Anesthesia perioperative patient emotion real-time monitoring and active pacification system

Abstract

The present application relates to the technical field of medical monitoring and nursing, in particular to a kind of anesthesia perioperative patient emotion real-time monitoring and active pacification system, including multidimensional biological signal perception module, space-time associated emotion prediction engine, neural feedback type active pacification module, surgical event linkage unit and medical intelligent decision terminal.Multidimensional biological signal perception module fuses electromyogram subconscious emotion capture, saliva cortisol microfluidic rapid detection and non-contact physiological monitoring technology, realize the synchronous acquisition of subjective emotion and objective biomarker;Space-time associated emotion prediction engine is based on the time sequence characteristics of surgical event and patient historical data, advance prediction emotional fluctuation risk;Neural feedback type active pacification module is through transdermal electrical stimulation and multi-sensory collaborative output, forms emotion regulation closed loop;Surgical event linkage unit and operating room equipment real-time interaction, automatically trigger pre-pacification mechanism at key operation node.

Description

Technical Field

[0001] This invention relates to the field of medical monitoring and nursing technology, specifically a real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia. Background Technology

[0002] The perioperative period after anesthesia is the stage in which patients experience the most dramatic physiological and psychological changes during the entire surgical process. During this stage, patients not only face the physical pain caused by surgical trauma but also bear immense psychological pressure regarding surgical risks, the effectiveness of anesthesia, and postoperative recovery. Studies show that over 70% of surgical patients experience varying degrees of anxiety, fear, and tension during the perioperative period. These negative emotions activate the hypothalamic-pituitary-adrenal axis and the sympathetic nervous system, leading to a surge in the secretion of stress hormones such as cortisol and adrenaline, causing a series of physiological responses including increased heart rate, elevated blood pressure, and elevated blood sugar.

[0003] Excessive perioperative stress not only increases the risks of anesthesia and surgery but also has long-term adverse effects on postoperative recovery. For example, severe preoperative anxiety increases the risk of cardiovascular events during anesthesia induction; strong intraoperative stress can lead to increased intraoperative bleeding and immunosuppression; and persistent negative postoperative emotions can prolong hospital stays, increase postoperative pain, raise the incidence of postoperative delirium and cognitive impairment, and may even lead to long-term psychological problems such as chronic pain and post-traumatic stress disorder.

[0004] Currently, there are still many shortcomings in the clinical management of perioperative patient emotions. Regarding emotion monitoring, traditional subjective assessment methods, such as the Self-Rating Anxiety Scale and the Self-Rating Depression Scale, can only reflect the patient's emotional state when conscious, and cannot monitor subconscious emotional changes during anesthesia induction, intraoperative maintenance, and the early stages of anesthesia recovery. Existing objective monitoring methods mostly focus only on routine vital signs such as heart rate and blood pressure. These indicators are affected by various factors such as anesthetic drugs and surgical procedures, and have poor correlation with emotional state, making it difficult to accurately reflect the patient's true emotional level.

[0005] In terms of emotional intervention, existing methods mainly include preoperative psychological counseling, playing music during surgery, and postoperative family visits. These methods are mostly passive and experience-based, lacking scientific theoretical guidance and quantitative effectiveness evaluation. Especially during surgery, because patients are under anesthesia, traditional verbal communication and visual comfort methods have limited effectiveness, and intervention techniques targeting subconscious emotions have not been widely applied. Furthermore, existing intervention methods are mostly implemented only after patients exhibit obvious emotional abnormalities, lacking proactivity and preventative measures, and failing to effectively avoid emotional crises.

[0006] In recent years, with the development of biomedical engineering and artificial intelligence technologies, some new emotion monitoring and intervention technologies have emerged. For example, some studies have used electroencephalogram (EEG) signals to monitor the state of consciousness and emotional changes in anesthetized patients, while others have used heart rate variability analysis to assess patients' autonomic nervous system function and stress levels. However, these technologies still have the following limitations: First, the monitoring dimensions are limited, lacking the integration of multi-dimensional biosignals. Emotion is a complex physiological and psychological process involving changes in multiple systems, including the nervous, endocrine, and immune systems. A single physiological signal can only reflect one aspect of emotion, making it difficult to comprehensively and accurately assess a patient's emotional state.

[0007] Second, they lack predictability and can only provide post-event intervention. Most existing emotion monitoring systems can only reflect the patient's current emotional state and cannot predict future emotional trends. Therefore, they can only provide passive intervention after the patient experiences emotional abnormalities, rather than taking preventative measures in advance.

[0008] Third, the intervention methods are simplistic, lacking specificity and effectiveness. Most existing emotional intervention methods only use single sensory stimuli, such as music therapy, without combining more effective intervention methods such as neuromodulation, and without personalizing them according to individual patient differences and surgical procedures.

[0009] Fourth, they are disconnected from the surgical procedure and lack a systematic approach. Most existing emotion management systems are independent of other equipment and workflows in the operating room, requiring manual operation and management by medical staff. This increases the workload of medical staff and makes it difficult to intervene in a timely manner at critical points in the surgery.

[0010] Therefore, there is an urgent need to develop a system that can comprehensively, accurately, and in real time monitor the emotional changes of patients during the perioperative period of anesthesia, predict the risk of emotional fluctuations in advance, and provide targeted, personalized, and proactive reassurance interventions to improve the quality of perioperative medical care and ensure patient safety. Summary of the Invention

[0011] To address the problems in existing technologies, this invention provides a real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia. This system can comprehensively assess the patient's emotional state through multi-dimensional biosignal fusion technology, provide early warning of emotional risks using a spatiotemporal correlation prediction model, and achieve proactive and personalized emotional intervention through a neurofeedback-based proactive reassurance module, effectively reducing the incidence of perioperative stress response and postoperative psychological complications.

