A postpartum phased functional rehabilitation training method and system

Abstract

The present application relates to the technical field of postpartum rehabilitation, and in particular to a postpartum phased functional rehabilitation training method and system, which comprises collecting core muscle group activity, breathing pattern and body posture data through wearable sensors, combining subjective feedback to generate an individualized rehabilitation baseline; dividing the recovery phase based on the baseline and calling an action library to generate a training plan; applying an action adaptation algorithm to optimize training parameters and dynamically adjusting the scheme through real-time monitoring. The present application ensures the safety, adaptability and scientificity of rehabilitation training through multi-modal data fusion and dynamic time warping algorithm, effectively solves the problem of lack of dynamic evaluation and individualized adjustment in traditional schemes, significantly improves rehabilitation efficiency and user experience, and has high practical value.

Description

Technical Field

[0001] This invention belongs to the field of medical rehabilitation technology, specifically a postpartum phased functional rehabilitation training method and system. Background Technology

[0002] Postpartum rehabilitation is a crucial step for women to restore physical function and prevent long-term health problems after childbirth, especially interventions for issues such as diastasis recti, pelvic floor dysfunction, and weakened core strength. Currently, existing technological solutions related to postpartum exercise rehabilitation mainly include general fitness courses / videos, physical therapy or physical rehabilitation, basic postpartum rehabilitation equipment / tools, and basic health tracking apps / software. While these solutions meet the needs of some users to a certain extent, they still have many shortcomings.

[0003] Regular fitness classes / videos are the most common form, usually designed and taught by fitness instructors, yoga teachers, or online influencers. The content may include exercises targeting the abdominal and pelvic floor muscles. However, these classes are mostly designed based on general fitness principles, lacking professional medical support, ignoring individual differences (such as the degree of diastasis recti, type of pelvic floor dysfunction, delivery method, etc.), and failing to fully consider the physiological characteristics and contraindications of postpartum women. This can lead to practitioners performing inappropriate or even harmful movements (such as premature high-intensity crunches or jumping), thereby worsening diastasis recti, exacerbating pelvic floor muscle damage, and even causing pain or other health risks.

[0004] Physical therapy or physical rehabilitation is provided by registered physical therapists who prescribe individualized exercise prescriptions based on assessments of pelvic floor muscle function, diastasis recti, core strength, and other issues. While this method is highly professional and safe, its assessments are typically sporadic and discontinuous (e.g., follow-up appointments every few weeks), making it difficult to dynamically track a postpartum woman's recovery and athletic performance in real time. Furthermore, physical therapy relies on frequent in-person appointments, which are costly and inconvenient, posing a significant obstacle, especially for postpartum women with limited time or those living in remote areas. Regular fitness classes and health tracking apps completely lack dynamic adjustment mechanisms and cannot optimize rehabilitation plans based on the user's recovery progress, potentially leading to inefficient rehabilitation or an increased risk of re-injury.

[0005] While basic postpartum rehabilitation equipment / tools (such as pelvic floor muscle trainers, yoga balls, and resistance bands) and health tracking apps can serve as supplementary tools, the former has limited functionality, and the latter has weak data integration capabilities. Existing health tracking apps mostly focus on general health indicators (such as weight and steps), lacking the ability to collect, analyze, and visualize key indicators specifically for postpartum rehabilitation (such as pelvic floor muscle awareness, diastasis recti width, and the quality of specific movement completion), making it difficult to form a holistic understanding of the rehabilitation process.

[0006] Furthermore, existing solutions generally suffer from a lack of user experience and low adherence. Online courses and apps are often dull and monotonous, lacking interactivity, engagement, and emotional support, failing to adequately consider the unique challenges of postpartum women's psychological state (such as fatigue, anxiety, and loneliness) and behavioral changes. Simultaneously, many solutions overemphasize isolated muscle training (e.g., focusing solely on Kegel exercises or crunches), neglecting the holistic reconstruction of the "core" as a dynamic functional unit (coordination of the diaphragm, pelvic floor muscles, transverse abdominis, and multifidus muscles), and lacking systematic integration of key elements such as breathing pattern reconstruction, posture correction, and psychological stress management. This results in incomplete rehabilitation outcomes, potentially addressing localized problems (such as urinary incontinence) while leaving behind overall functional deficiencies (such as core weakness, postural compensation, and chronic pain) or failing to effectively alleviate postpartum anxiety and depression.

