Patient pain assessment method and system

Abstract

The invention discloses a patient pain assessment method and system, and relates to the technical field of pain assessment. A patient pain assessment system comprises a data acquisition module, a self-assessment correction module, a physiological assessment correction module, an expression assessment correction module and a pain assessment module. According to the method, the personalized self-evaluation deviation P of the patient is calculated, and the score V of the visual simulation scale VAS filled by the patient is subjected to personalized correction, so that the subjective difference between individual patients can be effectively reduced, the final VAS score is closer to the actual pain level, the real pain state of the patient can be reflected more accurately, and the patient experience is improved. A doctor can make more accurate diagnosis, insufficient treatment or excessive treatment caused by misjudgment is avoided, and the assessment accuracy of the patient pain assessment method and system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pain assessment, and in particular to a method and system for assessing pain in patients. Background Art

[0002] Pain is a complex physiological and psychological experience, usually a response to physical injury or potential injury. It is a protective mechanism of the human body to alert us to possible problems in a certain part of the body and the need to take action to avoid further damage. Pain assessment is a very important part of medical care. It helps doctors and caregivers understand the degree, nature and changes of patients' pain, so as to develop effective treatment plans. With the development of technology, many hospitals and medical institutions have introduced electronic and intelligent pain assessment systems to improve the accuracy and efficiency of assessments.

[0003] However, existing patient pain assessment methods and systems mainly rely on the degree of deviation of the patient's physiological data, the range of changes in facial expressions, or the scores of pain assessment scales such as VAS, NRS and FPS-R to assess the patient's pain level and formulate corresponding treatment plans. If only a single type of assessment method is used for pain assessment, not only will the assessment results be incomplete, but the objectivity and accuracy of the assessment results will also be low. In addition, most of the above-mentioned assessment methods are more dependent on the patient's self-perception and self-report, and are therefore highly subjective and easily affected by individual differences, emotional fluctuations, cognitive abilities and other factors, resulting in low assessment accuracy of existing patient pain assessment methods and systems.

[0004] Based on the above situation, the present invention proposes a patient pain assessment method and system with high assessment accuracy. Summary of the invention

[0005] In order to overcome the disadvantage that the existing patient pain assessment methods and systems mainly rely on the degree of deviation of the patient's physiological data, the change of facial expression or the scores of pain assessment scales such as VAS, NRS and FPS-R to assess the patient's pain level and formulate corresponding treatment plans, and if only a single type of assessment method is used for pain assessment, not only will the assessment results be incomplete, but the objectivity and accuracy of the assessment results will also be low, and most of the above-mentioned assessment methods are more dependent on the patient's self-perception and self-report, and thus have strong subjectivity and are easily affected by factors such as individual differences, emotional fluctuations and cognitive abilities, thereby resulting in the disadvantage that the existing patient pain assessment methods and systems have low assessment accuracy, the present invention proposes a patient pain assessment method and system with high assessment accuracy.

[0006] A method for assessing pain in a patient, comprising the following steps: Use medical-grade electrical stimulation equipment to simulate different levels of pain on patients, use cameras to capture images of patients' facial expressions when they experience different levels of pain, use medical monitoring equipment to measure patients' various physiological data under different levels of pain, and read the visual analog scale (VAS) scores filled out by patients under different levels of pain , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; Based on pain simulation level The score of visual analogue scale (VAS) The difference between the two is used to calculate the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score ; Based on the patient's physiological data under different levels of pain, the cubic spline interpolation method is used to draw the pain level assessment curve of each physiological data changing with the pain level. The patient's physiological data at the onset of the disease are matched with the pain level assessment curve to obtain the pain assessment level corresponding to each physiological data. The pain assessment level corresponding to each physiological data is comprehensively calculated to obtain the pain physiological assessment score. ; The trained facial expression pain assessment model is fine-tuned based on the facial expression images of patients when they experience different levels of pain. The facial expression images collected when the patient is ill are input into the fine-tuned facial expression pain assessment model to obtain the pain expression assessment score. ; Pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The patient's comprehensive pain score This is considered as the result of the pain assessment, and the patient is assigned an appropriate treatment plan based on it.