[0012] The technical solution adopted by this invention to solve its technical problem is: a real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, comprising: A multi-dimensional biosignal sensing module is used to collect patients' physiological signals, neural electrical signals, and biomarker data, including an electromyography (EMG) acquisition patch, a microfluidic saliva detection unit, and a millimeter-wave radar sensor. The EMG acquisition patch is attached to the masseter and frontalis muscles of the patient's face to collect EMG signals related to subconscious emotions. The microfluidic saliva detection unit is integrated into the operating table armrest for rapid detection of salivary cortisol concentration. The millimeter-wave radar sensor is installed above the operating table for non-contact acquisition of respiratory rate, heart rate variability, and body movement data. The surgical event linkage unit communicates with the anesthesia machine, surgical instrument table, and surgical timing system to obtain information on the surgical process and key operational events in real time. It automatically identifies five stages: preoperative preparation, anesthesia induction, intraoperative maintenance, anesthesia recovery, and postoperative observation, and marks key operational nodes such as endotracheal intubation, surgical skin incision, and organ traction. The spatiotemporal correlation emotion prediction engine is connected to the multi-dimensional biosignal perception module and the surgical event linkage unit, respectively. It is used to construct an emotion prediction model that includes time and event dimensions, integrate multi-dimensional data to calculate the patient's real-time emotion index, and predict the trend of emotion changes in the near future. The neurofeedback-based active soothing module, connected to the spatiotemporal correlation emotion prediction engine, includes a transcutaneous electrical nerve stimulation unit, a multi-sensory collaborative output unit, and a pre-soothing strategy library. The transcutaneous electrical nerve stimulation unit outputs weak electrical stimulation through an electromyography (EMG) acquisition patch to regulate the patient's neural activity. The multi-sensory collaborative output unit is integrated into the operating table headrest and armrests, and can output audio, vibration, temperature, and weak light stimulation. The pre-soothing strategy library stores standardized soothing plans for different key surgical points and personalized patient preference data. The intelligent medical decision-making terminal is connected to the spatiotemporal correlation emotion prediction engine and the neurofeedback active soothing module, respectively. It is used to display the patient's real-time emotional state, emotion prediction curve and system operation information, receive intervention instructions from medical staff and generate intelligent decision suggestions.

[0013] Specifically, the electromyography (EMG) acquisition patch uses flexible nanoelectrode material, which can simultaneously acquire EMG signals and output transcutaneous nerve electrical stimulation. The electrode surface is coated with biocompatible conductive gel, and there is no skin irritation reaction after long-term wear.

[0014] Specifically, the microfluidic saliva detection unit uses a disposable detection chip that can quantitatively detect the concentration of salivary cortisol within minutes. The detection results are automatically uploaded to the spatiotemporal correlation emotion prediction engine as an objective biomarker for emotion assessment.

[0015] Specifically, the spatiotemporal correlation emotion prediction engine uses a long short-term memory network combined with an attention mechanism to learn the impact of different surgical stages and key operational nodes on the patient's emotions, resulting in a high accuracy rate in emotion prediction.

[0016] Specifically, the spatiotemporal correlation emotion prediction engine also includes an emotion risk grading unit, which divides the patient's emotion risk into three levels: low, medium, and high. Different levels correspond to different warning methods and intervention intensities.

[0017] Specifically, the transcutaneous electrical nerve stimulation unit uses low-frequency pulsed current, and the stimulation parameters are adjusted in real time according to the patient's electromyographic feedback signals, which can effectively inhibit sympathetic nerve excitation and relieve anxiety and tension.

[0018] Specifically, the pre-soothing strategy library automatically triggers a pre-soothing plan tens of seconds before key operation nodes, and adjusts the patient's emotional state in advance through multi-sensory collaborative output, reducing the stress response caused by key operations.

[0019] Specifically, the neurofeedback-based active soothing module employs a closed-loop regulation mechanism to monitor emotional changes after soothing intervention in real time, automatically adjusting stimulation parameters and soothing plans until the patient's emotions return to normal.

[0020] Specifically, the surgical event linkage unit can also receive patient special condition markers input by the anesthesiologist, such as history of drug allergies, history of mental illness, etc. The spatiotemporal correlation emotion prediction engine will adjust the parameters of the emotion assessment and prediction model based on these markers.

[0021] Specifically, the intelligent medical decision-making terminal can generate a complete perioperative emotion management report, including emotion change curves, emotional reactions to key events, and the effects of reassurance interventions, providing a basis for postoperative psychological intervention and optimization of subsequent surgical plans.

[0022] The beneficial effects of this invention are: By employing multi-dimensional biosignal fusion technology, combining electromyography to capture subconscious emotions, detection of objective biomarkers of salivary cortisol, and non-contact physiological monitoring, we can comprehensively and accurately assess the emotional state of patients at all stages of anesthesia, and in particular, capture subconscious emotional changes that cannot be monitored by traditional methods.

[0023] A spatiotemporal correlation emotion prediction engine was designed. Based on long short-term memory networks and attention mechanisms, it can learn the impact of surgical events on patients' emotions and predict the trend of emotional changes in the near future, thus realizing the transformation from passive intervention to proactive prevention.