[0007] In summary, existing technical solutions have significant shortcomings in terms of targeting, safety, dynamic assessment capabilities, accessibility, user experience, mind-body integration, and data-driven approaches. Therefore, there is an urgent need for a postpartum rehabilitation solution that can accurately match the physiological patterns of postpartum recovery, implement dynamic, progressive, and personalized adjustments, incorporate medical expertise, promote holistic physical and mental recovery, and improve user compliance. This invention aims to address these issues through a phased functional rehabilitation training method and system, providing postpartum women with safe, scientific, efficient, and easy-to-implement rehabilitation guidance. Summary of the Invention

[0008] This invention relates to the field of postpartum rehabilitation technology, specifically to a postpartum phased functional rehabilitation training method and system. The technical problem this invention addresses is that existing postpartum rehabilitation training programs lack targeted, dynamic assessment, and personalized adjustment mechanisms, making it difficult to meet the physiological needs of postpartum women at different recovery stages. Furthermore, traditional programs neglect mind-body integration and functional reconstruction, resulting in a limited user experience, low compliance, and consequently, limited rehabilitation effectiveness.

[0009] This invention provides a postpartum phased functional rehabilitation training method, comprising: collecting core muscle group activity data, breathing pattern data, and body posture data of the postpartum woman through wearable sensors, and combining subjective feedback information input from a mobile device to generate an individualized rehabilitation baseline; dividing the postpartum recovery stage based on the rehabilitation baseline, and calling the corresponding movement library to generate a training plan; optimizing the training plan by applying a movement adaptation algorithm to ensure the safety and suitability of the movements; and collecting movement performance data during the training process through a real-time monitoring module, dynamically adjusting training parameters, and generating a rehabilitation plan for the next stage.

[0010] The generation of an individualized rehabilitation baseline includes a first preprocessing step on core muscle group activity data collected by wearable sensors, which includes signal filtering and outlier removal; a second preprocessing step on respiratory pattern data, which includes respiratory rate calibration and depth normalization; using a multimodal data fusion algorithm to align and analyze core muscle group activity data, respiratory pattern data, and body posture data to generate a comprehensive rehabilitation baseline that includes core functional status, pelvic floor muscle health index, and psychological stress level; and extracting key feature vectors from the rehabilitation baseline using principal component analysis for subsequent stage division and movement adaptation.

[0011] The multimodal data fusion algorithm includes aligning timestamps from different data sources using time synchronization technology; calculating the similarity matrix between core muscle activity data and breathing pattern data using a dynamic time warping algorithm; and generating a comprehensive rehabilitation baseline using the weighted fusion formula F=α⋅M+β⋅R+γ⋅P, where M represents the core muscle activity score, R represents the breathing pattern health score, P represents the body posture stability score, and the weighting coefficients α, β, and γ are dynamically adjusted according to the number of postpartum weeks to ensure the prominence of characteristics at each stage.

[0012] The process of dividing postpartum recovery into stages involves constructing a stage division model based on key feature vectors in the comprehensive rehabilitation baseline and the postpartum timeline; dividing the postpartum period into the preparatory stage (week 1), early stage (weeks 2-4), middle stage (weeks 5-6), late stage (week 7), and final stage (week 8), with each stage corresponding to specific training objectives and movement complexity; classifying key feature vectors using a clustering algorithm to determine the best matching movement library for the current stage; and generating an initial training plan based on the classification results by calling preset movement parameter templates.

[0013] The application motion adaptation algorithm includes: assessing the safety of the motions in the initial training plan and filtering high-risk motions using a taboo rule base; classifying the remaining motions by difficulty and dynamically adjusting motion parameters (such as number of sets, number of repetitions per set, and rest intervals) based on the mother's real-time exercise performance data; and using formulas... Calculate the motion adaptation index ,in This indicates the quality score of the action completion. Indicates the duration of the action. Indicates the difficulty coefficient of the movement, decay factor Adjustments are made dynamically based on the number of weeks postpartum to ensure that the adaptability of movements gradually improves as the recovery progresses.

