Physiotherapy system for pelvic floor rehabilitation of cervical cancer patients

Abstract

This invention provides a physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients, relating to the field of pelvic floor physiotherapy. The system includes: a data acquisition module for acquiring pelvic floor electromyography (EMG) and ultrasound data from multiple historical patients; a data analysis module for constructing an EMG atlas and an EMG-factor correlation atlas based on the aforementioned data; the data acquisition module is also used to acquire the current patient's pelvic floor EMG data based on the EMG atlas, and to acquire the current patient's pelvic floor ultrasound data based on the current patient's pelvic floor EMG data and the EMG-factor correlation atlas; a treatment plan determination module for determining the current patient's physiotherapy plan based on the current patient's pelvic floor EMG and ultrasound data; and a physiotherapy execution module for controlling the physiotherapy equipment to perform pelvic floor function rehabilitation physiotherapy based on the current patient's physiotherapy plan, thereby improving the flexibility and effectiveness of pelvic floor function rehabilitation physiotherapy for cervical cancer patients.

Description

Technical Field

[0001] This invention relates to the field of pelvic floor physiotherapy, and in particular to a physiotherapy system for the rehabilitation of pelvic floor function in patients with cervical cancer. Background Technology

[0002] Cervical cancer, one of the most common malignant tumors of the female reproductive system worldwide, seriously threatens women's health and quality of life. In recent years, with continuous advancements in medical technology, the early diagnosis rate and treatment outcomes of cervical cancer have significantly improved, extending patients' survival time. However, treatments for cervical cancer, such as surgery, radiotherapy, and chemotherapy, while effectively combating the tumor, inevitably cause varying degrees of damage to the patient's pelvic floor function. In terms of surgical treatment, radical hysterectomy is one of the commonly used surgical methods for cervical cancer. This surgery removes the uterus and surrounding tissues, including the cardinal ligaments, uterosacral ligaments, and other pelvic floor support structures, directly disrupting the normal anatomical structure and function of the pelvic floor. The cardinal and uterosacral ligaments play a crucial role in maintaining the normal position of the uterus and vagina; their removal leads to decreased pelvic floor support, resulting in pelvic organ prolapse, such as uterine prolapse and anterior and posterior vaginal wall prolapse. Furthermore, surgery may damage the nerves that control the pelvic floor muscles and organs, affecting the normal regulation of the pelvic floor muscles and leading to weakness and decreased coordination of pelvic floor muscle contractions, resulting in problems such as urinary incontinence and bowel dysfunction. Radiotherapy also plays an important role in the treatment of cervical cancer, especially for patients in the middle and late stages. However, radiotherapy can cause radiation damage to the pelvic floor tissues, leading to local tissue fibrosis and vascular lesions. Fibrosis reduces the elasticity and stiffness of the pelvic floor muscles and ligaments, affecting their normal contractile function; vascular lesions affect the blood supply to the pelvic floor tissues, leading to tissue malnutrition and further exacerbating the damage to pelvic floor function. Although chemotherapy mainly targets tumor cells throughout the body, it can also have some indirect effects on the pelvic floor tissues. Chemotherapy drugs may cause systemic symptoms such as weakness and fatigue, leading to reduced patient activity. Prolonged bed rest prevents the pelvic floor muscles from getting sufficient exercise, gradually causing them to atrophy and decline in function. Simultaneously, chemotherapy may also affect the patient's endocrine system, leading to hormonal imbalances and further affecting the physiological function of the pelvic floor tissues.

[0003] Because each cervical cancer patient's condition, treatment method, physical status, and degree of pelvic floor dysfunction vary, personalized pelvic floor rehabilitation physiotherapy plans are necessary. Currently, there is a lack of a system and method in clinical practice that can comprehensively utilize multiple test data to accurately assess a patient's pelvic floor function and develop a personalized physiotherapy plan. Existing rehabilitation programs often adopt a uniform treatment model, failing to meet the individualized needs of different patients, resulting in poor rehabilitation outcomes for some patients and impacting their confidence in treatment and quality of life.

[0004] Therefore, there is a need to provide a physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients, in order to improve the flexibility and effectiveness of pelvic floor function rehabilitation physiotherapy for cervical cancer patients. Summary of the Invention

[0005] This invention provides a physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients, comprising: a data acquisition module for acquiring pelvic floor electromyography (EMG) data from multiple historical patients; a data analysis module for constructing an EMG atlas based on the EMG data from multiple historical patients, wherein the EMG atlas is used to record the correlation between multiple pelvic floor EMG detection locations; the data acquisition module is also used to acquire pelvic floor ultrasound data from multiple historical patients; the data analysis module is further used to construct an EMG-factor correlation atlas based on the pelvic floor EMG data and pelvic floor ultrasound data from multiple historical patients, wherein the EMG... The electromyography-factor correlation map is used to record the correlation between pelvic floor electromyography (EMG) detection locations and pelvic floor function assessment factors. The data acquisition module is also used to acquire the current patient's pelvic floor EMG detection data based on the EMG detection map, and to acquire the current patient's pelvic floor ultrasound detection data based on the current patient's pelvic floor EMG detection data and the EMG-factor correlation map. The treatment plan determination module is used to determine the current patient's physiotherapy plan based on the current patient's pelvic floor EMG detection data and pelvic floor ultrasound detection data. The physiotherapy execution module is used to control the physiotherapy equipment and perform pelvic floor function rehabilitation physiotherapy on the current patient based on the current patient's physiotherapy plan.

