A system for exercise monitoring and management for patients with diabetic foot

Abstract

The application discloses a kind of exercise monitoring management systems for diabetic foot patients, specifically relates to the field of diabetic patient monitoring, for solving the problem that existing exercise monitoring mode is difficult to effectively identify individualized risk trend and implement precise intervention;Through the pressure acquisition module, the main pressure area of the plantar under different gait is obtained, the pressure analysis module constructs the time sequence evolution track of the plantar stress direction, the electrical impedance acquisition module synchronously constructs the electrical impedance response sequence in the main pressure area of the plantar, the matrix construction module is jointly encoded into two-dimensional time sequence matrix of double channel, the lag extraction module analyzes the impedance lag area, the risk analysis module identifies the spatial overlap degree of impedance lag area and the main pressure area of the next cycle, the strategy construction module constructs risk trend turning point index with minimum point of coincidence degree and establishes individual exercise monitoring strategy based on different gait parameters, finally drives monitoring terminal to realize dynamic exercise intervention and prompt.

Description

Technical Field

[0001] This invention relates to the field of monitoring technology for diabetic patients, and more specifically, to a motion monitoring and management system for diabetic foot patients. Background Technology

[0002] Patients with diabetic foot face the dilemma of the "dual effect of exercise intervention" in daily life and rehabilitation: appropriate exercise can promote peripheral blood circulation in the foot, delay nerve degeneration, and enhance local tissue metabolism and functional recovery, making it an important part of diabetic foot rehabilitation; however, if exercise lacks control and monitoring, it can easily induce concentrated plantar pressure, edema spread, or accumulation of microtrauma, especially in high-risk individuals with atrophied plantar fat pads, loss of nerve sensation, and insufficient blood supply. In these cases, exercise itself may become a direct trigger for ulcer recurrence and structural damage. In reality, patients often face a dilemma between exercise and inactivity. On the one hand, they hope to improve foot function through walking training; on the other hand, due to the lack of professional monitoring mechanisms, they cannot determine whether their current exercise has turned into a risky state, thus passively bearing the irreversible consequences of injury. Currently used foot monitoring devices mostly focus on static pressure or temperature threshold monitoring, lacking the comprehensive ability to identify the changing trends of foot condition during exercise, and cannot determine whether the patient's current exercise behavior is promoting stability or inducing deterioration. Especially at home or in non-medical environments, patients are very likely to gradually enter a risk channel of high pressure retention or gait asymmetry accumulation in seemingly stable movements without realizing it, leading to avoidable re-ulceration events.

[0003] Therefore, there is an urgent need for an information-based monitoring system that can dynamically identify the evolution trend of foot risks, combine motor behavior characteristics to provide timely feedback and intervention, adapt to individual motor states at various stages of diabetic foot evolution, and support safe rehabilitation management. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a motion monitoring and management system for diabetic foot patients to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A motion monitoring and management system for patients with diabetic foot includes a pressure acquisition module, a pressure analysis module, an electrical impedance acquisition module, a matrix construction module, a hysteresis extraction module, a risk analysis module, a strategy construction module, and a monitoring and guidance module, wherein:

[0007] The pressure acquisition module collects the plantar pressure distribution of diabetic foot patients during different gait cycles and extracts the main pressure-bearing areas of the plantar surface.

[0008] The pressure analysis module extracts the force vector direction of the main pressure area of ​​the sole within each movement cycle and constructs the temporal evolution trajectory of the force direction of the sole.

[0009] The impedance acquisition module constructs an impedance response sequence by setting an impedance acquisition unit and sampling impedance in the main pressure area of ​​the sole of the foot at a preset multiple of the pressure acquisition frequency.

[0010] The matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction and the electrical impedance response sequence to generate a two-channel two-dimensional time-series matrix with spatial location as the index and time as the axis.

[0011] The hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the time window of pressure release.

[0012] The risk analysis module analyzes the spatial overlap between the current impedance recovery hysteresis region and the main pressure region of the foot in the next motion cycle;

[0013] The strategy construction module identifies the minimum point of the overlap index in the spatially overlapping area as the risk trend inflection point, and extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy.

[0014] The monitoring and guidance module dynamically drives the monitoring terminal to execute exercise guidance based on the output of the strategy construction module.

[0015] In a preferred embodiment, the pressure acquisition module acquires the plantar pressure distribution of diabetic foot patients during different gait cycles, and extracts the main pressure-bearing areas of the plantar surface, specifically including:

[0016] Multiple motion cycles are preset, and pressure is sampled from the patient's soles through a pressure acquisition unit. The pressure sampling values ​​within all motion cycles are then divided according to gait parameters.