[0007] As a preferred aspect of the invention, the level of pain simulation The range is 0-9, that is , and the pain self-assessment score , pain physiological assessment score and pain expression assessment score The score range is 0-10.

[0008] As a preferred aspect of the invention, the calculation obtains the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score The specific steps are: Calculate the level of pain simulation at different pain levels The score of visual analogue scale (VAS) The difference between the two values ​​was summed and averaged to obtain the patient's personalized self-assessment deviation. , the calculation formula is: ; The VAS score was calculated by the patient's and patients’ individualized self-assessment bias Sum the patient's pain self-assessment score , the calculation formula is: .

[0009] As a preferred aspect of the invention, the pain level assessment curves of various physiological data changing with pain level include: heart rate-pain level curve , Blood pressure-pain level curve and respiratory rate-pain level curve The pain assessment levels corresponding to the various physiological data include: heart rate-pain assessment level , blood pressure-pain assessment level and respiratory rate-pain assessment scale .

[0010] As a preferred aspect of the invention, the specific steps of drawing a pain level assessment curve of various physiological data changing with pain level based on various physiological data of the patient under different pain levels and using cubic spline interpolation method are: Obtain 10 raw data points based on the patient's physiological data at different levels of pain ,in , Indicates pain level, Represents physiological data; Construct a cubic polynomial function between each adjacent original data point , for each subinterval , the form of the cubic spline interpolation function is: ,in ,and , , and is the unknown coefficient, which is determined by solving the three bending moment equations on each subinterval, and the boundary relationship adopts the natural spline, that is, and ; A cubic polynomial function through each subinterval Generate more intermediate data points , the original data points and the intermediate data points Connect them one by one to get a smooth pain level assessment curve.

[0011] As a preferred aspect of the invention, the pain assessment level corresponding to each physiological data is comprehensively calculated to obtain a pain physiological assessment score. The calculation formula is: .

[0012] As a preferred aspect of the invention, the trained facial expression pain assessment model is specifically a Faster-RCNN model, including an input layer, a feature extraction layer, an RPN layer, a RAL layer and an output layer, wherein the input layer is used to receive the patient's facial expression image; wherein the feature extraction layer is used to extract feature information in the patient's facial expression image and output the feature image, and the specific structure of the feature extraction layer is a ResNet-50 network model, including a convolutional pooling layer and four residual layers; wherein the RPN layer is used to generate a series of candidate regions on the feature image output by the feature extraction layer; wherein the RAL layer is used to map candidate boxes of different sizes generated by the RPN layer to a feature image of a fixed size; wherein the output layer is used to output the facial expression pain assessment result, that is, the pain expression assessment score .

[0013] As a preferred aspect of the invention, the specific steps of fine-tuning the trained facial expression pain assessment model according to the facial expression images of the patient when experiencing different levels of pain are: Perform image preprocessing on the facial expression images of patients when they experience different levels of pain, use LabelImg to annotate the facial areas in the facial expression images, and assign corresponding pain level labels to each annotated box; Freeze some parameters in the feature extraction layer of the facial expression pain assessment model, that is, keep the weight parameters of the convolutional pooling layer and the first three residual layers of the ResNet-50 network model unchanged; The labeled facial expression images were input into the facial expression pain assessment model with some parameters frozen for secondary training. That is, the weight parameters of the fourth residual layer, RPN layer, RAL layer and output layer in the ResNet-50 network model were adjusted according to the labeled facial expression images until the performance of the fine-tuned facial expression pain assessment model reached the set standard.

[0014] As a preferred aspect of the invention, the pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The calculation formula is: ,in Self-assessment score for pain The weight coefficient of Scoring the physiological assessment of pain The weight coefficient of Pain expression assessment score The weight coefficient of .

