Comprehensive lymphedema detumescence self-monitoring management method and related equipment

Abstract

The invention discloses a comprehensive lymphedema detumescence self-monitoring management method and related equipment. The method comprises the following steps: collecting an edema treatment real-time operation video of a target patient; inputting the edema treatment real-time operation video into an edema treatment operation evaluation model so as to perform real-time analysis on the edema treatment operation of the target patient; and under the condition of determining that the edema treatment operation of the target patient has an error, generating an error operation prompt. The problems that household patients and family members are unfamiliar with the process and operation of comprehensive detumescence treatment of the lymphedema, operation errors are prone to occurring, the severity of the lymphedema cannot be evaluated timely and accurately, a treatment scheme is difficult to adjust according to changes of the illness state, and consequently the illness states of some patients are aggravated due to delayed treatment can be solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of smart medical care. More specifically, the present invention relates to a self-monitoring and management method for comprehensive detumescence of lymphedema and related equipment. Background Art

[0002] Lymphedema is a disease caused by impaired lymphatic circulation, resulting in the retention of lymph fluid in interstitial spaces. This leads to a series of pathological changes, including tissue edema, chronic inflammation, and tissue fibrosis, severely impacting human health and quality of life. Lymphedema is categorized into primary and secondary lymphedema. Secondary lymphedema is a chronic complication that is difficult to reverse and requires lifelong self-care. Treatment for lymphedema consists of two phases: in-hospital treatment and home maintenance. Therefore, standardized home self-care for patients with lymphedema is crucial for preventing and controlling lymphedema. However, multiple studies have shown that home self-care is ineffective. Key challenges include: 1. Lack of familiarity between patients and their families with the procedures and procedures for comprehensive lymphedema debulking treatment, leading to procedural errors; and overall poor compliance. 2. A lack of professional, real-time guidance and assessment makes it difficult to identify and correct operational issues. 3. It is impossible to timely and accurately assess the severity of lymphedema, and it is difficult to adjust the treatment plan according to changes in the condition, resulting in some patients' conditions worsening due to delayed treatment. Summary of the Invention

[0003] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention is not intended to limit the key features and essential features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] To address the problem that home patients and their families are unfamiliar with the process and operation of comprehensive lymphedema detumescence treatment, are prone to operational errors, are unable to timely and accurately assess the severity of lymphedema, and have difficulty adjusting treatment plans according to changes in the condition, resulting in some patients' conditions worsening due to delayed treatment, the present invention first proposes a comprehensive lymphedema detumescence self-monitoring and management method, which includes:

[0005] Collect real-time video of edema treatment for target patients;

[0006] Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient;

[0007] When it is determined that an error exists in the edema treatment operation for the target patient, an error operation prompt is generated.

[0008] Optionally, also include:

[0009] Collect real-life videos of clinical patients during different edema treatment steps, including cases of correct operation and various cases of incorrect operation;

[0010] Extracting features from the video to obtain a labeled data set, including at least one feature of action posture, limb angle, and time length;

[0011] The edema treatment operation evaluation model is trained based on the labeled data set.

[0012] Optionally, inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation on the target patient includes:

[0013] Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model;

[0014] The dynamic time warping algorithm is used to calculate the similarity between the target patient's operation sequence and the standard operation sequence. If the similarity is lower than the set threshold, it is judged as an operation error.

[0015] Optionally, also include:

[0016] obtaining pre-bandaging stretched state image information of the bandage during pressure therapy;

[0017] determining a current bandaging pressure of the bandage based on the pre-bandaging stretched state image information;

[0018] Generate tension adjustment operation prompts based on the difference between the current bandage pressure and the ideal bandage pressure;

[0019] The image information of the stretched state of the bandage before bandaging is continuously analyzed until the corresponding bandaging pressure reaches the ideal bandaging pressure, and a bandaging instruction message is generated.

[0020] Optionally, also include:

[0021] Acquiring image information of a stretched state of a bandage covering the surface of an affected limb after bandaging during pressure therapy;

[0022] determining a current bandaging pressure of the bandage based on the image information of the stretched state after bandaging;

[0023] Based on the difference between the current bandaging pressure and the ideal bandaging pressure, if the difference value is greater than a preset difference value, a re-bandaging prompt is generated.

[0024] Optionally, before the step of collecting the real-time operation video of edema treatment of the target patient, the step further includes:

[0025] Acquiring image information of the affected limb of the target patient;

[0026] determining the degree of edema of the affected limb based on the image information of the affected limb;

[0027] The treatment plan is determined based on the degree of edema.

[0028] Optionally, before the step of collecting the real-time operation video of edema treatment of the target patient, the step further includes:

[0029] Obtaining a circumference difference between an affected limb and a contralateral limb of the target patient;

[0030] determining the degree of edema of the affected limb based on the image information of the affected limb and the circumference difference;

[0031] The treatment plan is determined based on the degree of edema.

[0032] In a second aspect, the present invention further provides a lymphedema comprehensive swelling reduction self-monitoring and management device, comprising:

[0033] An acquisition unit, used to acquire real-time operation videos of edema treatment of target patients;

[0034] an evaluation unit, configured to input the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation on the target patient;

[0035] The prompt unit is used to generate an error operation prompt when it is determined that there is an error in the edema treatment operation of the target patient.

[0036] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for comprehensive lymphedema reduction and self-monitoring management as described in any one of the first aspects above when executing the computer program stored in the memory.

[0037] In a fourth aspect, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the comprehensive lymphedema self-monitoring and management method according to any one of the above items in the first aspect.