[0017] The gait parameters include cadence, stride length, and the proportion of support phase duration, with each movement cycle corresponding to a set of gait parameters;

[0018] The main pressure-bearing regions are merged using a pressure intensity clustering method, and the main pressure-bearing regions with stable structures under the same synchronous parameters are extracted through spatial connectivity screening.

[0019] In a preferred embodiment, the pressure analysis module extracts the force vector direction of the main pressure area of ​​the foot in each movement cycle, and constructs the temporal evolution trajectory of the force direction of the foot, specifically including:

[0020] Within each major pressure zone of the foot, the continuous displacement direction of the pressure centroid vector is calculated using the acquisition time as an index;

[0021] Trajectory smoothing and trend fitting operations are performed on the vector sequence within each motion cycle to generate the temporal evolution trajectory of the force direction on the sole of the foot. The evolution trajectory is represented in the form of spatial grid coordinates.

[0022] The spatial grid size is the coverage area of ​​a single pressure acquisition unit;

[0023] The trajectory smoothing method uses spatial vector aggregation to suppress local fluctuations and filter out local disturbance vectors.

[0024] In a preferred embodiment, the matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction with the electrical impedance response sequence to generate a dual-channel two-dimensional time-series matrix with spatial location as the index and time as the axis. Specifically, this includes:

[0025] Using the spatial coordinates of the evolution trajectory of the plantar force direction as the index dimension, pressure time series channels and electrical impedance time series channels are constructed based on pressure value and electrical impedance response value in each motion cycle;

[0026] A two-dimensional time series matrix is ​​established based on a dual-channel coding structure, where each spatial grid coordinate corresponds to a matrix unit, and the matrix unit contains time series channel values ​​in two dimensions.

[0027] In a preferred embodiment, the hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the pressure release time window, specifically including:

[0028] For each matrix cell in the matrix, the pressure release point is taken as the peak inflection point of the pressure value in the pressure time series channel. The time window in which the pressure value at the pressure release point drops to the preset baseline interval is traced back, and the corresponding impedance response value of the impedance time series channel is extracted within this time window.

[0029] Calculate the local recovery rate for the electrical impedance value within the spatial grid coordinates corresponding to each matrix element;

[0030] Set a threshold for impedance recovery rate, and perform cumulative area measurement on matrix units that do not reach the recovery rate to extract the hysteresis region of impedance recovery.

[0031] In a preferred embodiment, the risk analysis module analyzes the spatial overlap between the current impedance recovery hysteresis region and the main pressure region of the foot in the next motion cycle, specifically including:

[0032] After obtaining the hysteresis region of impedance recovery, extract the set of pressure peak inflection points in the pressure time series channel of the matrix unit in the next motion cycle, and label the pressure peak region in the form of grid coordinates.

[0033] Spatially register the pressure peak inflection point region of the next motion cycle with the impedance recovery lag region of the current cycle, calculate the percentage of their overlapping area, and output the spatial overlap region coincidence index.

[0034] In a preferred embodiment, the strategy construction module identifies the minimum point of the overlap index in spatially overlapping regions as a risk trend inflection point, and simultaneously extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy, specifically including:

[0035] Based on the overlap index of spatial overlapping areas throughout the entire motion cycle, the minimum value of the overlap index is selected as the risk trend inflection point.

[0036] Based on gait parameters, a set of motion cycles is divided, and the average overlap index of motion cycles corresponding to different gait parameters is statistically analyzed to conduct spatial risk assessment in the gait dimension.

[0037] The time index of the risk trend inflection point in the time series matrix is ​​extracted to determine the moment of risk trend inflection. At the same time, the gait parameter with the minimum average overlap index in all motion cycles is extracted. The two are combined as the parameters of the patient's motion monitoring strategy and output to the monitoring guidance module.

[0038] In a preferred embodiment, the dynamic drive monitoring terminal performs motion guidance specifically by:

[0039] Based on the motion monitoring strategy parameters output by the strategy construction module, the monitoring terminal is dynamically driven to execute motion time, gait guidance, and early warning prompts.