[0006] Furthermore, the pelvic floor electromyography (EMG) detection data includes electrical signals from multiple pelvic floor detection locations under various contraction states. The data analysis module constructs an EMG detection atlas based on EMG detection data from multiple historical patients, including: filtering effective pelvic floor detection locations from multiple pelvic floor detection locations based on the EMG detection data from multiple historical patients; for each historical patient, extracting effective pelvic floor EMG detection data from the historical patient's EMG detection data based on the effective pelvic floor detection locations; clustering multiple historical patients based on the effective pelvic floor EMG detection data from multiple historical patients to determine multiple clusters, where each cluster includes multiple historical patients; for each cluster, determining the correlation between multiple effective pelvic floor detection locations corresponding to the cluster based on the effective pelvic floor EMG detection data from the multiple historical patients included in the cluster; and constructing an EMG detection atlas based on the correlation between multiple effective pelvic floor detection locations corresponding to each cluster.

[0007] Furthermore, the data analysis module determines the correlation between multiple valid pelvic floor electromyography (EMG) detection locations corresponding to a cluster based on the valid EMG detection data of multiple historical patients included in the cluster. This includes: determining the correlation between any two valid pelvic floor detection locations and each contraction state based on the valid EMG detection data of multiple historical patients included in the cluster; and determining the correlation between two pelvic floor detection locations corresponding to a cluster based on the correlation between the two pelvic floor detection locations and each contraction state for any two pelvic floor detection locations. The correlation between the multiple valid EMG detection data of a cluster includes the correlation between any two pelvic floor detection locations corresponding to the cluster.

[0008] Furthermore, the data analysis module constructs an electromyography (EMG) and factor correlation map based on pelvic floor electromyography (PEMG) and ultrasound data from multiple historical patients. This includes: for each cluster, determining the correlation between the PEMG detection location corresponding to the cluster and the PEM function assessment factors based on the valid PEMG and ultrasound data from multiple historical patients included in the cluster; and constructing an EMG and factor correlation map based on the correlation between the PEMG detection location corresponding to each cluster and the PEM function assessment factors.

[0009] Furthermore, the data analysis module, based on valid pelvic floor electromyography (EMG) and pelvic floor ultrasound data from multiple historical patients included in the cluster, determines the correlation between the EMG locations corresponding to the cluster and pelvic floor function assessment factors. This includes: determining the correlation between any valid pelvic floor location and any pelvic floor function assessment factor for each contraction state based on the valid EMG and ultrasound data from multiple historical patients included in the cluster; and for any valid pelvic floor location and any pelvic floor function assessment factor, calculating the correlation between the valid pelvic floor location and the pelvic floor function assessment factor for each contraction state based on the correlation between the valid pelvic floor location and the pelvic floor function assessment factor. The correlation between the EMG locations corresponding to the cluster and the pelvic floor function assessment factors includes the correlation between any valid pelvic floor location and any pelvic floor function assessment factor.

[0010] Furthermore, the data acquisition module acquires the current patient's pelvic floor electromyography (EMG) data based on EMG atlases, including: determining multiple differentiated pelvic floor detection locations from valid pelvic floor detection locations based on EMG data from multiple historical patients; acquiring the current patient's initial EMG data based on the multiple differentiated pelvic floor detection locations; determining the cluster to which the current patient belongs based on the current patient's initial EMG data; acquiring the current patient's pelvic floor EMG data based on the current patient's cluster and EMG atlas; and acquiring the current patient's pelvic floor ultrasound data based on the current patient's pelvic floor EMG data.

[0011] Furthermore, the data acquisition module acquires the pelvic floor electromyography (EMG) data of the current patient based on the cluster to which the current patient belongs and the EMG detection atlas, including: determining the key pelvic floor detection location corresponding to each cluster from the valid pelvic floor detection locations based on the EMG detection atlas; and acquiring the pelvic floor EMG data of the current patient based on the key pelvic floor detection location corresponding to the cluster to which the current patient belongs.

[0012] Furthermore, the data acquisition module acquires the pelvic floor ultrasound data of the current patient based on the current patient's pelvic floor electromyography (EMG) data and EMG-factor correlation map, including: determining the key pelvic floor function assessment factors of the current patient based on the current patient's pelvic floor EMG data and EMG-factor correlation map; and acquiring the current patient's pelvic floor ultrasound data based on the key pelvic floor function assessment factors.

[0013] Furthermore, the treatment plan determination module determines the current patient's physical therapy plan based on the current patient's pelvic floor electromyography (EMG) data and pelvic floor ultrasound data, including: identifying similar historical patients based on the current patient's EMG and ultrasound data; and determining the current patient's physical therapy plan based on similar historical patients.

[0014] Furthermore, the scheme determination module identifies similar historical patients based on the current patient's pelvic floor electromyography (EMG) and pelvic floor ultrasound data, including: identifying similar historical patients based on the current patient's EMG and ultrasound data, the weights of key pelvic floor detection locations corresponding to the cluster to which the current patient belongs, and the weights of key pelvic floor function assessment factors for the current patient.

[0015] Compared with existing technologies, the physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients provided by this invention has at least the following beneficial effects: By collecting pelvic floor electromyography (EMG) and ultrasound data from historical patients, we constructed an EMG atlas and an EMG-factor correlation atlas to deeply explore the correlation between pelvic floor EMG detection locations and their relationship with pelvic floor function assessment factors, providing a data foundation for precise rehabilitation.