[0038] In summary, the comprehensive lymphedema self-monitoring management method proposed in this application collects real-time operation videos of edema treatment of target patients; inputs the real-time operation videos of edema treatment into the edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient; and generates an error operation prompt when it is determined that there is an error in the edema treatment operation of the target patient. Home patients often feel at a loss as to where to start due to lack of guidance. Video feedback allows patients to clearly understand whether they are doing it right, thereby enhancing their confidence and initiative. Video recognition can accurately locate the wrong points in the action (such as winding angle, operation rhythm) and correct them through a timely feedback mechanism, significantly reducing the risk of worsening of the condition due to operational errors. Combining model analysis data with long-term operation records, the system can gradually form a patient-specific treatment database and make differentiated adjustment suggestions based on changes in the condition (such as moderately increasing the massage time or switching to high-pressure treatment). All collection and analysis results can be uploaded to the doctor-side platform, and the doctor can remotely review the visual operation records and provide personalized assessment and adjustment suggestions.

[0039] The comprehensive lymphedema self-monitoring and management method for reducing swelling of the present invention, and other advantages, objectives and features of the present invention will be reflected in part through the following description, and will also be understood by those skilled in the art through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0041] Figure 1 A schematic flow chart of a comprehensive lymphedema self-monitoring and management method for reducing swelling provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of the structure of a comprehensive lymphedema self-monitoring and management device provided in an embodiment of the present application;

[0043] Figure 3 This is a schematic diagram of the structure of an electronic device for comprehensive lymphedema reduction and self-monitoring management provided in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0045] To address the issue of home patients and their families being unfamiliar with the procedures and procedures for comprehensive lymphedema detumescence treatment, which can lead to operational errors, inability to timely and accurately assess the severity of lymphedema, and difficulty adjusting treatment plans based on changes in the condition, which can worsen the condition of some patients due to delayed treatment, please refer to Figure 1 , is a flow chart of a comprehensive lymphedema reduction self-monitoring management method provided in an embodiment of the present application, which may specifically include: steps S110 to S130.

[0046] S110, collecting real-time operation video of edema treatment of the target patient.

[0047] S120: Input the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis on the edema treatment operation of the target patient.

[0048] S130: When it is determined that an error exists in the edema treatment operation for the target patient, an error operation prompt is generated.

[0049] Understandably, in traditional home care, patients often experience issues such as incorrect sequencing, improper operation methods, and insufficient duration due to a lack of feedback, which can affect treatment effectiveness and even worsen edema. Through video recognition and model-assisted approaches, the system can effectively address these blind spots in home care, empowering patients to manage themselves, improving treatment compliance, and reducing the risk of complications.

[0050] For example, real-time operation videos of edema treatment of target patients can be collected. When patients perform daily treatments at home (such as bandaging, massage, using pressure pumps, etc.), the operation process is continuously recorded through a fixed-position high-definition camera (such as a smartphone, desktop stand camera or portable wearable device). Video acquisition should cover the main operation parts (such as upper limbs, lower limbs or torso), and ensure clear shooting at appropriate angles and lighting conditions. The system automatically and synchronously records metadata such as acquisition time, shooting angle and patient ID, and provides semantic tags for subsequent identification and analysis. Through the video acquisition system, an objective data recording channel for patient behavior is established to replace traditional subjective text descriptions, thereby enhancing the comprehensiveness and accuracy of behavior capture. At the same time, it provides a stable data source for model recognition, provides a "visual basis" for subsequent feedback, and enhances patients' sense of responsibility and standard implementation rate of operations.

[0051] For example, edema treatment procedure videos can be fed into an edema treatment procedure assessment model for analysis. The system then inputs the captured video data into a trained multimodal edema treatment procedure assessment model in a time series format. Machine learning algorithms that can be used include convolutional neural networks (CNNs) for video image feature extraction and recognition, and decision tree algorithms for problem diagnosis and treatment decision making. This model typically also employs 3D convolutional neural networks (such as I3D and SlowFast) or temporal attention mechanisms (such as Transformers) for action recognition, hand path extraction, tool identification, and sequence discrimination. The model compares the patient's current procedure with a built-in standard procedure library, including: procedure step sequence, operating posture (such as massage direction, angle, and pressure trajectory), tool usage (such as the number of elastic bandage wraps, air pump setting), duration and rhythm, etc. The model automatically detects every potential error detail during the procedure, including sequence reversal (such as incorrect post-massage bandage application), improper method (such as bandage wrapping at an excessively large or loose angle), and time discrepancies (such as pressure pump application lasting less than 10 minutes), and outputs the analysis in a structured format. The technical effect of this link is to transform the operational behavior of "unprofessional behavior" into standardized structural information, break through the problems that cannot be evaluated in the traditional family therapy process, and realize the dual technical closed loop of "real-time evaluation + continuous supervision".

[0052] For example, when an error is detected, a prompt indicating an incorrect operation can be generated. When the model analysis results are compared with the standard procedure and a significant deviation is detected, the system will automatically generate customized feedback prompts on smart devices (such as mobile apps or desktop pop-ups). The prompts can be broken down into: text or voice instructions (e.g., "Current massage direction is inconsistent. Please adjust to axillary drainage."); replay comparison (the system displays an animation comparing the patient's operation with the standard procedure); and additional suggestions for detailed movements (e.g., suggesting that the wrap tension be controlled to "70%-80% of the original length"). Prompts can be graded based on the severity of the error (mild prompt, moderate warning, severe block), providing more structured guidance. Consequently, real-time prompts can significantly reduce the risk of treatment failure or worsening of the patient's condition due to unconscious operational errors. Especially for elderly patients or family members without a nursing background, the real-time prompt system significantly compensates for inexperience, achieving the auxiliary effect of a "family therapy smart coach."