[0040] The technical effects and advantages of the present invention, a sports monitoring and management system for diabetic foot patients:

[0041] By integrating multiple functional modules such as pressure acquisition, impedance response analysis, and dual-channel time-series matrix construction, the system can dynamically acquire the spatial structure and physiological response changes of the main pressure area of ​​the foot under different time states, enabling joint modeling of the evolution of plantar force direction and impedance recovery characteristics. The system employs spatial overlap analysis to identify the overlap between the impedance recovery lag region and the main pressure area in the next cycle, accurately judging potential risk evolution trends. It also identifies individualized risk turning points through the minimum overlap, effectively avoiding secondary tissue damage caused by blind exercise. By constructing a dynamic monitoring strategy including movement parameters such as cadence, stride length, and stance phase, the monitoring terminal possesses real-time response capabilities that are tailored to the individual and adjust according to the situation, significantly improving the safety and individual adaptability of exercise intervention for diabetic foot patients. While ensuring the physiological recovery of the foot, it provides a highly accurate support system for chronic disease self-management. Attached Figure Description

[0042] Figure 1This is a schematic diagram of a sports monitoring and management system for diabetic foot patients according to the present invention. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0044] Example 1

[0045] Figure 1 This invention provides a motion monitoring and management system for diabetic foot patients, comprising a pressure acquisition module, a pressure analysis module, an electrical impedance acquisition module, a matrix construction module, a hysteresis extraction module, a risk analysis module, a strategy construction module, and a monitoring and guidance module, wherein:

[0046] The pressure acquisition module collects the plantar pressure distribution of diabetic foot patients during different gait cycles and extracts the main pressure-bearing areas of the plantar surface.

[0047] The pressure analysis module extracts the force vector direction of the main pressure area of ​​the sole within each movement cycle and constructs the temporal evolution trajectory of the force direction of the sole.

[0048] The impedance acquisition module constructs an impedance response sequence by setting an impedance acquisition unit and sampling impedance in the main pressure area of ​​the sole of the foot at a preset multiple of the pressure acquisition frequency.

[0049] The matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction and the electrical impedance response sequence to generate a two-channel two-dimensional time-series matrix with spatial location as the index and time as the axis.

[0050] The hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the time window of pressure release.

[0051] The risk analysis module analyzes the spatial overlap between the current impedance recovery hysteresis region and the main pressure region of the foot in the next motion cycle;

[0052] The strategy construction module identifies the minimum point of the overlap index in the spatially overlapping area as the risk trend inflection point, and extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy.

[0053] The monitoring and guidance module dynamically drives the monitoring terminal to execute exercise guidance based on the output of the strategy construction module.

[0054] The pressure acquisition module collects the plantar pressure distribution of diabetic foot patients during different gait corresponding movement cycles and extracts the main pressure-bearing areas of the plantar surface.

[0055] Patients with diabetic foot were instructed to walk continuously on a designated path on a flat, hard surface, maintaining a natural gait throughout. The number of steps in each monitoring cycle was preset based on the severity of the patient's condition. If not preset, a cycle was defined as two foot-to-foot phases, i.e., the complete sequence of motion from foot landing, supporting the foot, to the next landing. In actual testing, pressure acquisition units embedded in insoles or smart plantar patches continuously sampled the contact pressure in different areas of the sole. The sampling frequency was set to 10Hz to ensure complete coverage of the dynamic contact process within the common gait frequency range (approximately 90 to 120 steps per minute). Each test included at least 20 complete cycle measurements, covering various postural variations in the patient.

[0056] After data acquisition, the pressure value sequence for each movement cycle is divided according to the synchronously acquired gait parameters. Gait parameters are extracted based on the shape of the pressure change curve within a continuous movement cycle. Specifically, step frequency is determined by the number of cycles per unit time; step length is estimated by the lateral displacement difference between the front and rear support points; and the support period duration ratio is calculated based on the ratio of the duration of stable support under pressure within a single step cycle to the total cycle duration. This set of gait parameters forms the status label for each movement cycle. The pressure distribution data is represented by pressure intensity.

[0057] After gait parameter segmentation, the pressure distribution data under various similar gait conditions are processed to identify the main pressure-bearing regions. Specifically, pressure data from multiple motion cycles corresponding to the same type of gait parameters are first aggregated and superimposed to form a statistically significant pressure heatmap cluster. Based on this, a region merging mechanism based on regional intensity clustering is introduced. The intensity clustering process uses a fixed hierarchical threshold setting method to divide the pressure value range into five levels: low-pressure zone (0–50 kPa), sub-low-pressure zone (51–100 kPa), medium-pressure zone (101–150 kPa), sub-high-pressure zone (151–200 kPa), and high-pressure zone (above 200 kPa). Each level of region is assigned a different clustering label on the heatmap. Merging is performed based on the adjacent distribution relationship of the clustering labels in the image space, retaining structurally coherent and relatively concentrated core regions as the initial candidate set of main pressure-bearing regions. After region merging, spatial connectivity filtering is performed on the resulting candidate set. Specifically, connectivity analysis is performed on each merged region in a two-dimensional plantar coordinate system, and the boundary closure and internal continuity of each region are determined based on the 8-neighbor connectivity structure. For regions whose connectivity does not meet the closure threshold (with a minimum closure area set to 1 cm²),... 2Areas with a longest non-connected boundary less than 2mm are discarded to ensure the extraction results have spatial anatomical rationality and structural stability. The extracted pressure zones above the set pressure level (default high-pressure zone) are ultimately designated as the main pressure zones.