[0016] Based on the constructed atlas, it can accurately acquire the current patient's pelvic floor electromyography and ultrasound detection data, reducing the blindness and error of detection and improving the efficiency and accuracy of data acquisition.

[0017] Personalized physiotherapy plans are determined based on the current patient's test data, taking into full account individual differences, making rehabilitation physiotherapy more targeted and effective. Attached Figure Description

[0018] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein: Figure 1 This is a schematic diagram of a physiotherapy system for pelvic floor function rehabilitation in patients with cervical cancer, as shown in some embodiments of this specification. Figure 2 This is a schematic diagram of electromyography (EMG) detection atlases shown in some embodiments of this specification; Figure 3 This is a schematic diagram of electromyography and factor correlation atlases shown in some embodiments of this specification. Detailed Implementation

[0019] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0020] Figure 1 This is a schematic diagram of a physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients, as shown in some embodiments of this specification. Figure 1 As shown, a physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients may include a data acquisition module, a data analysis module, a treatment plan determination module, and a physiotherapy execution module.

[0021] The data acquisition module is used to acquire pelvic floor electromyography (EMG) data from multiple historical patients.

[0022] The pelvic floor electromyography (EMG) data includes electrical signals from multiple pelvic floor detection locations under various contraction states.

[0023] Specifically, the pelvic floor muscles consist of multiple layers of muscles and connective tissue. Multiple pelvic floor testing locations need to cover key functional areas to comprehensively assess pelvic floor support, urinary continence, defecation continence, and sexual function. As examples only, multiple pelvic floor testing locations may include the perineal central tendon, bulbospongiosus muscle, ischiocavernosus muscle, superficial transverse perineal muscle, puborectalis muscle, and iliococcygeus muscle.

[0024] Multiple contraction states need to simulate everyday functional scenarios, covering three categories: static, dynamic, and induced contraction. For example: 1. Static contraction: Maximum voluntary contraction: The patient contracts their pelvic floor muscles as much as possible for 5-10 seconds, and the peak amplitude is recorded. This state is used to assess maximum muscle strength. For example, patients with stress urinary incontinence may have weakened pelvic floor muscles (MVC) and be unable to effectively close the urethra, leading to urinary leakage when abdominal pressure increases.

[0025] Sustained isometric contraction: The patient contracts the pelvic floor muscles with 50% of their maximum force for 30 seconds, and the degree of muscle fatigue is recorded. This state is used to assess muscle endurance. For example, patients with chronic pelvic pain may experience premature muscle fatigue due to sustained contraction, and a gradual decrease in amplitude can be observed during the test.

[0026] 2. Dynamic contraction: Rapid contraction-relaxation cycle: The patient rapidly contracts the pelvic floor muscles (1 second) and immediately relaxes (2 seconds), repeating 5-10 times. This state is used to assess muscle response speed and coordination. For example, patients with urinary urgency and incontinence may experience leakage because the rapid contraction is delayed and the urethra cannot be closed in time.

[0027] Stepwise contraction: The patient gradually increases the contraction strength (25%, 50%, 75%, 100% MVC), holding each strength for 5 seconds. This state is used to assess muscle strength grading ability. For example, postpartum recovery patients may experience fluctuations in contraction strength due to insufficient grading control, affecting the recovery of pelvic floor function.

[0028] 3. Induces contraction: Cough-induced contraction: The automatic contraction response of the pelvic floor muscles is detected when the patient coughs. This state is used to assess the function of neurological reflexes. For example, patients with neurogenic bladder may have impaired reflex arcs, resulting in no contraction of the pelvic floor muscles during coughing, which exacerbates urinary leakage.

[0029] Valsalva maneuver-induced contraction: The patient holds their breath and pushes downwards (simulating defecation), testing the pelvic floor muscles' ability to resist abdominal pressure. This state is used to assess pelvic floor support function; for example, in patients with pelvic organ prolapse, insufficient resistance may lead to a worsening of the prolapse.

[0030] Balloon dilation-induced contraction: A balloon is slowly dilated via the vagina or rectum to detect the pelvic floor muscle response to mechanical stimulation. This state is used to assess sensorimotor feedback function; for example, patients with interstitial cystitis may be hypersensitive, causing even slight balloon dilation to trigger strong contractions and induce pain.

[0031] Electrode patches can be set at each pelvic floor detection location to collect electrical signals at the pelvic floor detection location under various contraction states.

[0032] The data analysis module is used to construct an electromyography (EMG) atlas based on EMG data from multiple historical patients.

[0033] Among them, the electromyography (EMG) atlas is used to record the correlation between multiple pelvic floor EMG detection locations.

[0034] Specifically, it includes: Based on pelvic floor electromyography data from multiple historical patients, effective pelvic floor detection locations were selected from multiple pelvic floor detection locations. For each historical patient, effective pelvic floor electromyography (EMG) data is extracted from the historical patient's pelvic floor EMG data based on the effective pelvic floor detection location. The effective pelvic floor EMG data may include electrical signals at each effective pelvic floor detection location in multiple contraction states. Based on effective pelvic floor electromyography data from multiple historical patients, multiple historical patients were clustered to identify multiple clusters, where each cluster includes multiple historical patients; For each cluster, based on the valid pelvic floor electromyography data of multiple historical patients included in the cluster, the correlation between multiple valid pelvic floor detection locations corresponding to the cluster is determined; Based on the correlation between multiple valid pelvic floor detection locations corresponding to each cluster, an electromyography (EMG) detection atlas is constructed.