[0053] It is understandable that after accumulating patients' daily operation data over a long period of time, the system can draw operation behavior trend charts, evaluate the stability of daily treatment quality, and support the dataization of treatment decisions. For example, if a patient scores low in the winding tightness dimension for several consecutive days, the doctor may be advised to intervene in advance or change the treatment method. All video analysis results and error records can be uploaded to the cloud platform and reviewed by professional doctors on the remote visual interface, including error distribution charts, error type statistics, correction rate analysis, etc., to assist doctors in achieving refined remote guidance and plan adjustments. Compared with traditional static guidance (such as paper manuals or video tutorials), this system provides interactive, feedback-based, adaptive and individualized guidance. Through positive incentives (such as daily rewards for correct operation rates) or image scoring feedback, the system can stimulate patients' enthusiasm for daily participation and improve treatment compliance and treatment effectiveness.

[0054] For example, standard skin care procedures can be preset to observe the patient's skin condition, such as whether there is damage, redness, or rashes, and guide the patient in choosing skin care products and the frequency of use for different skin types and conditions. The system uses a camera to check whether the patient's skin care procedures meet the standards and prompts any omissions or incorrect operations. Standard procedures for pressure therapy can also be preset, including bandaging methods, pressure ranges, etc. The camera and pressure sensor are used to assess whether the patient's bandaging operation is correct and whether the pressure is within a safe range. Standard movements and frequencies for functional exercises can also be preset. The system uses the camera to analyze whether the patient's exercise movements are standard and the frequency is appropriate, providing real-time feedback.

[0055] In summary, the comprehensive lymphedema self-monitoring management method provided in the embodiment of the present application collects real-time operation videos of edema treatment of target patients; inputs the real-time operation videos of edema treatment into the edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient; and generates an error operation prompt when it is determined that there is an error in the edema treatment operation of the target patient. Home patients often feel at a loss as to where to start due to lack of guidance. Video feedback allows patients to clearly understand whether they are doing it right, thereby enhancing their confidence and initiative. Video recognition can accurately locate the wrong points in the action (such as winding angle, operation rhythm) and correct them through a timely feedback mechanism, significantly reducing the risk of worsening of the condition due to operational errors. Combining model analysis data with long-term operation records, the system can gradually form a patient-specific treatment database and make differentiated adjustment suggestions based on changes in the condition (such as moderately increasing the massage time or switching to high-pressure treatment). All collection and analysis results can be uploaded to the doctor-side platform, and the doctor can remotely review the visual operation records and provide personalized assessment and adjustment suggestions.

[0056] According to some embodiments, further comprising:

[0057] Collect real-life videos of clinical patients during different edema treatment steps, including cases of correct operation and various cases of incorrect operation;

[0058] Extracting features from the video to obtain a labeled data set, including at least one feature of action posture, limb angle, and time length;

[0059] The edema treatment operation evaluation model is trained based on the labeled data set.

[0060] For example, real-life videos of clinical patients undergoing different edema treatment steps can be collected. The system systematically collects videos of real patients undergoing edema treatment in a hospital setting. These videos must cover all treatment steps (such as elastic bandage wrapping, lymphatic massage, pressure pump use, limb elevation, etc.) and be divided into correct and incorrect operation examples based on the degree of procedural standardization. Incorrect operations can include a variety of scenarios, such as incorrect operation sequence, reversed posture direction, incomplete operation, and insufficient duration. These scenarios can be categorized and annotated by a team of clinical experts. Furthermore, typical errors can be simulated appropriately (for example, by having a nurse play the role of a patient and demonstrate errors) to supplement rare error samples. This diverse video sample establishes a comprehensive library of operation scenarios, providing the model with stronger generalization and error detection capabilities. The coverage of real data significantly improves training effectiveness, especially when dealing with ambiguous scenes, angle changes, or actions with blurred boundaries (such as the detailed wrist rotation during bandage wrapping). This step gives the system powerful error recognition capabilities, allowing it to not only identify common operations, but also capture small errors that are difficult to detect clinically but affect the treatment effect, such as micro-movement errors such as inconsistent bandage tightness or deviation in massage direction.

[0061] For example, feature extraction can be performed on videos to obtain an annotated dataset. After video capture, the system uses pose estimation technology (such as OpenPose or MediaPipe) to extract the motion trajectories of key parts (such as shoulders, elbows, wrists, knees, ankles, etc.) in each frame of video, generating structured features. Features include, but are not limited to: action posture features, such as the angle and speed of the palm relative to the limb when pressing; limb angle features, such as the angle between the calf and the direction of the bandage when wrapping a bandage; and time length features, such as the duration and pause length of each step. Combined with expert interpretation and labeling systems, each video is assigned labels such as standard / non-standard, error type category, and confidence score, forming a high-quality annotated dataset for supervised learning. Through a three-step data structure combining action decomposition with multidimensional feature extraction and expert label fusion, the system converts raw videos into training samples that the model can learn from. During the learning process, the model not only learns visual patterns but also understands the spatiotemporal logic and anatomical meaning of the actions. The resulting labeled dataset contains triple information: behavioral timing, anatomical geometry, and expert semantics. This allows the model to distinguish actions that are superficially similar but have different execution paths (such as forward massage and reverse sliding) during training, thereby achieving high-precision clinical error classification and identification.