[0058] The pressure analysis module extracts the force vector direction of the main pressure area on the sole of the foot within each movement cycle, and constructs the temporal evolution trajectory of the force direction on the sole of the foot.

[0059] The identified primary pressure-bearing area of ​​the foot is used as the analysis object, and the pressure values ​​of all pressure sampling points within this area are extracted at the current moment. Then, based on all pressure sampling points within the primary pressure-bearing area in the current frame, combined with their position coordinates in the two-dimensional plantar coordinate system, the pressure weighting center of this area is calculated to determine the position of the pressure center of gravity at the current moment. Connecting the pressure center of gravity positions at multiple consecutive time points forms a pressure center of gravity displacement path, which is the initial vector sequence within the current movement cycle. In actual sampling, this sequence often exhibits short-term directional jumps or spatial spikes due to minor perturbations, gait fluctuations, or local neural control abnormalities. To ensure the continuity and interpretability of the force direction trajectory, trajectory smoothing processing needs to be performed on this initial vector sequence.

[0060] Specifically, the trajectory smoothing method employs a spatial vector aggregation strategy. Within each vector segment consisting of three consecutive time sampling points, the changes in its directional angle and magnitude are evaluated. For midpoints where the direction deviates from a threshold (e.g., exceeding 25°), direction substitution is performed, replacing the midpoint with the weighted average direction of the preceding and following segments, thus eliminating the interference of local disturbances on the overall trajectory. Simultaneously, extremely short vectors (magnitude less than 1mm) are deleted to remove spurious vectors caused by brief periods of stillness in the foot region or data jitter. After trajectory smoothing, all valid vector segments are trend-fitted sequentially over time. Trend fitting uses a rolling window method, performing local direction fitting on each trajectory segment to identify its overall force trend direction. For example, during a five-frame window rolling process, the average direction vector within the window is extracted to form a representative set of trajectory segments.

[0061] The spatial grid division rules are determined based on the physical design structure of the pressure acquisition unit. A conventional approach involves deploying 64 to 128 pressure acquisition points to cover the entire sole area, with a spacing of 8mm to 10mm between points. In this embodiment, the spatial grid size is uniformly set to 8mm × 8mm (consistent with the coverage area of ​​a single pressure acquisition unit), strictly corresponding to the coverage area of ​​a single pressure acquisition unit to ensure the anchoring accuracy of subsequent trajectory and impedance data during spatial matching.

[0062] The matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction and the electrical impedance response sequence to generate a dual-channel two-dimensional time-series matrix with spatial location as the index and time as the axis.

[0063] The impedance acquisition module's impedance acquisition units are positioned precisely to correspond with the pressure acquisition units, covering the entire plantar area and focusing sampling on the primary pressure zone. The impedance acquisition units employ a high-frequency, multi-channel measurement mode to achieve real-time capture of the local tissue's electrical state. Considering that the response rate of impedance is generally slower than the rate of change in mechanical pressure, to fully reproduce the electrophysiological changes in tissue during pressure application and release, the impedance sampling frequency is set to 2.5 times the pressure sampling frequency. For example, if the pressure acquisition frequency is 10Hz, the impedance sampling frequency is set to 25Hz, ensuring that more than 25 sets of effective impedance response data can be obtained within each complete movement cycle. In actual execution, impedance acquisition uses the primary pressure zone as the boundary, collecting impedance values ​​from all sampling nodes within this area during each movement cycle, starting from the initial gait posture's ground contact moment until the end of the support phase foot lift. Each set of sampled data includes spatial location markers and real-time impedance values. The measurement uses a multi-electrode array with a four-electrode configuration to reduce interference from contact resistance. The sampling method is sequential scanning, and the switching time between channels is less than 1 millisecond to ensure spatial consistency.