[0014] The real-time monitoring module includes collecting data on core muscle activation, respiratory rate changes, and body posture stability during training through wearable sensors; comparing the collected data with preset standard thresholds to generate a training quality score; triggering a dynamic adjustment mechanism when the score is lower than the preset threshold, and re-invoking the motion adaptation algorithm to generate new training parameters; and displaying training progress and adjustment suggestions through a visual interface to enhance user engagement and compliance.

[0015] Another objective of this invention is to provide a postpartum phased functional rehabilitation training system, which, through an intelligent rehabilitation training system, solves the problems of lack of dynamic assessment and personalized adjustment in traditional programs, thereby improving rehabilitation efficiency. A postpartum phased functional rehabilitation training system includes: a data acquisition module for collecting core muscle group activity data, breathing pattern data, and body posture data via wearable sensors; a rehabilitation baseline generation module for generating an individualized rehabilitation baseline based on the collected data; a phase division module for dividing postpartum recovery phases according to the rehabilitation baseline; a movement adaptation module for calling up a movement library and generating an adapted training plan; and a real-time monitoring module for collecting movement performance data during training and dynamically adjusting training parameters.

[0016] The postpartum phased functional rehabilitation training method provided by this invention generates a comprehensive rehabilitation baseline through multimodal data fusion, which can accurately reflect the core functional status and recovery needs of postpartum women, effectively avoiding risky movements caused by the lack of professional assessment in traditional programs. This invention dynamically adjusts training parameters through a movement adaptation algorithm, enabling efficient planning of the postpartum woman's rehabilitation process and optimizing the problems of delayed or overtraining that may occur due to a one-size-fits-all design in traditional programs. Utilizing a dynamic time warping algorithm and a taboo rule base, the safety and adaptability of movements are precisely guaranteed, further improving the scientific nature and controllability of the rehabilitation process. Furthermore, the real-time monitoring and adjustment mechanism of this invention allows rehabilitation training to dynamically respond to individual differences and changes in recovery progress, optimizing the training program in real time and avoiding rehabilitation stagnation or secondary injury. Through this intelligent rehabilitation approach, this invention not only improves the accuracy and safety of postpartum rehabilitation but also significantly enhances user experience and long-term adherence, possessing higher practical value. Attached Figure Description

[0017] Figure 1 The diagram shows the modular structure of the postpartum phased functional rehabilitation training system in an embodiment of the present invention, including a data acquisition module, a rehabilitation baseline generation module, a phase division module, a movement adaptation module, and a real-time monitoring module.

[0018] Figure 2This is a flowchart illustrating the multimodal data fusion algorithm in this embodiment of the invention, showing the process of time synchronization and alignment of core muscle group activity data, breathing pattern data, and body posture data, similarity matrix calculation, and comprehensive rehabilitation baseline generation.

[0019] Figure 3 This is a logic block diagram of the action adaptation algorithm in this embodiment of the invention, which includes key steps such as security assessment, difficulty classification, dynamic parameter adjustment, and action adaptation index calculation. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a postpartum phased functional rehabilitation training method and system. Its core lies in generating an individualized rehabilitation baseline through multimodal data fusion, and combining dynamic assessment and personalized adjustment mechanisms to achieve scientific and efficient postpartum rehabilitation training. The following will be illustrated in conjunction with the appendix. Figure 1 To be continued Figure 3 The specific implementation process and technical details of this invention are described in detail.

[0022] First, the overall modular structure of the system is as follows: Figure 1 As shown, the system includes a data acquisition module, a rehabilitation baseline generation module, a stage division module, a movement adaptation module, and a real-time monitoring module. These modules work together to ensure that postpartum women receive targeted rehabilitation training programs at different recovery stages. In practice, the data acquisition module uses wearable sensors to collect data on the postpartum woman's core muscle activity, breathing patterns, and body posture. This data forms the basis for generating an individualized rehabilitation baseline and is also a crucial basis for subsequent training plan development. The wearable sensors are designed with high sensitivity and low power consumption to ensure that prolonged wear will not cause discomfort to the postpartum woman. The data collected by the sensors is uploaded to the rehabilitation baseline generation module for processing via wireless transmission technology.