[0024] A neurofeedback-based active soothing module was developed, which combines transcutaneous electrical nerve stimulation with multi-sensory synergistic output to form a closed-loop regulatory mechanism. This mechanism can automatically adjust intervention parameters based on the patient's real-time feedback, significantly improving the effectiveness of emotional intervention.

[0025] A pre-soothing mechanism was established, which automatically triggers the pre-soothing plan before key surgical procedures, thus adjusting the patient's neurological state in advance and effectively reducing the stress response caused by key procedures, rather than remedial measures afterward.

[0026] It integrates seamlessly with existing operating room equipment and workflows, automatically identifying surgical stages and key operational points without requiring manual operation by medical staff, thus reducing their workload and improving the system's practicality and scalability. Attached Figure Description

[0027] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0028] Figure 1 This is a diagram showing the overall architecture of the system described in this invention; Figure 2 This is an architecture diagram of a spatiotemporal correlation sentiment prediction engine. Figure 3 A flowchart for the surgical event linkage unit; Figure 4 This is a flowchart of the overall system workflow. Detailed Implementation

[0029] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0030] like Figures 1-4 As shown, the present invention provides a real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, comprising a multi-dimensional biosignal perception module, a surgical event linkage unit, a spatiotemporal correlation emotion prediction engine, a neurofeedback proactive reassurance module, and a medical and nursing intelligent decision-making terminal.

[0031] The multidimensional biosignal sensing module is used to collect patients' physiological signals, neural electrical signals, and biomarker data, including an electromyography acquisition patch, a microfluidic saliva detection unit, and a millimeter-wave radar sensor.

[0032] The electromyography (EMG) acquisition patch utilizes flexible nanoelectrode materials, specifically a flexible electrode substrate prepared from a composite of silver nanowires and polydimethylsiloxane (PDMS). The substrate thickness is controlled between 0.1-0.3 mm, exhibiting excellent tensile and bending properties, allowing it to perfectly conform to the contours of the patient's facial skin and maintain good electrical contact even during facial muscle activity or slight head movements. The electrode array uses a 3×3 layout, with each electrode having a diameter of 2 mm and a spacing of 5 mm, corresponding to different muscle fiber pathways in the masseter and frontalis muscles. The electrode surface is coated with a biocompatible conductive gel composed of sodium carboxymethyl cellulose, sodium chloride, and purified water, possessing good conductivity and moisturizing properties, maintaining stable conductivity for 24 hours, and being non-irritating and non-sensitizing to the skin. The EMG acquisition patch simultaneously functions as a signal acquisition device and an electrical stimulation output device. A multiplexed switch enables rapid switching between acquisition and stimulation modes, with a switching time of less than 10 milliseconds, ensuring uninterrupted signal acquisition.

[0033] The microfluidic saliva detection unit uses a disposable detection chip, with overall dimensions of 50 mm × 30 mm × 2 mm, fabricated from polymethyl methacrylate (PMMA) using an injection molding process. The chip integrates four functional areas: a sample inlet, microchannels, a reaction zone, and an electrochemical detection zone. The sample inlet has a diameter of 5 mm, enabling rapid collection of saliva samples from patients. The microchannels are 100 micrometers wide and 50 micrometers deep, with a hydrophilic surface treatment to guide the saliva sample automatically to the reaction zone. The reaction zone contains pre-immobilized cortisol monoclonal antibodies and horseradish peroxidase-labeled antigens, enabling a specific immune reaction with cortisol in saliva. The electrochemical detection zone uses screen-printed carbon electrodes to detect the current signal generated by the immune reaction; the current intensity is linearly correlated with the concentration of salivary cortisol. During detection, the patient's saliva sample is dripped into the sample inlet, enters the reaction zone under capillary action to undergo an immune reaction, and then enters the electrochemical detection zone for detection. The entire detection process is completed within 3-5 minutes. The test results are automatically uploaded to the spatiotemporal correlation emotion prediction engine via Bluetooth communication, eliminating the need for manual input.

[0034] The millimeter-wave radar sensor is installed 1.2-1.5 meters above the operating table, employing 24GHz frequency-modulated continuous wave radar technology with a transmission power of less than 10 milliwatts, meeting medical device safety standards. This sensor features adaptive dynamic clutter suppression, using digital signal processing algorithms to identify and filter clutter signals generated in the operating room in real time, such as equipment movement, staff movement, and air conditioning airflow. It accurately extracts echo signals generated by the patient's chest wall micro-movements, thereby calculating respiratory rate, heart rate variability, and body movement data. Compared to traditional contact-based ECG monitoring, millimeter-wave radar monitoring does not require electrodes to be attached to the patient, does not interfere with surgical sterilization or the operating area, and does not cause patient discomfort.

[0035] The surgical event linkage unit communicates with the anesthesia machine, surgical instrument table, surgical timing system, and anesthesia information management system in the operating room via wired Ethernet or 5G wireless, with a communication latency of less than 100 milliseconds, enabling real-time acquisition of surgical progress and key operational event information. The surgical event linkage unit incorporates a rule-based intelligent recognition algorithm that automatically identifies five surgical stages—preoperative preparation, anesthesia induction, intraoperative maintenance, anesthesia recovery, and postoperative observation—based on the anesthesia machine's drug administration type and dosage, the surgical instrument table's instrument usage records, the surgical timing system's cumulative time, and the anesthesiologist's electronic operation records. It also accurately marks key operational nodes such as endotracheal intubation, surgical incision, organ retraction, abdominal closure, and extubation.