[0015] A patient pain assessment system, comprising a data acquisition module, a self-assessment correction module, a physiological assessment correction module, an expression assessment correction module and a pain assessment module; The data acquisition module is used to simulate different levels of pain for patients and obtain the facial expression images, various physiological data and the visual analog scale (VAS) scores filled in by patients when they experience different levels of pain. , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; Self-assessment correction module for pain simulation based on level The score of visual analogue scale (VAS) The difference between the two values ​​is calculated to obtain the patient's personalized self-assessment deviation , and based on this, the visual analogue scale (VAS) score filled out by the patient at the onset of the disease was Perform personalized correction to obtain pain self-assessment score ; The physiological assessment correction module is used to draw a pain level assessment curve based on the various physiological data of the patient under different pain levels, and obtain the pain assessment level corresponding to various physiological data based on the pain level assessment curve. The pain assessment levels corresponding to various physiological data are summed and averaged to obtain the pain physiological assessment score. ; The facial expression assessment correction module is used to fine-tune the trained facial expression pain assessment model according to the facial expression images of patients when they experience different levels of pain, and to perform pain assessment on the facial expression images collected when the patient is ill based on the fine-tuned facial expression pain assessment model to obtain a pain expression assessment score. ; Pain assessment module, used to self-assess pain scores , pain physiological assessment score and pain expression assessment score Perform weighted summation to obtain the patient's comprehensive pain score , and assign appropriate treatment plans to patients accordingly.

[0016] The present invention has the following advantages: The present invention calculates the patient's personalized self-assessment deviation , and based on this, the visual analogue scale (VAS) score filled in by the patient Personalized correction can effectively reduce the subjective differences between individual patients, making the final VAS score closer to the actual pain level, thereby more accurately reflecting the patient's true pain state and helping doctors make more accurate diagnoses, avoiding under- or over-treatment due to misdiagnosis, and improving the accuracy of the patient's pain assessment method and system.

[0017] The present invention draws a pain level assessment curve based on various physiological data of patients under different levels of pain, and obtains the pain physiological assessment score corresponding to the physiological data of the patient at the onset of the disease based on the pain level assessment curve. , which can provide a more objective pain assessment standard, reduce the subjective bias caused by individual differences, and formulate treatment plans based on individualized assessment curves to better meet the needs of specific patients and achieve true personalized medicine, thereby significantly improving the objectivity of pain assessment and improving the accuracy of the patient's pain assessment method and system.

[0018] The present invention fine-tunes the facial expression pain assessment model through the facial expression images of the patient under different levels of pain, and performs pain assessment on the patient based on the fine-tuned facial expression pain assessment model, so that the facial expression pain assessment model can better adapt to the characteristics of the patient and can more accurately capture the subtle differences in the patient's facial expressions, thereby improving the accuracy of pain assessment and avoiding misjudgment due to individual differences. At the same time, for patients who are unable to effectively express their pain feelings, pain assessment of the patients through the facial expression pain assessment model can also help doctors understand the patient's pain condition more accurately.

[0019] The present invention is based on the pain self-assessment score , pain physiological assessment score and pain expression assessment score Perform weighted summation to obtain the patient's comprehensive pain score , and based on this, the patient's pain assessment can be achieved in multiple dimensions, making the pain assessment results more comprehensive. The comprehensive pain score obtained by integrating multiple indicators can balance the differences between individual patients, making the assessment results more objective and reliable, so that doctors can more accurately judge the degree of pain and its changing trends, avoid deviations and misjudgments caused by a single indicator, and improve the assessment accuracy of this patient pain assessment method and system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of the structure of a patient pain assessment system used in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0022] Embodiment 1, a method for assessing pain in a patient, comprising the following steps: Use medical-grade low-intensity electrical stimulation equipment to simulate different levels of pain on patients, use cameras to capture images of patients' facial expressions when they experience different levels of pain, use medical monitoring equipment to measure patients' various physiological data under different levels of pain, and read the visual analog scale (VAS) scores filled out by patients under different levels of pain , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; It should be noted that the principle of the above-mentioned medical-grade low-intensity electrical stimulation device is to use weak electric current to stimulate the skin nerve endings to simulate slight tingling or burning sensations, and this device has many advantages such as non-invasiveness, strong adjustability and good portability. It is often used in pain research, biofeedback training and certain types of physical therapy. A specific example is the "labor pain simulator" used by some medical institutions and educational institutions.