[0023] The workflow of the rehabilitation baseline generation module is as follows: Figure 2As shown, the process mainly includes three steps: data preprocessing, multimodal data fusion, and key feature extraction. In the data preprocessing stage, core muscle activity data undergoes a first preprocessing step of signal filtering and outlier removal to eliminate noise and invalid data; respiratory pattern data undergoes a second preprocessing step of respiratory rate calibration and depth normalization to ensure data consistency and accuracy. Subsequently, the multimodal data fusion algorithm synchronizes the core muscle activity data, respiratory pattern data, and body posture data in time. A similarity matrix between the core muscle activity data and respiratory pattern data is calculated using a dynamic time warping algorithm, and a comprehensive rehabilitation baseline is generated using the weighted fusion formula F=α⋅M+β⋅R+γ⋅P. Here, M represents the core muscle activity score, R represents the respiratory pattern health score, P represents the body posture stability score, and the weighting coefficients α, β, and γ are dynamically adjusted according to the postpartum weeks. For example, during the preparatory phase (week 1), due to the mother's relatively weak physical condition, the weight α of the core muscle activity score M is lower, while the weights β and γ of the breathing pattern health score R and the body posture stability score P are relatively higher, highlighting the importance of respiratory regulation and postural stability. Key feature vectors from the comprehensive rehabilitation baseline are extracted using principal component analysis for subsequent stage segmentation and movement adaptation.

[0024] The phase segmentation module constructs a phase segmentation model based on key feature vectors from the comprehensive rehabilitation baseline and the postpartum timeline, dividing postpartum recovery into five phases: preparatory phase (week 1), early phase (weeks 2-4), middle phase (weeks 5-6), late phase (week 7), and final phase (week 8). Each phase corresponds to specific training objectives and movement complexity. For example, in the preparatory phase, the main objectives are pain relief, improved blood circulation, and initial recovery of core muscle function; therefore, training movements focus on simple breathing exercises and gentle core muscle activation. In the final phase, the goal is to fully restore functional motor abilities, and training movements include higher-intensity core stability training and whole-body coordination exercises. A clustering algorithm is used to classify the key feature vectors, determine the best-matching movement library for the current phase, and generate an initial training plan by calling preset movement parameter templates.

[0025] The motion adaptation module is responsible for optimizing the initial training plan, ensuring the safety and suitability of the training movements. For example... Figure 3As shown, the motion adaptation algorithm includes three key steps: safety assessment, difficulty grading, and dynamic parameter adjustment. In the safety assessment phase, a taboo rule base is used to filter high-risk movements. For example, for postpartum women whose pelvic floor muscles have not fully recovered, the taboo rule base automatically excludes movements involving jumping or violent impacts. In the difficulty grading phase, movements are categorized into beginner, intermediate, and advanced levels based on their complexity and physical exertion, and the motion parameters are dynamically adjusted in conjunction with the postpartum woman's real-time performance data. For example, if a postpartum woman demonstrates a high quality of performance score when performing a certain set of movements... and shorter action execution time Then it can be done through the formula. Calculate the motion adaptation index ,in This indicates the quality score of the action completion. Indicates the duration of the action. Indicates the difficulty coefficient of the movement, decay factor Adjustments are made dynamically based on the number of weeks postpartum. In the early stages, The value is relatively large to reduce the difficulty of the movement; while in the final stage, The value is gradually decreased to progressively increase the training intensity. The final generated movement fit index... This will serve as the basis for adjusting movement parameters (such as number of sets, number of repetitions per set, and rest intervals) to ensure that the training plan can adapt to the postpartum recovery process.