[0036] The surgical event linkage unit can also receive patient-specific condition markers input by anesthesiologists, including but not limited to history of drug allergies, mental illness, previous surgeries, post-traumatic stress disorder, alcoholism, and drug dependence. Anesthesiologists can input these markers through a dedicated interface on the medical intelligent decision-making terminal; each marker corresponds to a unique code. The surgical event linkage unit transmits this marker information to the spatiotemporal correlation emotion prediction engine in real time. The engine adjusts the parameters of the emotion assessment and prediction model based on different marker types. For example, for patients with a history of mental illness, the model increases the weight of electromyography signals and cortisol levels; for patients with a history of post-traumatic stress disorder, the model increases sensitivity to changes in the sound and light of surgical instruments.

[0037] The spatiotemporal correlation emotion prediction engine is connected to the multi-dimensional biosignal perception module and the surgical event linkage unit, and is the core processing unit of the entire system.

[0038] The spatiotemporal correlation emotion prediction engine employs a deep learning architecture combining a Long Short-Term Memory (LSTM) network with an attention mechanism. The model is divided into five parts: an input layer, a feature extraction layer, an attention layer, a fusion layer, and an output layer. The input layer receives multi-dimensional biosignal data and surgical event context information. The biosignal data undergoes preprocessing at 1-second sampling intervals, including denoising, normalization, and feature extraction. The surgical event information is input in the form of event codes, including event type, occurrence time, and duration. The feature extraction layer consists of two bidirectional LSTM layers, capable of capturing long-term dependencies in time-series data and extracting emotional features at different time scales. The attention layer automatically learns the influence weights of different surgical stages and key operational nodes on the patient's emotions, assigning higher attention weights to high-stress operational nodes (such as endotracheal intubation and surgical skin incision). The fusion layer performs multimodal fusion of biosignal features and surgical event features to generate a comprehensive emotion feature vector. The output layer outputs the patient's real-time emotion index and an emotion prediction curve for the next 5-10 minutes. The model is pre-trained using a large amount of clinical labeled data and continuously optimized through incremental learning during actual use, achieving an accuracy rate of over 85% in emotion prediction.

[0039] The spatiotemporal correlation emotion prediction engine also includes an emotion risk grading unit. Based on real-time emotion indices and emotion prediction results, this unit uses a fuzzy comprehensive evaluation method to classify patients' emotional risks into three levels: low, medium, and high. Low risk corresponds to an emotion index within the normal range and a predicted stable future emotion; in this case, the system only performs routine monitoring and requires no special intervention. Medium risk corresponds to an emotion index approaching the abnormal threshold or a predicted upward trend in future emotion; in this case, the system automatically triggers preventative reassurance intervention and sends a yellow warning to medical staff. High risk corresponds to an emotion index exceeding the abnormal threshold or a predicted sharp deterioration in future emotion; in this case, the system immediately initiates emergency reassurance intervention, sends a red warning to medical staff, and provides both audio and visual cues. The emotion risk grading results are displayed in real-time on the medical intelligent decision-making terminal and serve as the basis for the proactive reassurance module to adjust its intervention strategies.

[0040] The neurofeedback-based active soothing module is connected to a spatiotemporally correlated emotion prediction engine, including a transcutaneous electrical nerve stimulation unit, a multi-sensory collaborative output unit, and a pre-soothing strategy library.

[0041] The transcutaneous electrical nerve stimulation (TENS) unit uses low-frequency pulsed current. Current parameters include frequency, pulse width, and intensity. The frequency range is 2-100 Hz, the pulse width is 50-500 microseconds, and the intensity is 0-50 mA. All parameters are adjustable within safe threshold ranges. This unit outputs low-frequency pulsed current through an electromyography (EMG) acquisition patch to stimulate the trigeminal and facial nerve branches in the patient's face. This modulates the activity of the limbic system and autonomic nervous system, inhibits sympathetic nerve excitation, and promotes parasympathetic nerve activity, thereby effectively relieving anxiety and tension. The stimulation parameters are adjusted in real time based on the patient's EMG feedback signals. When an increase in the amplitude of the patient's EMG signal is detected, the stimulation intensity is appropriately increased; when the amplitude of the EMG signal returns to normal, the stimulation intensity is gradually decreased, forming an adaptive adjustment mechanism. Simultaneously, the unit also has overcurrent and overheat protection functions. When the output current exceeds the safe threshold or the electrode temperature exceeds 40°C, the output is automatically cut off to ensure patient safety.

[0042] The multi-sensory collaborative output unit is integrated into the headrest and armrests of the operating table, providing four types of sensory stimulation: audio, vibration, temperature, and dim light. The headrest integrates bone conduction speakers and a temperature-sensing massage component. The bone conduction speakers have a frequency response range of 20Hz-20kHz, transmitting sound through the skull without the need for headphones, thus not interfering with surgical procedures or causing pressure on the patient's ears. The temperature-sensing massage component uses carbon fiber heating wires and a miniature vibration motor to provide 35-38℃ heat therapy and vibration massage of varying intensities, relieving neck muscle tension. The armrests integrate vibration feedback components and ambient lighting strips. The vibration feedback component provides vibration stimulation from 10-50Hz; the ambient lighting strips use RGB LED beads to emit soft, different colored light, creating a comfortable atmosphere. All output parameters are automatically controlled by the system and can also be manually adjusted by medical staff.