[0023] Based on pain simulation level The score of visual analogue scale (VAS) The difference between the two is used to calculate the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score ; Based on the patient's physiological data under different levels of pain, the cubic spline interpolation method is used to draw the pain level assessment curve of each physiological data changing with the pain level. The patient's physiological data at the onset of the disease are matched with the pain level assessment curve to obtain the pain assessment level corresponding to each physiological data. The pain assessment level corresponding to each physiological data is comprehensively calculated to obtain the pain physiological assessment score. ; The trained facial expression pain assessment model is fine-tuned based on the facial expression images of patients when they experience different levels of pain. The facial expression images collected when the patient is ill are input into the fine-tuned facial expression pain assessment model to obtain the pain expression assessment score. ; Pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The patient's comprehensive pain score This is considered as the result of the pain assessment, and the patient is assigned an appropriate treatment plan based on it.

[0024] Specifically, for the pain composite score For patients with a score of 0-3, non-drug treatments such as hot compress, cold compress, massage and acupuncture are mainly used. Patients with a score of 4-6 need to strengthen non-drug treatment or use weak opioids and non-steroidal anti-inflammatory drugs for analgesia. Patients with a score of 7-10 need to use strong opioids or even a combination of multiple analgesics with different drug mechanisms for analgesic treatment.

[0025] Level of pain simulation The range is 0-9, that is , and the pain self-assessment score , pain physiological assessment score and pain expression assessment score The score range is 0-10.

[0026] The calculation obtains the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score The specific steps are: Calculate the level of pain simulation at different pain levels The score of visual analogue scale (VAS) The difference between the two values ​​was summed and averaged to obtain the patient's personalized self-assessment deviation. , the calculation formula is: ; The VAS score was calculated by the patient's and patients’ individualized self-assessment bias Sum the patient's pain self-assessment score , the calculation formula is: .

[0027] It should be noted that the total score range of the visual analog scale (VAS) is 0-10 points. To fill in the form, the patient only needs to mark the degree of pain he or she feels on a straight line, where one end of the straight line represents "no pain", which is 0 points, and the other end represents "the most severe pain", which is 10 points. Medical staff can read the corresponding score according to the patient's marked position. This way of filling in the form is simple, fast and reasonable. Patients do not need to endure pain to fill out various types of questionnaires, which improves the user experience.

[0028] The above steps are performed by calculating the patient's personalized self-assessment deviation , and based on this, the visual analogue scale (VAS) score filled in by the patient Personalized correction can effectively reduce the subjective differences between individual patients, making the final VAS score closer to the actual pain level, thereby more accurately reflecting the patient's true pain state and helping doctors make more accurate diagnoses, avoiding under- or over-treatment due to misdiagnosis, and improving the accuracy of the patient's pain assessment method and system.

[0029] The pain level assessment curves of the physiological data changing with the pain level include: heart rate-pain level curve , Blood pressure-pain level curve and respiratory rate-pain level curve The pain assessment levels corresponding to the various physiological data include: heart rate-pain assessment level , blood pressure-pain assessment level and respiratory rate-pain assessment scale .

[0030] The specific steps of drawing a pain level assessment curve of various physiological data changing with pain level based on various physiological data of the patient under different pain levels by using cubic spline interpolation method are as follows: Obtain 10 raw data points based on the patient's physiological data at different levels of pain ,in , Indicates pain level, Represents physiological data; Construct a cubic polynomial function between each adjacent original data point , for each subinterval , the form of the cubic spline interpolation function is: ,in ,and , , and is the unknown coefficient, which is determined by solving the three bending moment equations on each subinterval, and the boundary relationship adopts the natural spline, that is, and ; A cubic polynomial function through each subinterval Generate more intermediate data points , the original data points and the intermediate data points Connect them one by one to get a smooth pain level assessment curve.

[0031] The pain assessment level corresponding to each physiological data is comprehensively calculated to obtain the pain physiological assessment score The calculation formula is .