[0064] Based on the generated trajectory of plantar force evolution as a spatial reference, all spatial grid coordinates involved are processed point-by-point. Within each motion cycle, using each coordinate point in the trajectory as an anchor unit, time period data corresponding to that coordinate are extracted from both the pressure acquisition sequence and the impedance acquisition sequence to construct a dual-channel time series for that coordinate. Specifically, within the entire motion cycle, pressure acquisition values ​​for individual grid coordinate points are obtained, and the time intervals are arranged according to the pressure sampling frequency to form a discrete time series. Each sequence point contains three elements: time index, pressure intensity value, and data validity marker (used to exclude non-primary pressure areas or low signal points caused by extremely small contact areas). To ensure time axis alignment between channels, interpolation is used to accurately map the sampling time to the time axis of the impedance channel. After processing, a complete pressure time series channel is formed. Subsequently, the corresponding impedance value is extracted at each spatial coordinate point to construct an impedance time series of the same length as the pressure channel. This series reflects the bioelectrical changes under specific stress conditions and has strong pathological relevance. Ultimately, each spatial grid coordinate point in the trajectory of force evolution will have a complete set of dual-channel time series, representing the dynamic curves of pressure and electrical impedance changing with time.

[0065] Subsequently, at the data structure organization level, the fusion of plantar physiological and mechanical responses was completed, constructing a two-dimensional time-series matrix with spatial location as the row index and dual-channel time evolution as the column structure. The matrix space is the set of grid coordinates that appeared in all force trajectories. For each grid coordinate point, a matrix cell was initialized, with two fields set inside the cell, corresponding to the pressure channel sequence and the electrical impedance channel sequence, respectively. During matrix construction, data was filled into the coordinate points on all plantar trajectories one by one, in units of motion cycles. If a spatial point was not within the main pressure trajectory in a certain cycle, the matrix cell for that point was filled with an empty sequence marker and was not included in the calculation region for analysis in subsequent processing. For coordinate points in the effective region, the previously constructed pressure and electrical impedance time channels were filled into the matrix cells respectively, and the data were arranged in a unified time axis order, exhibiting good longitudinal temporal consistency and lateral spatial comparability.

[0066] To enhance the integrity of the matrix in terms of spatial structure and increase the ability of grid neighborhood analysis, a spatial adjacency index structure is established for all matrix elements to record the relative positions between adjacent grid elements, facilitating region fusion and gradient diffusion identification during subsequent hysteresis recovery analysis. Simultaneously, a spatial labeling layer is added to the end of the entire matrix, recording information such as whether each point belongs to the main path region, whether it is located at the boundary of the impedance hysteresis region, or the overlapping core region, providing clear spatial labeling support for trend analysis and monitoring strategy formulation.

[0067] The hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the time window of pressure release.

[0068] For each matrix cell in the two-dimensional dual-channel time series matrix, the identification of the pressure release time window and the interception of impedance values ​​are performed. First, the pressure time series channel corresponding to the current cell is located in the matrix. This channel records the pressure response trend over time at the spatial grid coordinates within one motion cycle. Using this pressure sequence as input, its local peak inflection point is identified as a reference marker for determining the starting point of the release behavior. The peak inflection point is identified using the first derivative zero-crossing method. By calculating the difference between adjacent sampling points, the time point when the pressure curve reaches its maximum value and begins to decline is determined. This can be further confirmed by combining a monotonically decreasing trend of more than 5 consecutive frames (each frame considered a monitoring time point) to avoid misjudgment due to single-point fluctuations. After the peak inflection point is confirmed as the pressure release point, the pressure values ​​after that point are further traced back, scanning a certain length of time series until it drops to a set baseline interval. This baseline interval is not set as an absolute zero point, but is preset based on the average pressure response distribution of an individual in a resting state, usually set to an interval below 10 kPa, and requires a stable downward trend for at least 5 frames to form a clear release process window. The pressure drop period is the target window for impedance response analysis. The duration of the entire time window must be no less than 200 ms to ensure that the impedance response has a complete evolution process. After the time window is identified, the impedance time series channels in the same matrix unit are indexed to extract all impedance values ​​within the corresponding time window, forming a complete impedance response curve.

[0069] After extracting the impedance response sequence corresponding to the pressure release time window in each matrix unit, the calculation of local recovery rate and extraction of lag regions are performed. The recovery rate is calculated using the initial impedance value at the start of the time window as a reference point. The overall recovery amplitude and speed trend of the impedance value within the window time range are recorded, and the average recovery rate of each unit is calculated by dividing the difference between the final value and the initial value by the duration. To improve noise resistance, a 5-point moving average is introduced during the recovery rate calculation process to eliminate the influence of single-frame abrupt changes and improve the smoothness of the response curve fitting. Addressing the significant differences in baseline impedance levels among individuals in practical applications, the recovery rate threshold is not a uniform fixed value but is individually set based on the average baseline recovery rate of the patient at rest. This threshold is typically set to 70% of the average resting recovery rate. If the impedance recovery rate of a unit is lower than this set threshold, it indicates that the local tissue exhibits a relatively lagging recovery state during the motion cycle, posing a potential pathological risk.