[0062] For example, an edema treatment operation evaluation model can be trained based on a labeled dataset. An action recognition neural network architecture (such as I3D, TSN, SlowFast, or Transformer-based temporal networks) can be constructed or selected, using a structured labeled dataset as the training set and executing a supervised training process. The following strategies are introduced during training to optimize model capabilities. Uncertainty in a home environment can be simulated through angular perturbations, brightness changes, and time scaling; correctness classification tasks and error type identification tasks can be trained simultaneously; and structures such as LSTM / Transformer can be used to model the sequential nature of action processes. After training, the model will possess two core capabilities: determining whether a particular operation is standard and accurately identifying the type of error (such as sequence errors, posture deviations, or insufficient timing) in non-standard operations. This training approach, through cross-time action association modeling (temporal understanding) and a multi-label parallel learning mechanism, enables the model to not only identify objects from images but also understand the logic of actions, essentially simulating the process of physical therapists determining the continuity of actions. After training, the model can adapt to posture changes, background interference and non-ideal shooting conditions in real home environments. At the same time, it has high tolerance and error sensitivity to patients' diverse behaviors, significantly improving the subsequent system recognition accuracy and feedback timeliness.

[0063] According to some embodiments, inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation on the target patient includes:

[0064] Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model;

[0065] The dynamic time warping algorithm is used to calculate the similarity between the target patient's operation sequence and the standard operation sequence. If the similarity is lower than the set threshold, it is judged as an operation error.

[0066] It can be understood that dynamic time warping (DTW) is an algorithm that calculates the nonlinear similarity between two time series signals (such as motion trajectory sequences), allowing alignment comparison under speed changes. Its core idea is to compare each key point in the target sequence with multiple possible positions in the standard sequence, and dynamically find the global optimal alignment path by constructing a cumulative cost matrix to calculate the total matching cost. This method is widely used in scenarios such as speech recognition, handwriting recognition, and time series medical data analysis. In lymphedema treatment operations, different patients have different natural rhythms (such as some patients have a slower massage rhythm and uneven bandaging time). If only a fixed time window is used for comparison, it is easy to misjudge it as an incorrect operation. The introduction of DTW enables the model to have recognition capabilities that are insensitive to duration and sensitive to motion trajectory, and can accurately evaluate the actual action quality.

[0067] For example, when a patient performs an edema treatment procedure, a live video of the procedure is captured by a smart terminal and transmitted to the system in real time. The system first pre-processes the video using a motion recognition model, extracting key temporal feature points (such as hand position, joint angle, motion path, and operation rhythm). The sequence of feature points in these consecutive frames is then constructed into an operation sequence for the target patient. The system then compares this operation sequence with a standard operation sequence stored in a model database. To account for speed differences, pause time differences, or varying durations of actions between the patient's operation rhythm and the standard example, the system uses the dynamic time warping (DTW) algorithm to perform nonlinear alignment and matching of the two action sequences. This algorithm uses dynamic programming to find the "best match of the shortest path" between the two sequences, enabling frame-level action matching even with different action timelines. Based on this, the system calculates a similarity score (or "matching cost") between the two sequences. If the similarity falls below a preset threshold (e.g., 90%), it is determined to be an operation error, triggering the subsequent error notification module. With the introduction of DTW, the system can handle scenarios with large differences in operation duration. For example, if the standard massage technique is 10 seconds, but the patient actually performs it for 15 seconds, but the path and technique remain consistent, DTW can perform nonlinear alignment to avoid false positives. This feature significantly improves the model's inclusiveness and clinical adaptability. Even in cases where the rhythm is consistent but the path deviates (such as incorrect massage direction or bandage path deviation), DTW can still accurately capture the path jumps or bends caused by the trajectory change, identifying these actions as "low similarity" and accurately triggering corrective prompts. The similarity score, as a quantitative metric, can be used for subsequent system behavior scoring, training plan generation, and performance assessment during remote physician feedback. For example, the system might prompt, "The similarity of this action is 78%. Please note that the massage starting point and direction are consistent." For example, a patient performing a lower limb lymphatic drainage massage at home took 12 seconds, while the standard technique lasted 8 seconds. Despite the difference in time, the technique path, rhythm, and direction were largely consistent with the standard technique. The system used DTW to dynamically match the two action sequences and calculated a similarity of 94%, which exceeded the threshold (such as 90%) and determined that the operation was correct. However, the other patient reversed the direction during the execution, and the DTW calculated similarity was only 62%. The system identified it as a path error and generated a prompt: "Please perform drainage massage from the distal end to the proximal end."

[0068] In some examples, this also includes:

[0069] obtaining pre-bandaging stretched state image information of the bandage during pressure therapy;

[0070] determining a current bandaging pressure of the bandage based on the pre-bandaging stretched state image information;

[0071] Generate tension adjustment operation prompts based on the difference between the current bandage pressure and the ideal bandage pressure;

[0072] The image information of the stretched state of the bandage before bandaging is continuously analyzed until the corresponding bandaging pressure reaches the ideal bandaging pressure, and a bandaging instruction message is generated.

[0073] It is understandable that the core of this process is based on the physical model of elastic materials and computer vision feature extraction. When the elastic bandage is in a stretched state, the spacing, angle, and deformation rate of its surface texture or dedicated marks (such as black and white spaced dot matrix, colored ruler lines) will change regularly. These changes can be automatically captured by image processing models (such as texture detection based on OpenCV, pattern recognition based on CNN), and the stretching ratio λ = L / L0 (current length / initial length) can be calculated. Combined with a pre-built tension-pressure function model (for example, simplified according to the Laplace formula or experimentally fitted), the stretching ratio is converted into pressure per unit area, thereby realizing the function of inferring physical pressure from the image. The advantage of this mechanism is that it does not rely on expensive electronic pressure sensors, and reliable inference can be achieved only through images and known parameters.