[0035] Specifically, the following process can be used to select effective pelvic floor detection locations: S11. For each historical patient and each pelvic floor detection location, based on the electrical signals of the pelvic floor detection location in various contraction states of the historical patient, construct a first electrical signal vector corresponding to the historical patient for that pelvic floor detection location, wherein one element of the first electrical signal vector is the electrical signal of the pelvic floor detection location in a contraction state. S12. For each pelvic floor detection location, calculate the Euclidean distance between the first electrical signal vectors of any two historical patients corresponding to the pelvic floor detection location. S13. For each pelvic floor detection location, calculate the standard deviation of the Euclidean distance between the first electrical signal vectors of any two historical patients corresponding to the pelvic floor detection location, and use it as the standard deviation of the electrical signal vector corresponding to the pelvic floor detection location. S14. For each pelvic floor detection location, determine whether the standard deviation of the electrical signal vector corresponding to the pelvic floor detection location is greater than the standard deviation threshold (e.g., 0.5). If so, the pelvic floor detection location is taken as a valid pelvic floor detection location.

[0036] Multiple clusters can be identified using the following process: S21. For each historical patient and each effective pelvic floor detection location, based on the electrical signals of the effective pelvic floor detection location of the historical patient in various contraction states, construct a second electrical signal vector corresponding to the effective pelvic floor detection location for the historical patient, wherein one element of the second electrical signal vector is the electrical signal of the effective pelvic floor detection location in a contraction state. S22. For each historical patient, based on the second electrical signal vector corresponding to each effective pelvic floor detection position, construct the electrical signal matrix corresponding to the historical patient, where one row vector of the electrical signal matrix is ​​the second electrical signal vector corresponding to an effective pelvic floor detection position. S23. For any two historical patients, calculate the Euclidean distance between the electrical signal matrices corresponding to the two historical patients; S24. Using an improved clustering algorithm, multiple historical patients are clustered based on the Euclidean distance between the electrical signal matrices of any two historical patients to determine multiple clusters.

[0037] By way of example only, clustering multiple historical patients based on the Euclidean distance between the electrical signal matrices of any two historical patients using an improved clustering algorithm may include the following steps: S241. Determine k historical patients from multiple historical patients as initial cluster centers, where k is a positive integer and k is less than the total number of historical patients. Execute S242. S242. For each historical patient, assign the historical patient to the cluster containing the cluster center with the smallest Euclidean distance in the electrical signal matrix, and execute S243. S243. Determine whether all historical patients have been assigned. If yes, proceed to S244; otherwise, proceed to S242. S244. For each cluster, calculate the variance of the Euclidean distance between the electrical signal matrices corresponding to any two historical patients in the cluster, and use it as the intra-cluster variance. Then execute S245. S245. Determine whether there exists at least one cluster whose intra-cluster variance is greater than the intra-cluster variance threshold (e.g., 0.3). If yes, proceed to S246; otherwise, proceed to S247. S246. Select the clusters with intra-cluster variance greater than the intra-cluster variance threshold as the target clusters, determine the sum of Euclidean distances corresponding to each historical patient in the target cluster, select the historical patient with the largest sum of Euclidean distances that is not the cluster center as the new cluster center, and execute S242. Here, the sum of Euclidean distances corresponding to the historical patients is the sum of the Euclidean distances between the historical patient and the electrical signal matrices of each other historical patient included in the cluster. S247. Determine whether there exists at least one cluster containing a number of historical patients that is less than a preset number threshold (e.g., 10). If yes, proceed to S248; otherwise, proceed to S249. S248. Clusters with fewer than a preset threshold number of historical patients are identified as abnormal clusters. The cluster center of the abnormal cluster is removed. S242 is then executed. S249. Complete clustering.

[0038] In some embodiments, the data analysis module determines the correlation between multiple valid pelvic floor electromyography (EMG) locations corresponding to a cluster based on valid EMG data from multiple historical patients included in the cluster, including: Based on valid pelvic floor electromyography data from multiple historical patients included in the cluster, the correlation between any two valid pelvic floor detection locations and each contraction state is determined. For any two pelvic floor detection locations, the correlation between the two pelvic floor detection locations and each contraction state is determined based on the correlation between the two pelvic floor detection locations and each contraction state. The correlation between multiple valid pelvic floor electromyography data corresponding to a cluster includes the correlation between any two pelvic floor detection locations corresponding to the cluster.

[0039] Specifically, for any two valid pelvic floor detection locations and each contraction state, the electrical signals of each historical patient in that contraction state included in the cluster corresponding to the two valid pelvic floor detection locations are used as two variables. The correlation between the two valid pelvic floor detection locations and that contraction state is calculated using a correlation coefficient algorithm (e.g., Pearson correlation coefficient, Kendall rank correlation coefficient, etc.).

[0040] For any two valid pelvic floor detection locations, the average correlation between the two valid pelvic floor detection locations and each contraction state can be used as the correlation between the two pelvic floor detection locations corresponding to the cluster.