[0043] The pre-soothing strategy library automatically triggers pre-soothing protocols tens of seconds before key operational nodes. The specific triggering time is set according to the stress intensity of different operational nodes: 30 seconds in advance for high-stress operational nodes (such as endotracheal intubation and surgical skin incision), 20 seconds in advance for moderate-stress operational nodes (such as organ traction and abdominal closure), and 10 seconds in advance for low-stress operational nodes. The pre-soothing strategy library stores standardized soothing protocols for different surgical stages and key operational nodes. Each protocol includes specific parameter combinations of audio, vibration, temperature, dim light, and transcutaneous electrical nerve stimulation. The system automatically matches the corresponding pre-soothing protocol based on the key operational node information provided by the surgical event linkage unit and executes it through the multi-sensory collaborative output unit and the transcutaneous electrical nerve stimulation unit. For example, 30 seconds before endotracheal intubation, for patients in the light sedation and hazy state of anesthesia induction, the system will play soothing alpha wave audio, output gentle warming and low-frequency transcutaneous electrical nerve stimulation, soothe subconscious facial muscle tension and inhibit excessive sympathetic nerve excitation at the neural level, regulate the body's stress threshold in advance, and reduce physiological stress fluctuations caused by endotracheal intubation.

[0044] The neurofeedback-based active soothing module employs a closed-loop regulation mechanism. The entire regulation process includes the following four steps: First, a multi-dimensional biosignal sensing module collects real-time data related to the patient's emotions. Second, a spatiotemporal correlation emotion prediction engine calculates the patient's real-time emotion index and assesses the effectiveness of the soothing intervention. Third, if the emotion index does not reach the expected target, the system automatically adjusts the stimulation parameters and soothing plan, such as increasing the intensity of transcutaneous electrical nerve stimulation, changing the music type, or increasing vibration stimulation. Fourth, the above steps are repeated until the patient's emotion index returns to the normal range. The closed-loop regulation cycle is 10 seconds, enabling rapid response to changes in the patient's emotions. If the patient's mood still does not improve after three adjustments, the system sends a high-risk warning to the medical intelligent decision-making terminal, recommending manual intervention by medical staff, such as increasing the depth of anesthesia or administering sedatives.

[0045] The intelligent medical decision-making terminal features a 15.6-inch capacitive touchscreen, making it simple and convenient to operate. It is installed at the anesthesiologist's workstation. The terminal can display real-time information such as the patient's basic information, vital signs, surgical stage, real-time emotional index, emotional prediction curve, and system operating status. When the patient's emotional risk reaches a moderate level, the terminal will display a yellow warning and recommend corresponding preventative interventions; when the emotional risk reaches a high level, the terminal will display a red warning, provide audible and visual alerts, and recommend emergency interventions.

[0046] Healthcare professionals can view detailed emotion analysis reports through the terminal to understand the patterns of patients' emotional changes and the effectiveness of reassurance interventions. The terminal can also receive manual intervention commands from healthcare professionals, allowing them to manually adjust the parameters of the reassurance plan or pause the system's automatic reassurance function based on the patient's specific situation. After the surgery, the terminal automatically generates a complete perioperative emotion management report, including emotion change curves, emotional responses to key events, reassurance intervention records, and intervention effectiveness evaluations.

[0047] Example 1: This example provides an application of a real-time monitoring and active reassurance system for patients' emotions during the perioperative period of anesthesia in radical gastrectomy for gastric cancer under general anesthesia.

[0048] The patient, a middle-aged male, was admitted to the hospital for gastric cancer and scheduled for radical gastrectomy under general anesthesia. He had a history of chronic underlying diseases, was taking medication regularly, and his condition was stable. Preoperative psychological assessment indicated moderate anxiety. The day before surgery, medical staff explained the system's working principle and usage to the patient in detail and obtained informed consent. Then, electromyography (EMG) acquisition patches were applied to the patient's bilateral masseter and frontalis muscles. The patient underwent a simple pre-experience, choosing his preferred music type, vibration intensity, temperature, and light color, which was then stored in a pre-soothing strategy database. Simultaneously, a baseline saliva sample was collected to measure baseline cortisol levels.

[0049] On the day of the surgery, after the patient entered the operating room and lay on the operating table, the system automatically started. The multi-dimensional biosignal sensing module began to work: the electromyography (EMG) acquisition patch collected real-time EMG signals from the patient's masseter and frontalis muscles; the millimeter-wave radar sensor began to collect the patient's respiratory rate, heart rate variability, and body movement data; and the microfluidic saliva detection unit collected the patient's preoperative saliva sample, and the test results showed that the cortisol concentration was higher than the normal reference value.

[0050] The surgical event linkage unit identifies the current preoperative preparation stage based on information from the surgical timing system. The spatiotemporal correlation emotion prediction engine invokes the preoperative preparation stage model, integrating multi-dimensional data to calculate the patient's real-time emotion index, showing that the patient exhibits significant anxiety and fear, with an emotion risk level of medium. The system automatically triggers a preventative reassurance plan: playing the patient's preferred soft music through bone conduction speakers, while simultaneously activating the headrest's temperature-sensitive massage component to provide appropriate heat therapy and low-intensity vibration massage.

[0051] After a period of reassurance, the patient's electromyographic signal amplitude decreased significantly, respiratory rate and heart rate variability stabilized, and salivary cortisol concentration decreased significantly upon retesting. The spatiotemporal correlation emotion prediction engine calculated a significant improvement in the patient's emotional index, reducing emotional risk to low risk.