[0026] The real-time monitoring module continuously collects data on core muscle activation, respiratory rate changes, and body posture stability during training using wearable sensors. This data is compared to preset thresholds to generate a training quality score. When the score falls below the threshold, a dynamic adjustment mechanism is triggered, re-invoking the motion adaptation algorithm to generate new training parameters. For example, if the mother experiences insufficient core muscle activation during training, the system will automatically reduce the difficulty of the exercises or increase the rest intervals to avoid over-fatigue. Furthermore, the real-time monitoring module displays training progress and adjustment suggestions through a visual interface, enhancing user engagement and adherence. The visual interface design should be simple and intuitive, allowing mothers to easily understand their recovery status and training effectiveness.

[0027] To further illustrate the practical application of this invention, the following describes the rehabilitation training process of a postpartum woman in her third week of postpartum as an example. The woman began data collection by wearing wearable sensors, which recorded her core muscle activity data, breathing pattern data, and body posture data, and uploaded the data to the rehabilitation baseline generation module. After data preprocessing and multimodal data fusion, a comprehensive rehabilitation baseline was generated, showing a core muscle activity score of 0.6, a breathing pattern health score of 0.8, and a body posture stability score of 0.7. Based on the postpartum timeline and key feature vectors, the system determined that she was in the early stage and used the early stage movement library to generate an initial training plan. The initial training plan included exercises such as supine leg raises, bridge pose, and abdominal breathing exercises. In the movement adaptation module, the system performed a safety assessment of the above movements. After confirming that there were no high-risk movements, the system dynamically adjusted the movement parameters based on the difficulty level and real-time exercise performance data. For example, the number of sets for abdominal breathing exercises was adjusted from 3 to 4, the number of repetitions per set increased from 10 to 12, and the rest interval was shortened from 30 seconds to 20 seconds. During training, the real-time monitoring module continuously collects data and generates training quality scores. When it is found that the mother's core muscle activation is insufficient during the bridge exercise, the system automatically reduces the number of bridge sets from 3 to 2 and prompts the mother to appropriately extend the rest interval through a visual interface.

[0028] This invention addresses the lack of dynamic assessment and personalized adjustments in traditional rehabilitation programs through an intelligent rehabilitation training system, significantly improving rehabilitation efficiency. A multimodal data fusion algorithm ensures the accuracy and comprehensiveness of the rehabilitation baseline, while a movement adaptation algorithm dynamically adjusts training parameters to achieve the scientific rigor and controllability of the training plan. A real-time monitoring and adjustment mechanism effectively addresses individual differences and changes in recovery progress. Through this intelligent rehabilitation approach, this invention not only improves the accuracy and safety of postpartum rehabilitation but also significantly enhances user experience and long-term adherence, demonstrating broad practical value.

Claims

1. A postpartum phased functional rehabilitation training method, characterized in that, include: Wearable sensors collect data on the mother's core muscle activity, breathing patterns, and body posture. Combined with subjective feedback from mobile devices, an individualized rehabilitation baseline is generated. Based on the rehabilitation baseline, the postpartum recovery stages are divided, and corresponding action libraries are called to generate training plans. Action adaptation algorithms are applied to optimize the training plans to ensure the safety and adaptability of the actions. By collecting exercise performance data during training through a real-time monitoring module, training parameters are dynamically adjusted to generate a rehabilitation plan for the next stage.

2. The postpartum phased functional rehabilitation training method as described in claim 1, characterized in that, The generation of an individualized rehabilitation baseline includes: a first preprocessing step on core muscle group activity data, which includes signal filtering and outlier removal; a second preprocessing step on respiratory pattern data, which includes respiratory rate calibration and depth normalization; using a multimodal data fusion algorithm to align and analyze core muscle group activity data, respiratory pattern data, and body posture data to generate a comprehensive rehabilitation baseline that includes core functional status, pelvic floor muscle health index, and psychological stress level; and extracting key feature vectors from the rehabilitation baseline using principal component analysis for subsequent stage division and movement adaptation.