[0041] For any two valid pelvic floor detection locations, if the correlation between the two pelvic floor detection locations corresponding to a cluster is greater than a correlation threshold (e.g., 0.6), then the two pelvic floor detection locations are associated in that cluster. Figure 2 This is a schematic diagram of electromyography (EMG) detection atlases shown in some embodiments of this specification, such as... Figure 2 As shown, the electromyography (EMG) detection atlas includes multiple subgraphs, each corresponding to a cluster. In the subgraphs, nodes represent valid pelvic floor detection locations. If two valid pelvic floor detection locations are associated in a cluster, the corresponding nodes of the two valid pelvic floor detection locations in the subgraph corresponding to that cluster can be connected by edges.

[0042] By screening effective detection locations and extracting multi-state electrical signals, interference from low-quality data is eliminated, ensuring that the analysis is based on reliable signals and improving the accuracy of subsequent correlation calculations. Patients are clustered based on effective data to form clusters with homogeneous characteristics, making correlation analysis more aligned with the physiological / pathological characteristics of specific patient groups (such as postpartum and elderly patients). The dynamic correlations of detection locations under different contraction states are captured, revealing multi-dimensional patterns of pelvic floor muscle coordination and providing richer quantitative evidence for functional assessment. Electromyography atlases record the correlation networks between detection locations in a standardized format, supporting rapid clinical comparison of abnormal patterns, optimization of electrode placement, and providing a reusable data structure for research.

[0043] The data acquisition module is also used to acquire pelvic floor ultrasound data from multiple historical patients.

[0044] Specifically, the pelvic floor ultrasound data of historical patients can include data corresponding to multiple pelvic floor function assessment factors. These factors may include at least bladder wall thickness, urethral length, levator ani hiatus area, distance between bladder neck and pubic symphysis, urethral rotation angle, residual urine volume, and degree of bladder neck opening at maximum urinary flow rate.

[0045] The data analysis module is also used to construct an electromyography-factor correlation map based on pelvic floor electromyography and pelvic floor ultrasound data from multiple historical patients.

[0046] Among them, the electromyography-factor correlation atlas is used to record the correlation between the location of pelvic floor electromyography detection and pelvic floor function assessment factors.

[0047] Specifically, it includes: For each cluster, based on the valid pelvic floor electromyography (EMG) and pelvic floor ultrasound data of multiple historical patients included in the cluster, the correlation between the pelvic floor EMG locations corresponding to the cluster and pelvic floor function assessment factors is determined. Based on the correlation between the pelvic floor electromyography (EMG) detection locations corresponding to each cluster and the pelvic floor function assessment factors, an EMG-factor correlation map was constructed.

[0048] In some embodiments, the data analysis module determines the association between the pelvic floor electromyography (EMG) locations corresponding to a cluster and pelvic floor function assessment factors based on valid pelvic floor EMG and pelvic floor ultrasound data from multiple historical patients included in the cluster, including: Based on valid pelvic floor electromyography and pelvic floor ultrasound data from multiple historical patients included in the cluster, the correlation between any valid pelvic floor detection location and any pelvic floor function assessment factor corresponding to each contraction state is determined. For any valid pelvic floor detection location and any pelvic floor function assessment factor, the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor for each contraction state is calculated based on the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor. The association between the pelvic floor electromyography detection location corresponding to the cluster and the pelvic floor function assessment factor includes the correlation between any valid pelvic floor detection location and any pelvic floor function assessment factor.

[0049] Specifically, for each valid pelvic floor detection location, each pelvic floor function assessment factor, and each contraction state, the electrical signal of each historical patient in the contraction state corresponding to the cluster of valid pelvic floor detection locations and the factor value of each historical patient in the cluster of each pelvic floor function assessment factor are used as two variables. The correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor corresponding to the contraction state is calculated using a correlation coefficient algorithm (e.g., Pearson correlation coefficient, Kendall rank correlation coefficient, etc.).

[0050] For each valid pelvic floor detection location and each pelvic floor function assessment factor, the mean value of the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor for each contraction state is calculated, and this value is used as the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor for the corresponding cluster.

[0051] For any valid pelvic floor detection location and any pelvic floor function assessment factor, if the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor corresponding to a cluster is greater than a correlation threshold (e.g., 0.6), then the valid pelvic floor detection location and the pelvic floor function assessment factor are associated in that cluster. Figure 3 This is a schematic diagram of electromyography and factor correlation atlases shown in some embodiments of this specification, such as... Figure 3 As shown, the electromyography-factor association map includes multiple subgraphs, each corresponding to a cluster. In the subgraph, one type of node represents a valid pelvic floor detection location, and the other type of node represents a pelvic floor function assessment factor. If a valid pelvic floor detection location and a pelvic floor function assessment factor are associated in a certain cluster, then the corresponding node of the valid pelvic floor detection location and the pelvic floor function assessment factor in the subgraph corresponding to that cluster can be connected by an edge.

[0052] By combining electromyographic signals with pelvic floor function assessment factors, cross-validation of electrophysiological and anatomical functional factors can be achieved, thereby improving the comprehensiveness of pelvic floor dysfunction assessment.

[0053] The data acquisition module is also used to acquire the current patient's pelvic floor electromyography data based on the electromyography detection atlas, and to acquire the current patient's pelvic floor ultrasound detection data based on the current patient's pelvic floor electromyography data and the electromyography-factor correlation atlas.

