared parameter control module 104 can feed back corresponding control commands based on the event category of the stress event. That is, it completes the feedback of specific instructions for a specific event.

[0048] Preferably, in this embodiment of the invention, the control command corresponding to the event category of the stress event is fed back, including: If the event category is muscle spasm, a pause irradiation command is sent to the infrared light therapy device, and the patient's physiological information is fed back in real time. It should be noted that during this process, the system will continuously monitor the patient's data until the peripheral perfusion index approaches the set perfusion benchmark value and the heart rate variability also approaches the heart rate variability benchmark value and can be maintained stably for one minute. Then, the system can prompt the operator to manually resume the treatment.

[0049] If the event category is an inappropriate infrared light parameter event, a parameter downgrade command is sent to the infrared light therapy device. This parameter downgrade command controls the infrared light therapy device to make an adjustment according to a preset minimum adjustment unit. In this embodiment, the minimum adjustment unit is set to 5 mW / cm². 2 .

[0050] If the event category is uncertain, a stop irradiation command and an alarm will be sent to the infrared light therapy device. This is to ensure patient safety; the system will implement the most conservative strategy and request manual inspection from the operator via an alarm.

[0051] Preferably, in one embodiment of the present invention, in order to achieve individualized optimization of infrared light therapy parameters, the system has the ability to learn across treatment courses, constructing and maintaining a unique safe treatment parameter model for each patient. The system proposed in this embodiment also includes an event statistical comparison and analysis module.

[0052] The event statistics and comparison analysis module is used to create two independent parameter sets for each patient in local storage. These independent parameter sets include a patient-specific insecure parameter set and a patient-specific safe parameter set. The individualized insecure parameter set stores infrared information corresponding to events involving inappropriate infrared parameters, such as: {wavelength: 810nm, power density: 25mW / cm²}. 2 Cumulative time: 12 minutes}; The individualized safety parameter set stores infrared information used during the complete treatment process for stress events that were not reported, such as {wavelength: 810nm, power density: 30mW / cm²}. 2}

[0053] Before each new treatment session begins, the set infrared information is compared with the infrared information in an independent parameter set, and the optimal infrared information is recommended based on the comparison results. Specifically, this includes: first, a safety check is performed; the power density set by the operator must not exceed the lowest unsafe power density recorded for the same wavelength in the unsafe parameter set for that patient. If this is exceeded, the system will issue a warning and prevent the process from starting; simultaneously, the system recommends an initial parameter to the operator, whose power density is the highest power density recorded for the same wavelength in the patient's safe parameter set.

[0054] Through this two-way recording and application mechanism, the system can transform adverse events in a single treatment into "no-go" boundaries for future treatments, while simultaneously transforming successful treatment experiences into "recommended" starting points for future treatments. This cross-treatment data accumulation and iterative application allows treatment plans to evolve from standardized initial settings into data-driven, individualized optimization plans that truly suit the patient's unique physiological tolerance.

[0055] In summary, this invention acquires the patient's heart rate and trunk muscle movement signals in real time, and monitors stress events based on the spasm initiation impact and peripheral perfusion index, classifying stress events into unreliable events and events to be analyzed. For events to be analyzed, the vagal-sympathetic response delay is first quantified, and then the most unstable moment of heart rate dynamics during the event time period is analyzed, quantifying the response-instability time difference. Based on the response-instability time difference and the vagal-sympathetic response delay, the event category is accurately identified. Accurate control commands are then fed back based on the event category. This invention transforms two causal identification problems that are unclear in macroscopic physiological data representation into deterministic problems with a clear temporal order, and then combines the obtained time-domain features to accurately and effectively identify stress events, ensuring the accuracy of control during infrared light therapy and reducing the risk of medical accidents.

[0056] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A system for optimizing infrared light irradiation parameters for neuropathic pain after spinal cord injury, characterized in that, The system includes: The instantaneous state characterization module is used to extract the patient's peripheral perfusion index, heart rate variability, and spasm initiation impact at various moments during infrared light irradiation therapy, based on the patient's heart rate signal and trunk muscle movement signal. The physiological stress event detection module is used to detect stress events by prioritizing the spastic initiation impact as the first priority and the peripheral perfusion index as the second priority. Based on the time period, heart rate variability, and chaotic distribution of the peripheral perfusion index included in the stress event, unreliable events and events to be analyzed are classified. A multi-domain feature decoupling module is used to divide the sympathetic nerve response time period and the vagal nerve response time period in the event to be analyzed, and obtain the vagal-sympathetic response delay; a sliding window is set in the time period of the event to be analyzed, the variability fluctuation index of the heart rate in the sliding window is calculated, and the moment of the most unstable heart rate dynamics in the event to be analyzed is calculated based on the variability fluctuation index; the time interval between the start time of the sympathetic nerve response time period and the moment of the most unstable heart rate dynamics is used as the response-instability time difference; the event category of the event to be analyzed is identified based on the response-instability time difference and the vagal-sympathetic response delay; the event category includes muscle spasm, improper infrared parameters, and unreliable events; The infrared parameter control module is used to provide corresponding control commands based on the event category of the stress event.

2. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The motion signal is a triaxial acceleration signal. For each moment, the L2 norm of the triaxial acceleration is calculated, and the first derivative of the L2 norm signal at each moment is taken as the spasm initiation impact.

3. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The method for detecting stress events includes: For each moment, the initial spasm impact is compared with a preset impact threshold. If the initial spasm impact is greater than the preset impact threshold, the corresponding moment is directly used as the reference moment for the stress event. Otherwise, the peripheral perfusion index decay between the current moment and the previous moment is obtained. If the peripheral perfusion index decay is greater than a preset difference threshold, the current moment is used as the reference moment for the stress event. Otherwise, the current moment is ignored. Centered on the reference time of the stress event, the time period corresponding to the stress event is divided according to the preset size division rules.

4. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The method for distinguishing between unreliable events and events to be analyzed includes: An initial time period is defined within the time frame of the stress event. The standard deviations of the peripheral perfusion index and heart rate variability within the initial time period are statistically analyzed. If either the standard deviation of the peripheral perfusion index or the standard deviation of heart rate variability exceeds the preset standard deviation threshold for the corresponding dimension, the corresponding stress event is determined to be an unreliable event. Emergency events other than unreliable events are considered events to be analyzed.

5. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 3, characterized in that, The process of dividing the event to be analyzed into sympathetic nerve response time periods and vagal nerve response time periods includes: After the reference time of the stress event, the moment when the peripheral perfusion index first falls below the preset perfusion baseline value for the first consecutive preset number of times is taken as the starting point of the sympathetic nerve response; the moment when the peripheral perfusion index first recovers to the preset perfusion baseline value after the starting point of the sympathetic nerve response is taken as the ending point of the sympathetic nerve response; the time period between the starting point and the ending point of the sympathetic nerve response is the sympathetic nerve response time period. After the reference time of the emergency event, the moment when the heart rate variability is first continuously lower than the preset heart rate variability benchmark value for the second number of consecutive times is taken as the vagal nerve response start point; the moment when the heart rate variability first recovers to the preset perfusion benchmark value after the vagal nerve response start point is taken as the vagal nerve response end point; the time period between the vagal nerve response start point and the vagal nerve response end point is the sympathetic nerve response time period.

6. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The method for obtaining the variability volatility index includes: Within a sliding window, a heart rate interval sequence is constructed. The Poincaré diagram analysis algorithm is used to obtain the vertical standard deviation and the projected standard deviation along the diagonal of the heart rate interval sequence. The ratio obtained by using the vertical standard deviation as the numerator and the projected standard deviation as the denominator is used as the heart rate variability short-to-long ratio. The heart rate variability short-to-long ratio is used as the variability fluctuation index.

7. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 6, characterized in that, The method for obtaining the most unstable moment in heart rate dynamics includes: Traverse all sliding windows and select the center point of the sliding window corresponding to the minimum value of the variability fluctuation index as the moment when the heart rate dynamics are most unstable.

8. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The event category identified based on the response-instability time difference and the vagal-sympathetic response delay includes: If the response-instability time difference is greater than a preset first time difference threshold, it is determined to be muscle spasm; if the response-instability time difference is not greater than the preset first time difference threshold, and the vagal-sympathetic response delay is greater than a preset second time difference threshold, it is determined to be improper infrared parameters; other events to be analyzed are unreliable events.

9. The infrared irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The control commands fed back according to the event category of the stress event include: If the event category is muscle spasm, a pause irradiation command is sent to the infrared light therapy device, and the patient's physiological information is fed back in real time; If the event category is an inappropriate infrared light parameter event, a parameter downgrade instruction will be sent to the infrared light therapy device. If the event category is uncertain, a stop irradiation command will be sent to the infrared light therapy device, and an alarm will be triggered.

10. The infrared light irradiation parameter optimization system for neuropathic pain after spinal cord injury according to claim 1, characterized in that, The system also includes an event statistics comparison and analysis module, which creates two independent parameter sets for each patient in local storage. These independent parameter sets include a patient-specific insecure parameter set and a patient-specific safe parameter set. The insecure parameter set stores infrared information corresponding to events where infrared parameters are inappropriate. The safe parameter set stores infrared information used during the complete treatment process when no stress events were reported. Before each new treatment process begins, the set infrared information is compared with the infrared information in the independent parameter sets, and the optimal infrared information is recommended based on the comparison results.

Images