3. The postpartum phased functional rehabilitation training method as described in claim 2, characterized in that, The multimodal data fusion algorithm includes: aligning timestamps from different data sources using time synchronization technology; calculating the similarity matrix between core muscle activity data and breathing pattern data using a dynamic time warping algorithm; and generating a comprehensive rehabilitation baseline using the weighted fusion formula F=α⋅M+β⋅R+γ⋅P, where M represents the core muscle activity score, R represents the breathing pattern health score, P represents the body posture stability score, and the weighting coefficients α, β, and γ are dynamically adjusted according to the number of weeks postpartum.

4. The postpartum phased functional rehabilitation training method as described in claim 1, characterized in that, The process of dividing postpartum recovery into stages includes: constructing a stage division model based on key feature vectors in the comprehensive rehabilitation baseline and the postpartum timeline; dividing the postpartum period into the preparatory stage (week 1), early stage (weeks 2-4), middle stage (weeks 5-6), late stage (week 7), and final stage (week 8), with each stage corresponding to specific training objectives and movement complexity; classifying key feature vectors using a clustering algorithm to determine the best matching movement library for the current stage; and generating an initial training plan based on the classification results by calling preset movement parameter templates.

5. The postpartum phased functional rehabilitation training method as described in claim 1, characterized in that, The application motion adaptation algorithm includes: assessing the safety of motions in the initial training plan and filtering high-risk motions using a taboo rule base; classifying the remaining motions by difficulty and dynamically adjusting motion parameters based on the mother's real-time movement performance data; and using formulas... Calculate the motion adaptation index ,in This indicates the quality score of the action completion. Indicates the duration of the action. Indicates the difficulty coefficient of the movement, decay factor Adjust dynamically based on the number of weeks postpartum.

6. The postpartum phased functional rehabilitation training method as described in claim 1, characterized in that, The real-time monitoring module includes: collecting data on core muscle activation, respiratory rate changes, and body posture stability during training using wearable sensors; comparing the collected data with preset standard thresholds to generate a training quality score; triggering a dynamic adjustment mechanism when the score is lower than the preset threshold, and re-invoking the motion adaptation algorithm to generate new training parameters; and displaying training progress and adjustment suggestions through a visual interface.

7. A postpartum phased functional rehabilitation training system, characterized in that, include: The data acquisition module (1) is used to collect core muscle group activity data, breathing pattern data and body posture data through wearable sensors; The rehabilitation baseline generation module (2) is used to generate an individualized rehabilitation baseline based on the collected data; the stage division module (3) is used to divide the postpartum recovery stages according to the rehabilitation baseline; and the movement adaptation module (4) is used to call the movement library and generate an adapted training plan. The real-time monitoring module (5) is used to collect sports performance data during training and dynamically adjust training parameters.

8. The postpartum phased functional rehabilitation training system as described in claim 7, characterized in that, The rehabilitation baseline generation module (2) includes: a data preprocessing unit for filtering signals and removing outliers from core muscle group activity data, and calibrating respiratory frequency and normalizing depth for respiratory pattern data; a multimodal data fusion unit for synchronizing core muscle group activity data, respiratory pattern data and body posture data in time, and generating a comprehensive rehabilitation baseline through a weighted fusion formula; and a feature extraction unit for extracting key feature vectors from the comprehensive rehabilitation baseline using principal component analysis.

9. The postpartum phased functional rehabilitation training system as described in claim 7, characterized in that, The action adaptation module (4) includes: a safety assessment unit, used to filter high-risk actions through a taboo rule base; a difficulty grading unit, used to grade actions according to their complexity and physical exertion; and a parameter adjustment unit, used to dynamically adjust action parameters in conjunction with real-time sports performance data and calculate the action adaptation index using a formula.

10. The postpartum phased functional rehabilitation training system as described in claim 7, characterized in that, The real-time monitoring module (5) includes: a data acquisition unit for acquiring core muscle activation, respiratory rate changes and body posture stability data through wearable sensors; a scoring generation unit for comparing the acquired data with a preset standard threshold to generate a training quality score; a dynamic adjustment unit for re-calling the action adaptation algorithm to generate new training parameters when the score is lower than the preset threshold; and a visualization unit for displaying training progress and adjustment suggestions through a visualization interface.

Images