[0054] In some embodiments, the data acquisition module acquires the current patient's pelvic floor electromyography (EMG) data based on an EMG atlas, including: Based on pelvic floor electromyography data from multiple historical patients, several differential pelvic floor detection locations were identified from the effective pelvic floor detection locations. Based on multiple differentiated pelvic floor detection locations, the initial pelvic floor electromyography (EMG) data of the current patient is obtained. The initial EMG data of the current patient includes electrical signals of each differentiated pelvic floor detection location in multiple contraction states. Based on the current patient's initial pelvic floor electromyography data, determine the cluster to which the current patient belongs; Based on the current patient's cluster and electromyography (EMG) atlas, obtain the current patient's pelvic floor EMG data.

[0055] Specifically, multiple differentiated pelvic floor detection locations can be determined according to the following process: S31. For each cluster and each valid pelvic floor detection location, calculate the mean of the Euclidean distance between the second electrical signal vectors of any two historical patients included in the cluster corresponding to the pelvic floor detection location, and use it as the mean of the Euclidean distance between the pelvic floor detection location and the cluster. Valid pelvic floor detection locations with a mean less than the mean threshold (e.g., 2) are selected as candidate pelvic floor detection locations corresponding to the cluster. S32. For each valid pelvic floor detection location, determine the candidate ratio of valid pelvic floor detection locations based on the candidate pelvic floor detection locations corresponding to each cluster. The candidate ratio is the ratio of the number of clusters that use the valid pelvic floor detection location as a candidate pelvic floor detection location to the total number of clusters. Valid pelvic floor detection locations with a candidate ratio less than the ratio threshold (e.g., 40%) are selected as pelvic floor detection locations to be screened. S33. For each pelvic floor detection location to be screened and each cluster, the mean value of the second electrical signal vector of each historical patient included in the cluster corresponding to the pelvic floor detection location to be screened is used as the mean value vector of the second electrical signal of the pelvic floor detection location to be screened for the corresponding cluster. S33. For each pelvic floor detection location to be screened, calculate the Euclidean distance between the mean vectors of the second electrical signals of any two clusters corresponding to the pelvic floor detection location to be screened, and calculate the variance to obtain the distance variance corresponding to the pelvic floor detection location to be screened. S34. Sort the pelvic floor detection locations to be screened according to the distance variance corresponding to each location, and select the top n (e.g., 5, 6, etc.) locations as the differentiated pelvic floor detection locations.

[0056] Based on the initial pelvic floor electromyography (EMG) data of the current patient, a differential electrical signal matrix can be constructed. In this matrix, each row vector is a second electrical signal vector representing a differential pelvic floor detection location.

[0057] For each cluster, based on the second electrical signal mean vector corresponding to each pelvic floor detection location to be screened, a differential electrical signal mean matrix corresponding to the cluster is constructed, wherein one row vector of the differential electrical signal mean matrix is ​​the second electrical signal mean vector of a differential pelvic floor detection location.

[0058] For each cluster, calculate the Euclidean distance between the differential electrical signal matrix of the current patient and the mean differential electrical signal matrix of the corresponding cluster, and select the cluster with the smallest Euclidean distance as the cluster to which the current patient belongs.

[0059] In some embodiments, the data acquisition module acquires the pelvic floor electromyography (EMG) data of the current patient based on the cluster to which the current patient belongs and the EMG atlas, including: Based on electromyography (EMG) atlases, key pelvic floor detection locations corresponding to each cluster were identified from the valid pelvic floor detection locations. Based on the key pelvic floor detection locations corresponding to the current patient's cluster, obtain the current patient's pelvic floor electromyography (EMG) data.

[0060] Specifically, for each cluster and each valid pelvic floor detection location, the variance of the Euclidean distance between the second electrical signal vectors of any two historical patients included in the cluster is used. Valid pelvic floor detection locations with a variance greater than a variance threshold (e.g., 0.8) are designated as key pelvic floor detection locations corresponding to the cluster. Based on the electromyography (EMG) atlas, valid pelvic floor detection locations connected to the key pelvic floor detection locations corresponding to the cluster are also designated as key pelvic floor detection locations.

[0061] The current patient's pelvic floor electromyography data can include electrical signals from each cluster corresponding to each key pelvic floor detection location in various contraction states.

[0062] In some embodiments, the data acquisition module acquires pelvic floor ultrasound data of the current patient based on the current patient's pelvic floor electromyography (EMG) data and EMG-factor correlation atlas, including: Based on the current patient's pelvic floor electromyography (EMG) data and EMG-factor correlation map, the key pelvic floor function assessment factors for the current patient were identified. Based on the key pelvic floor function assessment factors of the current patient, obtain the current patient's pelvic floor ultrasound examination data.

[0063] Specifically, pelvic floor electromyography (EMG) data of patients without pelvic floor dysfunction were obtained. The EMG data of patients without pelvic floor dysfunction included electrical signals at each valid pelvic floor testing location in multiple contraction states.

[0064] Based on the pelvic floor electromyography (EMG) data of patients without pelvic floor dysfunction and the current patient's pelvic floor EMG data, abnormal key pelvic floor detection locations are identified. For example, for each key pelvic floor detection location, the Euclidean distance between the second electrical signal vector of the patient without pelvic floor dysfunction and the second electrical signal vector of the current patient is calculated. Key pelvic floor detection locations with an Euclidean distance greater than a Euclidean distance threshold (e.g., 2, etc.) are identified as abnormal key pelvic floor detection locations.