[0032] The above steps draw a pain level assessment curve based on the patient's physiological data at different levels of pain, and obtain the pain physiological assessment score corresponding to the patient's physiological data at the time of onset based on the pain level assessment curve. , which can provide a more objective pain assessment standard, reduce the subjective bias caused by individual differences, and formulate treatment plans based on individualized assessment curves to better meet the needs of specific patients and achieve true personalized medicine, thereby significantly improving the objectivity of pain assessment and improving the accuracy of the patient's pain assessment method and system.

[0033] The trained facial expression pain assessment model is specifically a Faster-RCNN model, including an input layer, a feature extraction layer, an RPN layer, a RAL layer and an output layer, wherein the input layer is used to receive the patient's facial expression image; wherein the feature extraction layer is used to extract feature information in the patient's facial expression image and output the feature image, and the specific structure of the feature extraction layer is a ResNet-50 network model, including a convolutional pooling layer and four residual layers; wherein the RPN layer is used to generate a series of candidate regions on the feature image output by the feature extraction layer; wherein the RAL layer is used to map the candidate boxes of different sizes generated by the RPN layer to the feature image of a fixed size; wherein the output layer is used to output the facial expression pain assessment result, that is, the pain expression assessment score .

[0034] The specific steps of fine-tuning the trained facial expression pain assessment model according to the facial expression images of the patient when experiencing different levels of pain are: Perform image preprocessing on the facial expression images of patients when they experience different levels of pain, use LabelImg to annotate the facial areas in the facial expression images, and assign corresponding pain level labels to each annotated box; Freeze some parameters in the feature extraction layer of the facial expression pain assessment model, that is, keep the weight parameters of the convolutional pooling layer and the first three residual layers of the ResNet-50 network model unchanged; The labeled facial expression images were input into the facial expression pain assessment model with some parameters frozen for secondary training. That is, the weight parameters of the fourth residual layer, RPN layer, RAL layer and output layer in the ResNet-50 network model were adjusted according to the labeled facial expression images until the performance of the fine-tuned facial expression pain assessment model reached the set standard.

[0035] The above steps fine-tune the facial expression pain assessment model through the facial expression images of the patient under different levels of pain, and assess the patient's pain based on the fine-tuned facial expression pain assessment model, so that the facial expression pain assessment model can better adapt to the characteristics of the patient and can more accurately capture the subtle differences in the patient's facial expressions, thereby improving the accuracy of pain assessment and avoiding misjudgment due to individual differences. At the same time, for patients who cannot effectively express their pain feelings, pain assessment of patients through the facial expression pain assessment model can also help doctors understand the patient's pain condition more accurately.

[0036] The pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The calculation formula is: ,in Self-assessment score for pain The weight coefficient of Scoring the physiological assessment of pain The weight coefficient of Pain expression assessment score The weight coefficient of .

[0037] It should be noted that the weight coefficient , and It can be flexibly adjusted according to the specific situation of each patient (such as medical history, disease characteristics, etc.) to achieve truly personalized pain assessment. For example, the weight coefficient can be reduced for patients who cannot effectively express their pain feelings (such as children, the elderly, or patients with language barriers or hand and limb defects). The value of The value is set to 0, and for patients with facial defects (such as facial paralysis, etc.), the weight coefficient can be reduced The value of The value of is set to 0.

[0038] The above steps were evaluated by pain self-assessment scores. , pain physiological assessment score and pain expression assessment score Perform weighted summation to obtain the patient's comprehensive pain score , and based on this, the patient's pain assessment can be achieved in multiple dimensions, making the pain assessment results more comprehensive. The comprehensive pain score obtained by integrating multiple indicators can balance the differences between individual patients, making the assessment results more objective and reliable, so that doctors can more accurately judge the degree of pain and its changing trends, avoid deviations and misjudgments caused by a single indicator, and improve the assessment accuracy of this patient pain assessment method and system.