[0074] For example, during the treatment of lymphedema, the system obtains image information of the patient's bandage stretching state before the elastic bandage is wrapped through a smart terminal (such as a mobile phone camera or a home auxiliary wearable device). The image can capture the degree of stretching by shooting the surface texture of the first circle of bandage on the limb or built-in reference marks (such as a specific color marking line, a spacing ruler). After the image acquisition is completed, the system calls the visual analysis model to extract the unit length stretch ratio information of the bandage, and calculates the current bandage pressure in combination with the known physical elastic parameters of the bandage (such as the elongation coefficient, the initial tension-pressure mapping table). Subsequently, the system compares the current pressure with the ideal bandage pressure range set by medicine. If the pressure value is lower than the target lower limit, an operation prompt of "Insufficient tension, please stretch appropriately" is generated; if it exceeds the target upper limit, a prompt of "Excessive tension, easy to cause ischemia, please relax appropriately" is generated. Before each bandage is applied to the patient, the system continuously collects image information and performs real-time analysis until the bandaging pressure of the current bandage reaches the ideal range. The system then generates a bandaging instruction message stating that the next bandage can be started, thus forming a closed-loop feedback mechanism from bandaging action to stretching analysis, pressure verification, and feedback prompts.

[0075] Understandably, traditional home bandaging relies heavily on experience and feel, making it prone to being too loose and ineffective, or too tight and causing injury. With the introduction of image-based stretch analysis, the system provides quantitative feedback for each bandage application, ensuring pressure remains within the medically recommended range (e.g., 20-40 mmHg), significantly improving the effectiveness and safety of bandaging. The system continuously provides voice or graphic prompts indicating "current tension is too low / appropriate / too high," enabling patients to develop a correct force perception during application, shifting from passive reliance to active adjustment. This helps patients quickly establish a correct skill path and mitigate localized edema rebound caused by repeated incorrect bandaging. This system replaces traditional manual acupressure assessment or specialized equipment monitoring with visual technology, enabling patients to complete standardized bandaging procedures even without a caregiver, automating home treatment and significantly improving the efficiency of chronic disease management. The stretch value and corresponding pressure of each bandage application are recorded and uploaded, creating a comprehensive bandaging process graph. This allows physicians to remotely assess patient performance, identify trends of chronically low tension or misapplication, and initiate timely intervention.

[0076] For example, consider a patient using an elastic bandage with two strips of ribbon to wrap their lower limbs. The system captures an image of the current wrapping area and identifies that the ribbon spacing has been stretched from the original 5mm to 9mm, corresponding to a stretch ratio of λ = 1.8. Based on the tension curve for known bandage types, the system calculates the current pressure to be 28mmHg, which is within the recommended range. The system prompts, "The current tension is moderate. Please proceed to the next wrap." However, during the next capture, the ribbon spacing is 3mm (λ ≈ 0.6). The system detects insufficient tension and prompts, "The current tension is too low. Please tighten appropriately before wrapping."

[0077] In some examples, this also includes:

[0078] Acquiring image information of a stretched state of a bandage covering the surface of an affected limb after bandaging during pressure therapy;

[0079] determining a current bandaging pressure of the bandage based on the image information of the stretched state after bandaging;

[0080] Based on the difference between the current bandaging pressure and the ideal bandaging pressure, if the difference value is greater than a preset difference value, a re-bandaging prompt is generated.

[0081] It is understandable that even if the stretching is well controlled during the bandaging process, if the final fixation is improper or sliding and loosening occurs, the actual bandaging pressure will not reach the ideal value. Then it can still be based on the functional relationship between the optical deformation of the elastic fabric and the physical tension. When the bandage is in a stretched state, its surface markings (such as double-line scales, printed ribbons, and variable reflective layers) will deform with the degree of stretching, and these deformations can still be captured by images after bandaging. The system compares the measured features in the image with the marking features in the known unstretched state to deduce the local stretch rate λ=L / L0 at each point, and further substitutes it into the physical model to deduce the local bandaging pressure P(λ) at that point. Since the pressure may be locally uneven or unstable due to loosening, gravity, or uneven winding after the bandage is completed, the introduction of this mechanism can be regarded as an image quality control verification of the expected treatment effect.

[0082] For example, after the patient completes the elastic bandage on the target limb, the system uses an external camera to capture image information of the stretched state of the bandage covering the surface of the affected limb. This image acquisition position should cover multiple typical areas on the limb surface (such as the mid-section of the limb, near the joints, and distal fixed points), and prioritize the acquisition of bandage marking areas (such as ribbons, specific textures, grid deformation, etc.). The system calls the image processing module to extract the stretching features in the captured area, including ribbon spacing, grid deformation amplitude, angular offset, weave density, etc., and calculates the current bandage pressure value based on the tension-pressure mapping model of the bandage material. Subsequently, this pressure value is compared with the ideal bandage pressure range set in the system; if the difference (ΔP) between the actual pressure and the target pressure in any area exceeds a preset threshold (such as ±10 mmHg), the system automatically generates a re-bandaging prompt, prompts the possible consequences of excessive or insufficient pressure at the corresponding location, and recommends re-bandaging.

[0083] It's understandable that the pre-bandaging tension analysis mentioned above ensures the appropriate initial pressure, while the post-bandaging image verification step is the final verification step, preventing the problem of starting correctly but ending incorrectly. For example, if the end fasteners are not securely fastened, and the actual tension in the last few turns of the bandage falls far below the target, the system can promptly identify and provide a warning, preventing uncontrolled or even exacerbated edema due to insufficient local pressure. By comparing the tensile pressure distribution across various regions in the image, the system can identify areas where excessive pressure is present, posing a risk of compression, or areas where pressure is too low, creating a blind spot in treatment, prompting a re-bandaging request. High pressure, in particular, can cause blood flow obstruction in joints. This mechanism helps dynamically adjust pressure distribution, improving the physiological adaptability and tissue tolerance of the bandage. The system can upload the pressure distribution of each region after bandaging to a remote platform as a heat map or quantitative distribution map, allowing doctors to remotely assess the quality of the patient's home treatment. Doctors can also recommend personalized optimization solutions based on repeated abnormal bandaging of a specific area, such as changing the bandage type or using a pressure monitoring device.