[0065] Based on the electromyography and factor correlation map, the effective pelvic floor detection locations corresponding to the abnormal key pelvic floor detection locations are identified and connected to the pelvic floor function assessment factors in the subgraph corresponding to the cluster to which the current patient belongs. These factors are then used as the key pelvic floor function assessment factors for the current patient.

[0066] The current patient's pelvic floor ultrasound data can include factor values ​​for each key pelvic floor function assessment factor.

[0067] The treatment plan determination module is used to determine the current patient's physical therapy plan based on the current patient's pelvic floor electromyography and pelvic floor ultrasound data.

[0068] Specifically, it includes: Based on the current patient’s pelvic floor electromyography and pelvic floor ultrasound data, identify similar historical patients; Based on similar patient histories, determine the current patient's physical therapy plan.

[0069] In some embodiments, the protocol determination module identifies similar historical patients based on the current patient's pelvic floor electromyography (EMG) data and pelvic floor ultrasound data, including: Based on the current patient's pelvic floor electromyography and pelvic floor ultrasound data, as well as the weights of key pelvic floor detection locations corresponding to the current patient's cluster and the weights of key pelvic floor function assessment factors for the current patient, similar historical patients are identified.

[0070] Specifically, the weight of the key pelvic floor detection location corresponding to the current patient's cluster can be determined based on the Euclidean distance between the second electrical signal vector of the key pelvic floor detection location corresponding to patients without pelvic floor dysfunction and the second electrical signal vector of the current patient. The larger the Euclidean distance, the greater the weight of the key pelvic floor detection location. For example, the sum of the Euclidean distances between the second electrical signal vectors of all key pelvic floor detection locations corresponding to patients without pelvic floor dysfunction and the second electrical signal vector of the current patient can be calculated. For each key pelvic floor detection location, the ratio of the Euclidean distance between the key pelvic floor detection location and the second electrical signal vector of the current patient to the sum of the Euclidean distances can be used as the weight of the key pelvic floor detection location corresponding to the current patient's cluster.

[0071] The weights of key pelvic floor function assessment factors for the current patient can be determined based on the weights of the key pelvic floor detection locations connected to them. The greater the weight of the connected key pelvic floor detection locations, the greater the weight of the key pelvic floor function assessment factor for the current patient. For example, for each key pelvic floor function assessment factor for the current patient, the mean weight of the key pelvic floor detection locations connected to the key pelvic floor function assessment factor can be calculated as the mean weight corresponding to the key pelvic floor function assessment factor. The ratio of the mean weight corresponding to each key pelvic floor function assessment factor to the sum of the mean weights corresponding to each key pelvic floor function assessment factor can be used as the weight of the key pelvic floor function assessment factor for the current patient.

[0072] For each key pelvic floor detection location corresponding to the cluster to which the current patient belongs, calculate the Euclidean distance between the second electrical signal vector of the historical patients corresponding to the key pelvic floor detection location and the second electrical signal vector of the current patient.

[0073] For the key pelvic floor function assessment factors of the current patient, calculate the factor value difference between the factor value of the key pelvic floor function assessment factor and the factor value of the current patient.

[0074] Based on the weights of key pelvic floor detection locations corresponding to the current patient's cluster and the weights of key pelvic floor function assessment factors for the current patient, a weighted sum is calculated of the Euclidean distance between the second electrical signal vectors of historical patients and the second electrical signal vector of the current patient corresponding to each key pelvic floor detection location, and the factor value difference of each key pelvic floor function assessment factor for the current patient. The difference between the current patient and historical patients is then calculated. The historical patient with the smallest difference value is considered a similar historical patient.

[0075] The physical therapy plans of patients with similar histories can be used as the physical therapy plans of the current patients. The physical therapy plans of the current patients can include physical therapy methods (e.g., electrical stimulation therapy, magnetic stimulation therapy, etc.) and physical therapy parameters (e.g., stimulation location, stimulation frequency and waveform, stimulation intensity and time, etc. of electrical stimulation therapy).

[0076] The physiotherapy execution module is used to control the physiotherapy equipment based on the current patient's physiotherapy plan, and to perform pelvic floor function rehabilitation physiotherapy on the current patient.

[0077] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients, characterized in that, include: The data acquisition module is used to acquire pelvic floor electromyography data from multiple historical patients. The data analysis module is used to construct an electromyography (EMG) atlas based on pelvic floor EMG data from multiple historical patients. The EMG atlas is used to record the correlation between multiple pelvic floor EMG detection locations. The data acquisition module is also used to acquire pelvic floor ultrasound detection data from multiple historical patients; The data analysis module is also used to construct an electromyography-factor correlation map based on pelvic floor electromyography and pelvic floor ultrasound data from multiple historical patients. The electromyography-factor correlation map is used to record the correlation between pelvic floor electromyography detection locations and pelvic floor function assessment factors. The data acquisition module is also used to acquire the current patient's pelvic floor electromyography data based on the electromyography detection atlas, and to acquire the current patient's pelvic floor ultrasound detection data based on the current patient's pelvic floor electromyography data and the electromyography-factor correlation atlas. The treatment plan determination module is used to determine the current patient's physical therapy plan based on the current patient's pelvic floor electromyography data and pelvic floor ultrasound data. The physiotherapy execution module is used to control the physiotherapy equipment based on the current patient's physiotherapy plan, and to perform pelvic floor function rehabilitation physiotherapy on the current patient.

2. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 1, characterized in that, The pelvic floor electromyography (EMG) data includes electrical signals from multiple pelvic floor detection locations under various contraction states. The data analysis module constructs an electromyography (EMG) atlas based on pelvic floor EMG data from multiple historical patients, including: Based on pelvic floor electromyography data from multiple historical patients, effective pelvic floor detection locations were selected from multiple pelvic floor detection locations. For each historical patient, valid pelvic floor electromyography (EMG) data are extracted from the historical patient's pelvic floor EMG data based on the valid pelvic floor detection location. Based on effective pelvic floor electromyography data from multiple historical patients, multiple historical patients were clustered to identify multiple clusters, where each cluster includes multiple historical patients; For each cluster, based on the valid pelvic floor electromyography data of multiple historical patients included in the cluster, the correlation between multiple valid pelvic floor detection locations corresponding to the cluster is determined; Based on the correlation between multiple valid pelvic floor detection locations corresponding to each cluster, an electromyography (EMG) detection atlas is constructed.

3. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 2, characterized in that, The data analysis module, based on valid pelvic floor electromyography (EMG) data from multiple historical patients included in the cluster, determines the correlation between multiple valid pelvic floor detection locations corresponding to the cluster, including: Based on valid pelvic floor electromyography data from multiple historical patients included in the cluster, the correlation between any two valid pelvic floor detection locations and each contraction state is determined. For any two pelvic floor detection locations, the correlation between the two pelvic floor detection locations and each contraction state is determined based on the correlation between the two pelvic floor detection locations and each contraction state. The correlation between the multiple valid pelvic floor electromyography detection data corresponding to the cluster includes the correlation between any two pelvic floor detection locations corresponding to the cluster.

4. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 3, characterized in that, The data analysis module constructs an electromyography (EMG) and factor correlation map based on pelvic floor electromyography (EMG) and pelvic floor ultrasound data from multiple historical patients, including: For each cluster, based on the valid pelvic floor electromyography (EMG) and pelvic floor ultrasound data of multiple historical patients included in the cluster, the correlation between the pelvic floor EMG locations corresponding to the cluster and pelvic floor function assessment factors is determined. Based on the correlation between the pelvic floor electromyography (EMG) detection locations corresponding to each cluster and the pelvic floor function assessment factors, an EMG-factor correlation map was constructed.

5. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 4, characterized in that, The data analysis module, based on valid pelvic floor electromyography (EMG) and pelvic floor ultrasound data from multiple historical patients included in the cluster, determines the correlation between the corresponding pelvic floor EMG locations and pelvic floor function assessment factors, including: Based on valid pelvic floor electromyography and pelvic floor ultrasound data from multiple historical patients included in the cluster, the correlation between any valid pelvic floor detection location and any pelvic floor function assessment factor corresponding to each contraction state is determined. For any valid pelvic floor detection location and any pelvic floor function assessment factor, the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor for each contraction state is calculated based on the correlation between the valid pelvic floor detection location and the pelvic floor function assessment factor. The association between the pelvic floor electromyography detection location corresponding to the cluster and the pelvic floor function assessment factor includes the correlation between any valid pelvic floor detection location and any pelvic floor function assessment factor.

6. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 2, characterized in that, The data acquisition module acquires the current patient's pelvic floor electromyography data based on electromyography atlases, including: Based on pelvic floor electromyography data from multiple historical patients, several differential pelvic floor detection locations were identified from the effective pelvic floor detection locations. Based on multiple differentiated pelvic floor detection locations, the initial pelvic floor electromyography data of the current patient is obtained; Based on the current patient's initial pelvic floor electromyography data, determine the cluster to which the current patient belongs; Based on the current patient's cluster and electromyography (EMG) atlas, obtain the current patient's pelvic floor EMG data.

7. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 6, characterized in that, The data acquisition module acquires the pelvic floor electromyography (EMG) data of the current patient based on the current patient's cluster and EMG atlas, including: Based on electromyography (EMG) atlases, key pelvic floor detection locations corresponding to each cluster were identified from the valid pelvic floor detection locations. Based on the key pelvic floor detection locations corresponding to the current patient's cluster, obtain the current patient's pelvic floor electromyography (EMG) data.

8. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 7, characterized in that, The data acquisition module acquires the current patient's pelvic floor ultrasound data based on the current patient's pelvic floor electromyography (EMG) data and EMG-factor correlation atlas, including: Based on the current patient's pelvic floor electromyography (EMG) data and EMG-factor correlation map, the key pelvic floor function assessment factors for the current patient were identified. Based on the key pelvic floor function assessment factors of the current patient, obtain the current patient's pelvic floor ultrasound examination data.

9. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 8, characterized in that, The treatment plan determination module determines the current patient's physical therapy plan based on the patient's current pelvic floor electromyography and pelvic floor ultrasound data, including: Based on the current patient’s pelvic floor electromyography and pelvic floor ultrasound data, identify similar historical patients; Based on similar patient histories, determine the current patient's physical therapy plan.

10. The physiotherapy system for pelvic floor function rehabilitation in cervical cancer patients according to claim 9, characterized in that, The scheme determination module identifies similar historical patients based on the current patient's pelvic floor electromyography and pelvic floor ultrasound data, including: Based on the current patient's pelvic floor electromyography and pelvic floor ultrasound data, as well as the weights of key pelvic floor detection locations corresponding to the current patient's cluster and the weights of key pelvic floor function assessment factors for the current patient, similar historical patients are identified.

Images