[0052] Subsequently, the anesthesiologist began anesthesia induction, administering anesthetic drugs intravenously. The surgical event linkage unit automatically identified the current anesthesia induction phase based on the anesthesia machine's drug administration information. The spatiotemporal correlation emotion prediction engine invoked the model for the anesthesia induction phase, increasing the weight of electromyographic (EMG) signals. At this time, the patient's consciousness gradually faded, but the EMG signals could still reflect subconscious emotional changes. The system detected a brief increase in EMG signals when the patient received a portion of the anesthetic, indicating mild discomfort. The neurofeedback-based active reassurance module immediately increased the intensity of transcutaneous electrical nerve stimulation (TENS) while simultaneously increasing the music volume. A few seconds later, the patient's EMG signals returned to a stable state.

[0053] After anesthesia induction, the physician performed endotracheal intubation. The surgical event linkage unit detected the impending intubation in advance and automatically triggered a pre-soothing protocol: increasing the frequency of transcutaneous electrical nerve stimulation (TENS) while providing gentle vibration stimulation. During intubation, the patient's vital signs remained stable, electromyography (EMG) signals showed no significant abnormal fluctuations, and cortisol levels did not increase significantly, indicating that the pre-soothing protocol effectively reduced the stress response induced by intubation.

[0054] Once the surgery entered the maintenance phase, the patient was under deep anesthesia. The spatiotemporal correlation emotion prediction engine invoked the model from the maintenance phase, increasing the weights of millimeter-wave radar physiological signals and salivary cortisol concentration. The system periodically collected saliva samples from the patient to detect cortisol concentration, while continuously monitoring electromyography and physiological signals. When the surgery reached the main operational steps, the surgeon performed gastrectomy and lymph node dissection; the surgical event linkage unit marked this critical operational node. The system pre-triggered a pre-reassurance program, increasing the intensity of warm massage. During the procedure, the patient's heart rate and blood pressure fluctuated slightly but quickly returned to normal, and cortisol concentration remained at a stable level.

[0055] As the surgery neared its end, the system detected a gradual increase in the amplitude of the patient's electromyographic (EMG) signals and a slight increase in respiratory rate, predicting a significant rise in the patient's stress level in the short term. The spatiotemporal correlation-based emotion prediction engine upgraded the emotional risk to medium risk and sent a warning to the medical intelligent decision-making terminal. The neurofeedback-based active reassurance module automatically adjusted the reassurance plan: increasing the intensity and frequency of transcutaneous electrical nerve stimulation (TENS) while activating the vibration feedback component of the armrests. Shortly after the intervention, the patient's EMG and physiological signals returned to stability, and the expected increase in stress level did not occur.

[0056] After the surgery, the anesthesiologist stopped administering medication, and the patient entered the anesthesia recovery phase. The surgical event linkage unit automatically identified the current anesthesia recovery phase based on information from the anesthesia machine and changes in the patient's vital signs. The spatiotemporal correlation emotion prediction engine invoked the anesthesia recovery phase model, further increasing the weight of electromyographic signals. At this time, the patient's consciousness gradually returned, making them prone to agitation and anxiety. The system detected a significant increase in the patient's electromyographic signals, indicating a rise in anxiety. The neurofeedback-based active reassurance module adjusted the reassurance plan: reducing the intensity of transcutaneous electrical nerve stimulation while playing a pre-recorded message from the patient's family.

[0057] As the patient gradually regained consciousness, the system gradually reduced the intensity of transcutaneous electrical nerve stimulation while increasing the intensity of visual and auditory stimulation. Once the patient was able to open their eyes and understand speech, the system shut down the transcutaneous electrical nerve stimulation and continued playing music and voice messages from family members. After the patient was fully awake, the doctor successfully removed the endotracheal tube. During the extubation process, the patient remained emotionally stable, without any agitation induced by fear or other negative emotions, nor any coughing reaction triggered by increased airway sensitivity. The coughing reaction is essentially a natural reflex triggered by physical stimulation of the airway. Although it is not directly controlled by subjective emotions, the body's stress state (such as sympathetic excitation and increased airway tension caused by fear) significantly increases its probability and intensity.

[0058] After the patient was transferred to the postoperative observation room, the surgical event linkage unit automatically identified that they were in the postoperative observation phase. The system continued to monitor the patient's emotional changes and periodically collected saliva samples to test cortisol levels. After a period of time post-surgery, the patient's mood index returned to the normal range, and their cortisol level also returned to normal. Once the patient's mood stabilized, the system discontinued its automatic calming function.

[0059] On the first day after surgery, medical staff followed up with the patient. The patient reported that they did not experience significant tension or fear during the procedure and felt reassured upon waking and hearing their family members' voices. The patient's postoperative recovery was good, with no significant postoperative pain or psychological discomfort. The patient was discharged smoothly approximately one week after surgery.

[0060] Example 2: This example provides an application of a real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia in cesarean section under spinal anesthesia, and compares it with traditional emotion management methods.

[0061] Several women who underwent cesarean section under spinal anesthesia at our hospital were randomly divided into an observation group and a control group. There were no statistically significant differences in age, gestational age, weight, preoperative anxiety level, anesthesia method, or other general characteristics between the two groups, making them comparable. The observation group received emotional management using the perioperative patient emotion real-time monitoring and proactive reassurance system described in this invention, while the control group received traditional emotional management methods, including preoperative psychological counseling, intraoperative nurse accompaniment, and postoperative family visits.