[0070] After evaluating the recovery rate of all units, all units in the matrix below the recovery rate threshold are spatially labeled, and their spatial distribution map in a two-dimensional coordinate system is constructed. Using connectivity rules, adjacent low-rate recovery units are merged into a hysteresis region, and its boundary contour is recorded. To quantitatively assess the spatial impact of this hysteresis region, an area measurement operation is performed. Based on the area reference of the sensing grid corresponding to each unit, point-by-point accumulation calculations are performed to finally obtain the total area of ​​the hysteresis recovery region.

[0071] The risk analysis module analyzes the spatial overlap between the current impedance recovery lag region and the main pressure region of the foot in the next motion cycle.

[0072] After extracting the impedance recovery hysteresis region, spatial labeling of the plantar pressure peak region is performed for the next adjacent motion cycle. Based on this, the two-dimensional time series matrix in the next motion cycle is traversed, and for each spatial matrix unit, the full sequence response of its pressure time series channel is extracted. Local peak inflection point identification is then performed, and the process is consistent with that in the hysteresis extraction module.

[0073] After completing the meshing and labeling of the pressure peak region, a spatial overlap analysis is performed between the corresponding region and the impedance recovery hysteresis region of the previous cycle. The structure of the impedance recovery hysteresis region extracted from the previous cycle is imported, ensuring that its boundary information, mesh cell index, and hysteresis labels are fully loaded. Due to slight differences in the motion rhythm between the two cycles, a static registration operation of the coordinate system is performed before performing the region comparison to avoid spatial misalignment interfering with the registration results. The registration adopts an alignment mechanism based on the plantar center reference point, using the geometric center of the foot as the reference origin, aligning the spatial mask images in the two cycles to a unified reference coordinate system to ensure the consistency of position mapping.

[0074] After registration, all grid coordinate cells in the impedance recovery lag region are traversed sequentially to determine if they intersect with the corresponding coordinate cells in the current cycle's pressure peak region. A point-by-point matching mechanism is used; if both masks have a value of 1 at the same coordinate, they are considered overlapping cells and added to the overlap counter. The total number of cells in all impedance lag regions and the number of overlapping cells are recorded simultaneously, and the overlap index is calculated, which is the percentage of overlapping cells in the total number of cells in the impedance lag region. This index quantifies the severity of whether the pressure path is acting again on the unhealed area and is an important reference dimension for evaluating the improvement or deterioration of the movement trend. The spatial overlap index is output in floating-point form, ranging from 0 to 1. A higher value indicates a more significant re-superposition effect of the plantar high-pressure path on the unhealed area, suggesting a risk of further tissue damage. When the value approaches 0, it indicates that the pressure path in the new cycle spatially avoids the lag region or that the patient's foot impedance recovery ability has improved, indicating positive training potential. By clearly defining spatial geometric relationships, this solution enables cross-modal coupling recognition between physiological tissue recovery and external movement patterns, demonstrating its ability to perform patho-motor fusion analysis in diabetic foot exercise monitoring.

[0075] The strategy construction module identifies the minimum point of the overlap index in the spatially overlapping area as the risk trend inflection point, and extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy.

[0076] After completing the spatial registration and overlap index calculation of the impedance recovery lag region and pressure peak region within each cycle of the entire exercise cycle, a sequence of overlap indices arranged in chronological order was obtained. This sequence uses the exercise cycle as the time unit, with each time point corresponding to a precise floating-point value of spatial overlap. To identify key points where the risk of diabetic foot patients changes from worsening to improving, or from improving to worsening, during exercise, a minimum point extraction operation was performed on this sequence as a risk trend inflection point. Stability filtering was applied to the overlap index sequence, employing a five-point median smoothing mechanism to eliminate local measurement fluctuations and ensure the trend stability of extreme value identification. In the preprocessed sequence, a first-order difference operation was performed to find the sequence turning point from decreasing to increasing, and to determine whether this point is a local minimum of the current sequence. To avoid short-cycle fluctuations forming false minimums, a minimum interval threshold of 3 cycles was set for minimum value identification to ensure the continuous trend representativeness of the extracted points.