[0084] For example, after a patient completes bandaging their lower limbs, the system guides them to take a complete bandaging image (centered on the mid-calf and bounded by the knee and ankle). Image recognition shows that the grid point spacing near the ankle of the calf is 1.15 times the original value. The system calculates the bandaging pressure there to be approximately 15 mmHg; the ideal value set by the system is 30-40 mmHg, so ΔP = -15 mmHg. The system generates a red prompt: "The ankle pressure is clearly insufficient, which may affect the proximal drainage effect. It is recommended to re-bandage and increase the terminal tension."

[0085] In some examples, before the step of acquiring a real-time video of edema treatment operation of a target patient, the method further includes:

[0086] Acquiring image information of the affected limb of the target patient;

[0087] determining the degree of edema of the affected limb based on the image information of the affected limb;

[0088] The treatment plan is determined based on the degree of edema.

[0089] It is understandable that limb edema manifests as abnormal increase in soft tissue volume, increased skin tension, subcutaneous fluid retention, etc., and quantitative indicators can be obtained through image analysis. The system identifies the area of ​​volume expansion by comparing with a normal limb model in a non-edema state; and then identifies skin tension and indentation recovery (such as rebound speed after finger pressure) through color texture changes and surface contour analysis. Some systems also support comparative image recognition (patients upload historical images or healthy side images of the same limb) to achieve more accurate contralateral comparative analysis or temporal progression judgment. The system automatically selects the appropriate bandage pressure range (such as 20-30 mmHg for mild edema and 30-40 mmHg for moderate edema) and operating specifications based on the "parameter matching table" of the degree of swelling and the treatment plan.

[0090] For example, before capturing a video of the patient's treatment procedure, the system first guides the user to capture standardized images of the target limb (e.g., front, side, and top views). Image acquisition can be performed using a smartphone, home tablet, or portable assistive device. The system then uses an image analysis module to process the images, identifying the target limb's edge contours, skin tension distribution, visible swelling areas, and contrast features compared to the contralateral limb. Image processing can employ techniques such as edge detection, depth estimation, 3D reconstruction, or keypoint gesture recognition to extract key visual indicators such as limb volume change, skin tension, and tissue protrusions. The system can integrate previously trained classification models to match the current image to a predefined edema classification standard (e.g., mild, moderate, severe, indurated, or pitting). Once the edema level is identified, the system automatically generates a personalized comprehensive edema treatment plan, including recommended bandaging schemes (e.g., bandage type, number of wraps, recommended pressure range), lymphatic drainage frequency, elevation duration, and pressure pump usage cycle. This serves as a reference baseline for subsequent procedure quality analysis and feedback evaluation. Therefore, subsequent video recognition module decisions regarding bandage pressure, massage rhythm, and tool usage must be based on edema severity. By assessing edema severity through images and generating personalized treatment plans, the system dynamically adjusts the standard judgment criteria based on severity, improving the accuracy of feedback and avoiding one-size-fits-all misjudgments. This system avoids the risk of over- or undertreatment, resulting from the misuse of high-pressure bandages for mild edema or inefficient techniques for severe edema. Precisely setting pressure ranges is crucial for ensuring blood supply and preventing necrosis, particularly in high-risk areas (such as the distal lower limbs and joint depressions). Edema assessment, as a starting point, provides the system with a degree of decision-making support, providing rationale for subsequent feedback (such as prompts regarding insufficient bandage pressure or insufficient massage time). The system saves each image and edema recognition result, creating a time-series image comparison archive. If images over the past week indicate continued worsening of edema, the system can automatically adjust the treatment plan to an enhanced one, increasing the frequency of pressure pump use or recommending referral.

[0091] Taking a frontal image of the patient's left calf as an example, the system identified that the limb's outer contour was approximately 12% larger than the reference image of the healthy side, with enhanced skin surface reflection, blurred texture, and indentation recovery time exceeding 15 seconds. The model matching result was moderate edema. The system then generated a treatment plan: a bandaging pressure of 30-40 mmHg was recommended, a high-extensibility gradient compression elastic bandage was recommended as the bandage type, massage was performed twice a day for 15 minutes each time, and a pressure pump was used once a day at a setting of 40 mmHg for 20 minutes. The system will use this as a standard baseline when analyzing bandaging videos and pressure images.

[0092] In some examples, before the step of acquiring a real-time video of edema treatment operation of a target patient, the method further includes:

[0093] Obtaining a circumference difference between an affected limb and a contralateral limb of the target patient;

[0094] determining the degree of edema of the affected limb based on the image information of the affected limb and the circumference difference;

[0095] The treatment plan is determined based on the degree of edema.

[0096] It is understandable that the circumference difference between the affected limb and the healthy limb is one of the most commonly used and reliable quantitative indicators in the clinical evaluation of edema, and is often used to confirm early edema, track treatment effects, and identify symmetrical diseases. Traditional measurement methods are mostly based on manual measurement with a soft ruler. This system establishes an image-geometry calibration model by combining the limb circumference contour extracted from the image with the reference circumference actually input by the patient or automatically identified, thereby obtaining an accurate circumference difference value. At the same time, image information (such as edge sharpness, surface tension, and concave indentation) provides auxiliary visual dimensional judgment, allowing the system to more accurately perform grading when the numerical boundaries are blurred or there are slight differences. Combining the advantages of both, the system achieves comprehensive judgment of edema and promotes fine stratification of subsequent treatment intensity.