[0062] The day before the procedure, electromyography (EMG) collection patches were applied to the observation group of mothers for a pre-experience. The mothers were allowed to choose their preferred music type, vibration intensity, temperature, and light color, and their family members recorded encouraging messages, which were then stored in the system. Simultaneously, basal saliva samples were collected from both groups of mothers to measure their basal cortisol levels.

[0063] On the day of the surgery, after the mothers in the observation group entered the operating room, the system automatically started, began real-time monitoring of their emotional changes, and proactively provided personalized reassurance interventions based on their emotional state and the progress of the surgery. In the control group, the mothers received routine psychological counseling and companionship from the circulating nurses after entering the operating room.

[0064] Saliva samples were collected from both groups of mothers before surgery, before anesthesia puncture, during fetal delivery, during placental delivery, and postoperatively to measure cortisol concentration. Heart rate, systolic blood pressure, and diastolic blood pressure changes were also recorded for both groups. Postoperative pain levels were assessed using the visual analog scale (VAS) and the Edinburgh Postpartum Depression Scale (EDS) was used to assess the risk of postpartum depression.

[0065] The results showed no statistically significant differences in cortisol concentration, heart rate, systolic blood pressure, and diastolic blood pressure between the two groups before the procedure. From the time of anesthesia and puncture, the cortisol concentration, heart rate, and systolic blood pressure in the observation group were significantly lower than those in the control group, with this difference being particularly pronounced during key operative stages such as fetal and placental delivery. Postoperatively, the cortisol concentration, heart rate, and systolic blood pressure in the observation group remained significantly lower than those in the control group.

[0066] Postoperative pain assessment results showed that the pain scores of the observation group at all postoperative time points were significantly lower than those of the control group. Postpartum depression assessment results showed that the postpartum depression scores of the observation group were significantly lower than those of the control group, and the incidence of postpartum depression was also significantly lower in the observation group. Furthermore, the intraoperative blood loss, postoperative analgesic dosage, and postoperative hospital stay were significantly less in the observation group than in the control group.

[0067] The results of this embodiment show that the real-time monitoring and active reassurance system for patients' emotions during the perioperative period of anesthesia described in this invention can effectively reduce perioperative stress response in women undergoing cesarean section, stabilize vital signs, alleviate postoperative pain, reduce the risk of postpartum depression, and promote postoperative recovery.

[0068] Example 3: This example provides an application of a real-time monitoring and active reassurance system for patients' emotions during the perioperative period of anesthesia in elderly patients undergoing hip replacement surgery, and compares it with a single-modal emotion monitoring system.

[0069] Several elderly patients who underwent hip replacement surgery at our hospital were randomly divided into an experimental group and a control group. There were no statistically significant differences in age, sex, weight, preoperative physical condition, and operation time between the two groups, making them comparable. The experimental group used the multi-dimensional biosignal fusion emotion monitoring and neurofeedback active soothing system described in this invention, while the control group used a single heart rate variability emotion monitoring system and traditional music soothing methods.

[0070] The day before the procedure, electromyography (EMG) acquisition patches were applied to patients in the experimental group for a pre-experience. Patients were allowed to choose their preferred music type, vibration intensity, temperature, and light color, and this information was stored in the system. Simultaneously, baseline saliva samples were collected from both groups of patients to measure baseline cortisol levels.

[0071] On the day of the surgery, after the two groups of patients entered the operating room, their respective systems were activated to monitor the patients' emotional changes in real time and provide corresponding reassurance and intervention.

[0072] The accuracy rates of emotion recognition, emotion prediction, and soothing effectiveness were recorded for both systems. Emotion recognition accuracy was assessed using the subjective evaluation of experienced anesthesiologists as the gold standard; accurate recognition was considered achieved when the emotion index calculated by the system matched the anesthesiologist's assessment. Emotion prediction accuracy refers to the proportion of patients whose predicted emotion trends matched the actual emotion trends. Soothing effectiveness refers to the proportion of patients whose emotion index significantly improved within a short period after the system implemented soothing intervention.

[0073] In addition, intraoperative changes in vital signs, postoperative delirium incidence, and postoperative cognitive impairment incidence were recorded in both groups of patients. Postoperative delirium was assessed using the confusion assessment method, and postoperative cognitive impairment was assessed using the Mini-Mental State Examination.

[0074] The results showed that the experimental group's emotion recognition accuracy was significantly higher than that of the control group. This advantage was particularly pronounced during the anesthesia induction and recovery phases, while the control group exhibited lower accuracy in these two phases. The experimental group's system also possessed an emotion prediction function, accurately predicting patients' emotional changes in advance, a function absent in the control group's system. Furthermore, the experimental group's system was significantly more effective in calming patients than the control group.

[0075] Intraoperative vital sign monitoring results showed that the experimental group had significantly smaller fluctuations in heart rate and blood pressure than the control group, and the amount of vasoactive drugs used during the operation was significantly less than that in the control group.

[0076] Postoperative follow-up results showed that the incidence of postoperative delirium and postoperative cognitive impairment was significantly lower in the experimental group than in the control group. Furthermore, the postoperative hospital stay was significantly shorter in the experimental group than in the control group.