[0077] Once a minimum point satisfying the above conditions is identified, the system records the corresponding exercise cycle index as a turning point in the risk trend. Within the cycle corresponding to this node, the spatial overlap between the pressure peak area and the hysteresis recovery area is minimized, representing that the tissue structure achieves maximum pressure avoidance during exercise, or the patient achieves maximum impedance recovery exercise effect. This turning point will serve as a reference time point in the subsequent generation of individualized monitoring strategies.

[0078] After acquiring the spatial overlap index for all movement cycles, gait parameters are introduced as a classification basis to assess the risk level under individual gait patterns, thus dividing the movement cycles. Several discrete interval boundaries for gait parameters are preset, and the movement cycles are mapped to corresponding gait parameter sets based on the movement parameters collected within each cycle. After grouping, the overlap index values ​​for all cycles under each gait parameter group are statistically analyzed, and the average value within the group is calculated to form a "gait parameter-average overlap" mapping table. A weighted average method is used in the statistics, where the weight is the total pressure integral value within the current cycle to reflect the actual contribution of high-load cycles to risk. This gait dimension overlap index reflects whether the plantar high-pressure area repeatedly acts on the spatial structural behavior of areas that have not yet recovered tissue under different movement patterns, and is a core indicator for assessing the safety of movement patterns.

[0079] After identifying risk trend inflection points and assessing gait parameter risks, the two are jointly analyzed to generate individualized exercise monitoring strategy parameters. The specific location of the risk trend inflection point in the two-dimensional time-series matrix is ​​extracted from its time index, including the corresponding exercise cycle number and its start timestamp. This time point is defined as a critical moment of risk change, and changes in gait patterns, stress path variations, or impedance response characteristics before and after it are representative. Next, the gait parameter group corresponding to the minimum overlap index is retrieved from the "gait parameter-average overlap" mapping table, and the specific parameter values ​​of this group are extracted, including the stride frequency range, stride length range, and support phase duration percentage. This set of parameters is output as the most suitable exercise mode for the current individual and combined with the risk trend inflection point to form a monitoring strategy parameter set. This parameter set contains two core dimensions: first, the time anchor point for organizing risk trend changes in the current monitoring cycle; and second, the parameter combination corresponding to the optimal exercise mode. This strategy parameter set is encapsulated and synchronously pushed to the monitoring guidance module as direct input for dynamically adjusting exercise guidance.

[0080] The dynamic driving monitoring terminal performs motion guidance as follows:

[0081] The monitoring and guidance module reads the risk trend inflection time index and optimal gait parameter combination output by the strategy construction module, including cadence (e.g., 95 times / minute), stride length (e.g., 50 cm), and stance phase ratio (e.g., stance phase accounts for 65% of the cycle). Based on the cadence and stride length values, the system automatically calculates the single-cycle duration and the desired movement rhythm to generate a rhythm guidance signal. Simultaneously, a gait switching reminder mechanism is sent to the monitoring terminal based on the risk trend inflection time index. For example, when the patient's current gait has continuously exceeded the strategy's recommended range and the time is close to the identified inflection point (e.g., ±5 seconds), a "movement time too long" prompt is proactively issued, with the prompt content based on the real-time deviation between the actual sampled gait parameters and the strategy gait parameters.

[0082] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0083] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. 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. 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 wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0084] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0085] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0087] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0089] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0090] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0091] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A motion monitoring and management system for patients with diabetic foot, characterized in that, It includes a pressure acquisition module, a pressure analysis module, an electrical impedance acquisition module, a matrix construction module, a hysteresis extraction module, a risk analysis module, a strategy construction module, and a monitoring and guidance module, among which: The pressure acquisition module collects the plantar pressure distribution of diabetic foot patients during different gait cycles and extracts the main pressure-bearing areas of the plantar surface. The pressure analysis module extracts the force vector direction of the main pressure area of ​​the sole within each movement cycle and constructs the temporal evolution trajectory of the force direction of the sole. The impedance acquisition module constructs an impedance response sequence by setting an impedance acquisition unit and sampling impedance in the main pressure area of ​​the sole of the foot at a preset multiple of the pressure acquisition frequency. The matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction and the electrical impedance response sequence to generate a two-channel two-dimensional time-series matrix with spatial location as the index and time as the axis. The hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the time window of pressure release. The risk analysis module analyzes the spatial overlap between the current impedance recovery hysteresis region and the main pressure region of the foot in the next motion cycle; The strategy construction module identifies the minimum point of the overlap index in the spatially overlapping area as the risk trend inflection point, and extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy. The monitoring and guidance module dynamically drives the monitoring terminal to execute exercise guidance based on the output of the strategy construction module. The hysteresis extraction module extracts the hysteresis region of impedance recovery based on the impedance response sequence within the pressure release time window, specifically including: For each matrix unit in the two-dimensional time series matrix, the pressure release point is taken as the peak inflection point of the pressure value in the pressure time series channel. The time window in which the pressure value at the pressure release point drops to the preset baseline interval is traced back, and the corresponding impedance response value of the impedance time series channel is extracted within the time window. Calculate the local recovery rate for the electrical impedance value within the spatial grid coordinates corresponding to each matrix element; Set a threshold for impedance recovery rate, and perform cumulative area measurement on matrix units that do not reach the recovery rate to extract the hysteresis region of impedance recovery.