[0039] Embodiment 2, a patient pain assessment system, such as Figure 1 As shown, it includes a data acquisition module, a self-assessment correction module, a physiological assessment correction module, an expression assessment correction module and a pain assessment module; The data acquisition module is used to simulate different levels of pain for patients and obtain the facial expression images, various physiological data and the visual analog scale (VAS) scores filled in by patients when they experience different levels of pain. , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; Self-assessment correction module for pain simulation based on level The score of visual analogue scale (VAS) The difference between the two values ​​is calculated to obtain the patient's personalized self-assessment deviation , and based on this, the visual analogue scale (VAS) score filled out by the patient at the onset of the disease was Perform personalized correction to obtain pain self-assessment score ; The physiological assessment correction module is used to draw a pain level assessment curve based on the various physiological data of the patient under different pain levels, and obtain the pain assessment level corresponding to various physiological data based on the pain level assessment curve. The pain assessment levels corresponding to various physiological data are summed and averaged to obtain the pain physiological assessment score. ; The facial expression assessment correction module is used to fine-tune the trained facial expression pain assessment model according to the facial expression images of patients when they experience different levels of pain, and to perform pain assessment on the facial expression images collected when the patient is ill based on the fine-tuned facial expression pain assessment model to obtain a pain expression assessment score. ; Pain assessment module, used to self-assess pain scores , pain physiological assessment score and pain expression assessment score Perform weighted summation to obtain the patient's comprehensive pain score , and assign appropriate treatment plans to patients accordingly.

[0040] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A method for assessing pain in a patient, characterized in that: The following steps are involved: Use medical-grade electrical stimulation equipment to simulate different levels of pain on patients, use cameras to capture images of patients' facial expressions when they experience different levels of pain, use medical monitoring equipment to measure patients' various physiological data under different levels of pain, and read the visual analog scale (VAS) scores filled out by patients under different levels of pain , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; Based on pain simulation level The score of visual analogue scale (VAS) The difference between the two is used to calculate the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score ; Based on the patient's physiological data under different levels of pain, the cubic spline interpolation method is used to draw the pain level assessment curve of each physiological data changing with the pain level. The patient's physiological data at the onset of the disease are matched with the pain level assessment curve to obtain the pain assessment level corresponding to each physiological data. The pain assessment level corresponding to each physiological data is comprehensively calculated to obtain the pain physiological assessment score. ; The trained facial expression pain assessment model is fine-tuned based on the facial expression images of patients when they experience different levels of pain. The facial expression images collected when the patient is ill are input into the fine-tuned facial expression pain assessment model to obtain the pain expression assessment score. ; Pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The patient's comprehensive pain score This is considered as the result of the pain assessment, and the patient is assigned an appropriate treatment plan based on it.

2. A method for assessing pain in a patient according to claim 1, characterized in that: Level of pain simulation The range is 0-9, that is , and the pain self-assessment score , pain physiological assessment score and pain expression assessment score The score range is 0-10.

3. A method for assessing pain in a patient according to claim 1, characterized in that: The calculation obtains the patient's personalized self-assessment deviation , based on the patient's personalized self-assessment of bias The score of the visual analogue scale (VAS) filled in by the patient at the onset of the disease Perform personalized correction to obtain pain self-assessment score The specific steps are: Calculate the level of pain simulation at different pain levels The score of visual analogue scale (VAS) The difference between the two values ​​was summed and averaged to obtain the patient's personalized self-assessment deviation. , the calculation formula is: ； The VAS score was calculated by the patient's and patients’ individualized self-assessment bias Sum the patient's pain self-assessment score , the calculation formula is: .

4. A method for assessing pain in a patient according to claim 1, characterized in that: The pain level assessment curves of the physiological data changing with the pain level include: heart rate-pain level curve , Blood pressure-pain level curve and respiratory rate-pain level curve The pain assessment levels corresponding to the various physiological data include: heart rate-pain assessment level , blood pressure-pain assessment level and respiratory rate-pain assessment scale .

5. A method for assessing pain in a patient according to claim 4, characterized in that: The specific steps of drawing a pain level assessment curve of various physiological data changing with pain level based on various physiological data of the patient under different pain levels by using cubic spline interpolation method are as follows: Obtain 10 raw data points based on the patient's physiological data at different levels of pain ,in , Indicates pain level, Represents physiological data; Construct a cubic polynomial function between each adjacent original data point , for each subinterval , the form of the cubic spline interpolation function is: ,in ,and , , and is the unknown coefficient, which is determined by solving the three bending moment equations on each subinterval, and the boundary relationship adopts the natural spline, that is, and ; A cubic polynomial function through each subinterval Generate more intermediate data points , the original data points and the intermediate data points Connect them one by one to get a smooth pain level assessment curve.