[0097] For example, before capturing a video of the target patient undergoing treatment, the system first guides the patient to take images of their affected limb and contralateral healthy limb in a standard posture (e.g., both calves, calf and thigh segments, both upper limbs, etc.). The system then uses a doctor's order or an intelligent device (e.g., a tape measure, electronic ruler, or volume sensor) to obtain limb circumference data at specific reference locations (e.g., ankle, mid-calf, upper third of thigh). These measurements can be manually entered by the patient or automatically identified through image registration and proportional estimation. The system fuses the image information with the circumference difference (ΔC = affected limb circumference - healthy limb circumference). Combining features such as the edge morphology, tension state, and tissue texture of the affected limb in the image, the system uses a deep assessment model (e.g., a joint classification network of image and numerical features) to determine the degree of edema in the affected limb. Once the degree is established, the system develops an individualized treatment plan based on pre-set criteria, including parameters such as recommended bandage pressure range, bandage type, elevation time, lymphatic drainage plan, and whether to recommend auxiliary pump therapy, providing standard reference values ​​for subsequent operation recognition and feedback judgment. For example, a patient inputs a circumference of the affected limb in the middle of the calf of 33.5cm and a circumference of the contralateral healthy limb of 31.0cm. The system calculates ΔC = 2.5cm, which is within the moderate edema judgment range. Simultaneously, the image captured shows that the contours of the affected limb are unclear and local indentations are obvious. The model combines visual features with numerical judgment and ultimately assesses the degree of edema as being in the moderate-to-severe borderline range. Based on this, the system generates a treatment plan: high-pressure elastic bandages are recommended, with a target bandage pressure of 35-45mmHg; drainage techniques are increased to twice a day; a low-pressure sheath is recommended for nighttime elevation assistance; and the system uses this pressure range as a benchmark for action recognition and feedback evaluation when subsequently analyzing the operation video.

[0098] It is understandable that compared with simple image recognition, the circumference difference, as a reliable indicator of physical size changes, can accurately reflect the degree of swelling of the affected limb, especially with high sensitivity in early mild edema or repeated changes in the chronic stage. The combination of images and numerical values ​​forms a double chain of evidence, which can significantly reduce the risk of missed diagnosis and misjudgment. Based on the corresponding standards of circumference difference, the system calls differentiated treatment strategies in a coordinated manner to ensure that the operation feedback module guides judgment based on quantitative thresholds, thereby achieving dynamic matching of pressure setting, drainage frequency, and bandaging density. The image and circumference data collected each time can be included in the system archive to form a time series record of images and circumferences. The system can automatically compare historical records to identify whether the edema has decreased, remained unchanged, or rebounded, and recommend enhancement plans or relief plans in a timely manner to improve chronic disease monitoring capabilities. Doctors can see patient data at a glance on the background platform, no longer relying solely on patients' subjective statements, significantly improving data visualization and trust in treatment decisions.

[0099] According to some embodiments, during the patient's evaluation phase before edema treatment, the system prompts the user to press the target measurement segments of the left and right arms (such as the middle section of the forearm or the middle section of the upper arm) in a vertical direction on the mobile phone screen in turn, and keep the wrist or elbow naturally extended to ensure that the skin is fully in contact with the screen. The system activates the screen touch point detection function through the built-in App or WeChat applet, and records the maximum width (W) and area (A) of the contact area. The system regards the contact area as an approximate elliptical or flattened circular cross-section, and estimates the approximate value of the original circumference (i.e., the arm circumference C) in combination with the preset arm flattening rate model or calibration database. The user repeats the above operation for the healthy arm to obtain an estimated value of the circumference difference of ΔC = affected limb C1-healthy limb C2. The estimated value is automatically archived by the system and used together with the image recognition results for edema grade identification and treatment plan formulation. The core of this method is to use the contact cross-section of the user's arm pressed on the mobile phone screen as a compressed deformation representation of the actual cross-section of the limb. Taking the approximate elliptical cross section as an example, the contact width W ≈ the minor axis of the ellipse (b), the screen contact area A ≈ π·a·b ≈ π·a·W, the major axis a ≈ A / (π·W), the circumference C ≈ the ellipse circumference estimation formula, which can be approximated as By combining the screen resolution (e.g., 1080px width corresponds to 6.5cm) and the unit area of ​​the screen touch points, pixel measurements can be converted into actual centimeters. The system can also automatically introduce a compression correction factor K (generally between 1.05 and 1.20) based on user parameters such as gender, age, and body shape to offset errors caused by incomplete skin contact or different degrees of compression. As a result, users can complete basic circumference measurements with the help of their daily mobile phones without the need for external tools such as tape measures, providing physical calibration data for the image recognition system. Compared to manual measurement, which is greatly affected by position, operating technique, and tape tightness, this method uses a fixed operating posture and standard process (e.g., aligning the screen edge and pressing naturally for 5 seconds), improving repeatability and user compliance. By fusing the circumference difference obtained from the contact area estimate with visual image features (e.g., skin tension and tissue morphology), a more robust assessment result can be given in the fuzzy range of mild to moderate edema, reducing the risk of over- or under-treatment. Easy to integrate into remote assessment system: Since the information required by this method can be directly uploaded to the system backend, doctors can remotely review the measurement process and estimation curve, forming a reliable remote edema status initial screening program. For example, the user presses the forearm on the screen, and the App records the contact width as 5.8cm and the area as approximately 48cm. 2 The calculated short axis b ≈ 2.9 cm, the long axis a ≈ 2.63 cm, and the estimated original circumference C ≈ 18.2 cm. The contralateral forearm was estimated to be 16.5 cm using the same method, with a final ΔC ≈ 1.7 cm. Image recognition revealed mild increased surface tension in the affected limb, and the system assessed mild to moderate edema in the arm. Low-pressure bandaging (20-30 mmHg) and two daily elevations were recommended as the initial treatment plan.