2. The exercise monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The pressure acquisition module collects the plantar pressure distribution of diabetic foot patients during different gait cycles, and extracts the main pressure-bearing areas of the plantar surface, specifically including: Multiple motion cycles are preset, and pressure is sampled from the patient's soles through a pressure acquisition unit. The pressure sampling values ​​within all motion cycles are then divided according to gait parameters. The gait parameters include cadence, stride length, and the proportion of support phase duration, with each movement cycle corresponding to a set of gait parameters; The main pressure-bearing regions are merged using a pressure intensity clustering method, and the main pressure-bearing regions with stable structures under the same synchronous parameters are extracted through spatial connectivity screening.

3. The exercise monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The pressure analysis module extracts the force vector direction of the main pressure area on the sole of the foot within each movement cycle, and constructs the temporal evolution trajectory of the force direction on the sole, specifically including: Within each major pressure zone of the foot, the continuous displacement direction of the pressure centroid vector is calculated using the acquisition time as an index; Trajectory smoothing and trend fitting operations are performed on the vector sequence within each motion cycle to generate the temporal evolution trajectory of the force direction on the sole of the foot. The evolution trajectory is represented in the form of spatial grid coordinates. The spatial grid size is the coverage area of ​​a single pressure acquisition unit; The trajectory smoothing method uses spatial vector aggregation to suppress local fluctuations and filter out local disturbance vectors.

4. The exercise monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The matrix construction module jointly encodes the temporal evolution trajectory of the plantar force direction with the electrical impedance response sequence to generate a dual-channel two-dimensional time-series matrix with spatial location as the index and time as the axis. Specifically, it includes: Using the spatial coordinates of the evolution trajectory of the plantar force direction as the index dimension, pressure time series channels and electrical impedance time series channels are constructed based on pressure value and electrical impedance response value in each motion cycle; A two-dimensional time series matrix is ​​established based on a dual-channel coding structure, where each spatial grid coordinate corresponds to a matrix unit, and the matrix unit contains time series channel values ​​in two dimensions.

5. A motion monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The risk analysis module analyzes the spatial overlap between the current impedance recovery hysteresis region and the main pressure region of the foot in the next motion cycle, specifically including: After obtaining the hysteresis region of impedance recovery, extract the set of pressure peak inflection points in the pressure time series channel of the matrix unit in the next motion cycle, and label the pressure peak region in the form of grid coordinates. Spatially register the pressure peak inflection point region of the next motion cycle with the impedance recovery lag region of the current cycle, calculate the percentage of their overlapping area, and output the spatial overlap region coincidence index.

6. A motion monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The strategy construction module identifies the minimum point of the overlap index in spatially overlapping areas as the inflection point of risk trends, and extracts the average overlap index under different time parameters to establish an individual motion monitoring strategy, specifically including: Based on the overlap index of spatial overlapping areas throughout the entire motion cycle, the minimum value of the overlap index is selected as the risk trend inflection point. Based on gait parameters, a set of motion cycles is divided, and the average overlap index of motion cycles corresponding to different gait parameters is statistically analyzed to conduct spatial risk assessment in the gait dimension. The time index of the risk trend inflection point in the two-dimensional time series matrix is ​​extracted to determine the moment of risk trend inflection. At the same time, the gait parameter with the minimum average overlap index in all motion cycles is extracted. The two are combined as the parameters of the patient's motion monitoring strategy and output to the monitoring guidance module.

7. A motion monitoring and management system for diabetic foot patients according to claim 1, characterized in that, The dynamic driving monitoring terminal performs motion guidance as follows: Based on the motion monitoring strategy parameters output by the strategy construction module, the monitoring terminal is dynamically driven to execute motion time, gait guidance, and early warning prompts.

Images