6. A method for assessing pain in a patient according to claim 4, characterized in that: The pain assessment level corresponding to each physiological data is comprehensively calculated to obtain the pain physiological assessment score The calculation formula is: .

7. A method for assessing pain in a patient according to claim 1, characterized in that: The trained facial expression pain assessment model is specifically a Faster-RCNN model, comprising an input layer, a feature extraction layer, an RPN layer, a RAL layer and an output layer, wherein the input layer is used to receive the patient's facial expression image; The feature extraction layer is used to extract feature information from the patient's facial expression image and output the feature image. The specific structure of the feature extraction layer is a ResNet-50 network model, including a convolutional pooling layer and four residual layers; the RPN layer is used to generate a series of candidate regions on the feature image output by the feature extraction layer; the RAL layer is used to map the candidate boxes of different sizes generated by the RPN layer to the feature image of a fixed size; the output layer is used to output the facial expression pain assessment result, that is, the pain expression assessment score .

8. A method for assessing pain in a patient according to claim 7, characterized in that: The specific steps of fine-tuning the trained facial expression pain assessment model according to the facial expression images of the patient when experiencing different levels of pain are: Perform image preprocessing on the facial expression images of patients when they experience different levels of pain, use LabelImg to annotate the facial areas in the facial expression images, and assign corresponding pain level labels to each annotated box; Freeze some parameters in the feature extraction layer of the facial expression pain assessment model, that is, keep the weight parameters of the convolutional pooling layer and the first three residual layers of the ResNet-50 network model unchanged; The labeled facial expression images were input into the facial expression pain assessment model with some parameters frozen for secondary training. That is, the weight parameters of the fourth residual layer, RPN layer, RAL layer and output layer in the ResNet-50 network model were adjusted according to the labeled facial expression images until the performance of the fine-tuned facial expression pain assessment model reached the set standard.

9. A method for assessing pain in a patient according to claim 1, characterized in that: The pain self-assessment score , pain physiological assessment score and pain expression assessment score Comprehensive calculation is performed to obtain the patient's comprehensive pain score The calculation formula is: ,in Self-assessment score for pain The weight coefficient of Scoring the physiological assessment of pain The weight coefficient of Pain expression assessment score The weight coefficient of .

10. A patient pain assessment system, applied to a patient pain assessment method according to any one of claims 1 to 9, characterized in that: It includes a data acquisition module, a self-assessment correction module, a physiological assessment correction module, an expression assessment correction module and a pain assessment module; The data acquisition module is used to simulate different levels of pain for patients and obtain the facial expression images, various physiological data and the visual analog scale (VAS) scores filled in by patients when they experience different levels of pain. , obtain the patient's facial expression images, various physiological data and visual analog scale VAS scores at the time of onset ; Self-assessment correction module for pain simulation based on level The score of visual analogue scale (VAS) The difference between the two values ​​is calculated to obtain the patient's personalized self-assessment deviation , and based on this, the visual analogue scale (VAS) score filled out by the patient at the onset of the disease was Perform personalized correction to obtain pain self-assessment score ; The physiological assessment correction module is used to draw a pain level assessment curve based on the various physiological data of the patient under different pain levels, and obtain the pain assessment level corresponding to various physiological data based on the pain level assessment curve. The pain assessment levels corresponding to various physiological data are summed and averaged to obtain the pain physiological assessment score. ; The facial expression assessment correction module is used to fine-tune the trained facial expression pain assessment model according to the facial expression images of patients when they experience different levels of pain, and to perform pain assessment on the facial expression images collected when the patient is ill based on the fine-tuned facial expression pain assessment model to obtain a pain expression assessment score. ; Pain assessment module, used to self-assess pain scores , pain physiological assessment score and pain expression assessment score Perform weighted summation to obtain the patient's comprehensive pain score , and assign appropriate treatment plans to patients accordingly.

Images