[0100] See also Figure 2 An embodiment of the lymphedema comprehensive swelling reduction self-monitoring and management device in the embodiment of the present application may include:

[0101] The acquisition unit 21 is used to acquire a real-time operation video of edema treatment of a target patient;

[0102] An evaluation unit 22 is configured to input the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis on the edema treatment operation of the target patient;

[0103] The prompting unit 23 is configured to generate an error operation prompt when it is determined that an error exists in the edema treatment operation for the target patient.

[0104] In summary, the lymphedema comprehensive detumescence self-monitoring management device provided in the embodiment of the present application collects the real-time operation video of the edema treatment of the target patient; inputs the real-time operation video of the edema treatment into the edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient; and generates an error operation prompt when it is determined that there is an error in the edema treatment operation of the target patient. Home patients often feel at a loss as to where to start due to lack of guidance. Video feedback allows patients to clearly understand whether they are doing it right, thereby enhancing their confidence and initiative. Video recognition can accurately locate the wrong points in the action (such as winding angle, operation rhythm) and correct them through a timely feedback mechanism, significantly reducing the risk of worsening of the condition due to operational errors. Combining model analysis data with long-term operation records, the system can gradually form a patient-specific treatment database and make differentiated adjustment suggestions based on changes in the condition (such as moderately increasing the massage time or switching to high-pressure treatment). All collection and analysis results can be uploaded to the doctor-side platform, and the doctor can remotely review based on the visual operation records to provide personalized assessment and adjustment suggestions.

[0105] like Figure 3 As shown, an embodiment of the present application further provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 320 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for comprehensive lymphedema reduction and self-monitoring management are implemented:

[0106] Collect real-time video of edema treatment for target patients;

[0107] Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient;

[0108] When it is determined that an error exists in the edema treatment operation for the target patient, an error operation prompt is generated.

[0109] Since the electronic device introduced in this embodiment is a device used to implement a lymphedema comprehensive swelling reduction self-monitoring and management device in the embodiment of this application, based on the method introduced in the embodiment of this application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of this application falls within the scope of protection of this application.

[0110] In the specific implementation process, the computer program 311 can be implemented when executed by the processor Figure 1 Any implementation manner in the corresponding embodiment:

[0111] Collect real-time video of edema treatment for target patients;

[0112] Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient;

[0113] When it is determined that an error exists in the edema treatment operation for the target patient, an error operation prompt is generated.

[0114] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0115] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0116] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0117] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0118] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0119] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device is caused to execute the following Figure 1 The process of comprehensive lymphedema reduction self-monitoring management in the corresponding embodiment.

[0120] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0121] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0122] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0123] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0124] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0125] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0126] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A self-monitoring and management method for comprehensive detumescence of lymphedema, characterized in that: include: Collect real-time video of edema treatment for target patients; Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient; When it is determined that an error exists in the edema treatment operation for the target patient, an error operation prompt is generated.

2. The method according to claim 1, wherein Also includes: Collect real-life videos of clinical patients during different edema treatment steps, including cases of correct operation and various cases of incorrect operation; Extracting features from the video to obtain a labeled data set, including at least one feature of action posture, limb angle, and time length; The edema treatment operation evaluation model is trained based on the labeled data set.

3. The method according to claim 1, wherein Inputting the real-time edema treatment operation video into the edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation of the target patient includes: Inputting the real-time edema treatment operation video into an edema treatment operation evaluation model; The dynamic time warping algorithm is used to calculate the similarity between the target patient's operation sequence and the standard operation sequence. If the similarity is lower than the set threshold, it is judged as an operation error.

4. The method according to claim 3, wherein Also includes: obtaining pre-bandaging stretched state image information of the bandage during pressure therapy; determining a current bandaging pressure of the bandage based on the pre-bandaging stretched state image information; Generate tension adjustment operation prompts based on the difference between the current bandage pressure and the ideal bandage pressure; The image information of the stretched state of the bandage before bandaging is continuously analyzed until the corresponding bandaging pressure reaches the ideal bandaging pressure, and a bandaging instruction message is generated.

5. The method according to claim 4, wherein Also includes: Acquiring image information of a stretched state of a bandage covering the surface of an affected limb after bandaging during pressure therapy; determining a current bandaging pressure of the bandage based on the image information of the stretched state after bandaging; Based on the difference between the current bandaging pressure and the ideal bandaging pressure, if the difference value is greater than a preset difference value, a re-bandaging prompt is generated.

6. The method according to any one of claims 1 to 5, characterized in that Before the step of collecting the real-time operation video of edema treatment of the target patient, it also includes: Acquiring image information of the affected limb of the target patient; determining the degree of edema of the affected limb based on the image information of the affected limb; The treatment plan is determined based on the degree of edema.

7. The method according to claim 6, wherein Before the step of collecting the real-time operation video of edema treatment of the target patient, it also includes: Obtaining a circumference difference between an affected limb and a contralateral limb of the target patient; determining the degree of edema of the affected limb based on the image information of the affected limb and the circumference difference; The treatment plan is determined based on the degree of edema.

8. A lymphedema comprehensive swelling reduction self-monitoring and management device, characterized by: include: An acquisition unit, used to acquire real-time operation videos of edema treatment of target patients; an evaluation unit, configured to input the real-time edema treatment operation video into an edema treatment operation evaluation model to perform real-time analysis of the edema treatment operation on the target patient; The prompt unit is used to generate an error operation prompt when it is determined that there is an error in the edema treatment operation of the target patient.

9. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the lymphedema comprehensive swelling reduction self-monitoring management method according to any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the comprehensive lymphedema reduction self-monitoring and management method according to any one of claims 1 to 7 is implemented.

Images