[0077] Further analysis revealed that the advantages of the experimental group system are mainly reflected in the following aspects: First, the multi-dimensional biosignal fusion technology can comprehensively and accurately assess the patient's emotional state, especially capturing the subconscious emotional changes of elderly patients under anesthesia; second, the emotion prediction function can detect potential emotional risks in advance and take preventive intervention measures; third, the neurofeedback active soothing module combines transcutaneous electrical nerve stimulation and multi-sensory synergistic output, and the intervention effect is significantly better than music therapy alone.

[0078] The results of this embodiment show that the real-time monitoring and active reassurance system for patients in the perioperative period of anesthesia described in this invention is significantly superior to single-modality emotion monitoring systems in terms of emotion recognition accuracy, emotion prediction ability, and reassurance effect. It is particularly suitable for elderly patients and other groups with special emotional reactions and a high risk of postoperative psychological complications.

[0079] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, characterized in that, include: A multi-dimensional biosignal sensing module is used to collect patients' physiological signals, neural electrical signals, and biomarker data, including an electromyography (EMG) acquisition patch, a microfluidic saliva detection unit, and a millimeter-wave radar sensor. The EMG acquisition patch is attached to the masseter and frontalis muscles of the patient's face to collect EMG signals related to subconscious emotions. The microfluidic saliva detection unit is integrated into the operating table armrest for rapid detection of salivary cortisol concentration. The millimeter-wave radar sensor is installed above the operating table for non-contact acquisition of respiratory rate, heart rate variability, and body movement data. The surgical event linkage unit communicates with the anesthesia machine, surgical instrument table, and surgical timing system to obtain information on the surgical process and key operational events in real time. It automatically identifies five stages: preoperative preparation, anesthesia induction, intraoperative maintenance, anesthesia recovery, and postoperative observation, and marks key operational nodes such as endotracheal intubation, surgical skin incision, and organ traction. The spatiotemporal correlation emotion prediction engine is connected to the multi-dimensional biosignal perception module and the surgical event linkage unit, respectively. It is used to construct an emotion prediction model that includes time and event dimensions, integrate multi-dimensional data to calculate the patient's real-time emotion index, and predict the trend of emotion changes in the near future. The neurofeedback-based active soothing module, connected to the spatiotemporal correlation emotion prediction engine, includes a transcutaneous electrical nerve stimulation unit, a multi-sensory collaborative output unit, and a pre-soothing strategy library. The transcutaneous electrical nerve stimulation unit outputs weak electrical stimulation through an electromyography (EMG) acquisition patch to regulate the patient's neural activity. The multi-sensory collaborative output unit is integrated into the operating table headrest and armrests, and can output audio, vibration, temperature, and weak light stimulation. The pre-soothing strategy library stores standardized soothing plans for different key surgical points and personalized patient preference data. The intelligent medical decision-making terminal is connected to the spatiotemporal correlation emotion prediction engine and the neurofeedback active soothing module, respectively. It is used to display the patient's real-time emotional state, emotion prediction curve and system operation information, receive intervention instructions from medical staff and generate intelligent decision suggestions.

2. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia as described in claim 1, characterized in that: The electromyography (EMG) acquisition patch uses flexible nanoelectrode material, which can simultaneously acquire EMG signals and output transcutaneous nerve electrical stimulation. The electrode surface is coated with biocompatible conductive gel, and there is no skin irritation reaction after long-term wear.

3. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia as described in claim 1, characterized in that: The microfluidic saliva detection unit uses a disposable detection chip that can quantitatively detect salivary cortisol concentration within minutes. The detection results are automatically uploaded to the spatiotemporal correlation emotion prediction engine as an objective biomarker for emotion assessment.

4. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia as described in claim 1, characterized in that: The spatiotemporal correlation emotion prediction engine uses a long short-term memory network combined with an attention mechanism to learn the impact of different surgical stages and key operational nodes on the patient's emotions.

5. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia according to claim 1, characterized in that: The spatiotemporal correlation emotion prediction engine also includes an emotion risk grading unit, which divides the patient's emotion risk into three levels: low, medium, and high. Different levels correspond to different warning methods and intervention intensities.

6. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia according to claim 1, characterized in that: The transcutaneous electrical nerve stimulation unit uses low-frequency pulsed current, and the stimulation parameters are adjusted in real time according to the patient's electromyographic feedback signals, which can effectively inhibit sympathetic nerve excitation and relieve anxiety and tension.

7. The real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia according to claim 1, characterized in that: The pre-soothing strategy library automatically triggers a pre-soothing plan tens of seconds before key operation nodes, and adjusts the patient's emotional state in advance through multi-sensory collaborative output, reducing the stress response caused by key operations.

8. A real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, as described in claim 1, is characterized in that: The neurofeedback-based active soothing module employs a closed-loop regulation mechanism to monitor emotional changes after soothing intervention in real time, automatically adjusting stimulation parameters and soothing plans until the patient's emotions return to normal.

9. A real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, as described in claim 1, is characterized in that: The surgical event linkage unit can also receive patient special condition markers input by the anesthesiologist, and the spatiotemporal correlation emotion prediction engine will adjust the parameters of the emotion assessment and prediction model based on these markers.

10. A real-time monitoring and proactive reassurance system for patients' emotions during the perioperative period of anesthesia, as described in claim 1, characterized in that: The intelligent medical decision-making terminal can generate a complete perioperative emotion management report, including emotion change curves, emotional reactions to key events, and the effects of reassurance interventions, providing a basis for postoperative psychological intervention and optimization of subsequent surgical plans.

Images