Pediatric penile development assessment method and system based on multi-modal sensor measurements

Abstract

The application belongs to the technical field of medical devices, and discloses a pediatric penis development evaluation method and system based on multi-modal sensor measurement, which comprises the following steps: calling hardware parameter group corresponding to the scene to configure current scene parameters, collecting and verifying environmental conditions, and then starting the measurement process; collecting double-frequency tissue impedance data of the pubic symphysis region, locking the measurement reference zero point, identifying the interface channel, and obtaining the fat pad thickness; starting constant force standardization traction, maintaining constant force creep after reaching the force threshold of the age group, latching the current displacement to generate the stretching length, collecting double-path homologous strain data at the same time, and obtaining the circumference value; checking the stretching length and the circumference value according to the preset physiological boundary, determining the measurement quality grade, obtaining the final development parameter, evaluating and grading the development based on the fat pad thickness, and generating the development evaluation conclusion; receiving the doctor's calibration instruction, updating the physiological boundary based on the development evaluation conclusion and the final development parameter.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and more specifically, to a method and system for assessing pediatric penile development based on multimodal sensor measurements. Background Technology

[0002] With the improvement of children's health management systems, the demand for early screening and intervention of penile developmental abnormalities (such as concealed penis and micropenis) in children is increasing. Delayed treatment of these diseases will seriously affect the physical and mental health of affected children. Currently, clinical practice still relies on manual measurement and assessment, which has shortcomings such as strong subjectivity and low standardization, making it difficult to meet the needs of large-scale children's health screening.

[0003] Most existing solutions suffer from low standardization in measurement, with manual operation relying entirely on physician experience, leading to discrepancies in measurement results between different doctors. Existing automated equipment fails to consider the creep and stress relaxation characteristics of pediatric penile soft tissue and is susceptible to interference from penile rigidity and ambient temperature, resulting in poor measurement accuracy and consistency. Furthermore, the assessment logic is flawed, relying solely on stretch length grading without considering fat pad thickness to differentiate concealed penis, easily misdiagnosing physiologically buried penises in obese children as true micropenis. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a method for assessing pediatric penile development based on multimodal sensor measurement, comprising: S1: Obtain the power-on command and scene selection command, call the hardware parameter group of the corresponding scene to configure the current scene parameters, collect and verify that the environmental conditions are qualified, start the measurement process, and generate a measurement ready status signal. S2: Based on the measurement ready state signal, collect dual-frequency tissue impedance data of the pubic symphysis region, determine the optimal contact surface through peak detection, lock the measurement reference zero point, and then identify the interface channel to obtain the fat pad thickness. S3: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is locked through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are collected, and the circumference value without interference is obtained through differential logic. S4: By using multidimensional physiological logic rules to verify the stretch length and circumference values ​​according to the preset physiological boundaries and determine the measurement quality level, the final development parameters are obtained. Then, based on the thickness of the fat pad, a development assessment conclusion is generated through the assessment of concealed penis and development grading. S5: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

[0005] Furthermore, the method for generating the measurement ready state signal includes: Based on the power-on command of the device and the scene selection command of the operator, the corresponding scene preset hardware parameter group is loaded to perform one-time parameter configuration of the multimodal monitoring unit; Simultaneously, ambient temperature and humidity data are collected, and environmental conditions are verified through preset temperature and humidity windows. If the environmental conditions are qualified, a measurement ready status signal is generated; if the environmental conditions are not qualified, the measurement function is locked.

[0006] Furthermore, the method for locking the measurement reference zero point includes: Based on the measurement ready state signal, the bony identification excitation frequency and tissue layering excitation frequency of each measurement point in the pubic symphysis region were collected as dual-frequency tissue impedance data. For any measurement point, a comparison is made using a preset bone tissue threshold. When the tissue impedance values ​​at two frequencies simultaneously meet the threshold, the measurement point is determined to be an effective bone channel. The total number of effective bony channels at each measurement location is counted to form an effective channel sequence that changes with the movement of the device. Peak detection is used to mark the location where the number of effective channels is at a local maximum and stable as the optimal contact surface of the pubic symphysis. Lock the current displacement measurement data at the optimal contact surface position as the measurement reference zero point for the entire measurement process.

[0007] Furthermore, the method for obtaining the fat pad thickness includes: After locking the measurement reference zero point, single-frequency depth impedance sequences of tissue layer excitation frequencies at each measurement point are collected in sequence from the skin surface to the deeper layers. Based on the difference in impedance characteristics between fat and muscle, the interface channel between fat and muscle is identified by using a preset tissue boundary threshold. The thickness of the subcutaneous fat pad in the pubic symphysis region is calculated by multiplying the difference in the serial number between the interface channel and the skin surface reference measurement channel by the electrode spacing, based on the pre-calibrated electrode spacing.

[0008] Furthermore, the method for generating the draw length includes: Based on the measurement reference zero point, a measurement start signal is generated and triggered to start clamping and execute segmented variable speed traction preloading and constant force standardized traction. Once the tensile force reaches the preset age-specific force threshold, it enters a constant force creep state, and the creep process is monitored in real time in stages. By using a dual judgment logic of strain rate and force stability, the creep equilibrium point is determined and locked, and the corresponding current displacement value is stored. Based on the stress relaxation characteristics of soft tissue, the current displacement value is corrected through stress relaxation compensation, and the inherent length compensation value of the clamping device is superimposed to generate the final standardized drawing length.

[0009] Furthermore, the circumference value is obtained in the following ways: Based on the measurement start signal, while preloading is being stretched, dual-source strain data from the measurement surface and the reference surface are simultaneously acquired and aligned by timestamp; Based on the principle of common-mode interference suppression, the intensity of common-mode interference is determined by differential logic layering, and the hardness and temperature in the strain data are layered to cancel out common-mode interference. Then, through stability and effectiveness verification, the interference-free circumference value is obtained.

[0010] Furthermore, the method for obtaining the final developmental parameters includes: The circumference value and the draw length are standardized and aligned to generate basic development parameters; Then, based on the pre-set multidimensional physiological logic rules, the basic developmental parameters are sequentially checked for age-group physiological boundaries, physiological logic consistency, and full-process validity backtracking. The measurement quality level is determined based on the verification results, and the final developmental parameters are generated from the valid measurement results.

[0011] Furthermore, the methods for generating developmental assessment conclusions include: Based on the final developmental parameters, the thickness of the fat pad is stratified into different thickness levels. Then, by combining the stretching length and the thickness of the fat pad, the pediatric penis is assessed for concealed penis and classified into confidence levels through a pre-set multi-condition stratification rule for concealed penis. At the same time, based on the stretch length, the penile development is divided into different developmental levels according to age group, and auxiliary developmental assessment is carried out in combination with the circumference to obtain the final basic developmental grading results; By integrating the results of concealed penis identification, confidence level, and basic developmental grading, a final developmental assessment conclusion is generated.

[0012] Furthermore, the methods for dynamically updating physiological boundaries include: Based on the developmental assessment conclusions and final developmental parameters, the system receives calibration instructions triggered by the doctor and gold standard measurement values. The final developmental parameters are used as valid measurement results and paired with the corresponding confidence markers and gold standard measurement values ​​to be stored in the calibration sample library. Based on the calibration sample library, the direction of measurement deviation is determined by age group according to high-confidence samples, and the physiological boundary threshold correction amount for each age group is dynamically updated and bounded by a fixed step size; then, through the continuous updating of the calibration sample library, a closed-loop iterative calibration is formed.

[0013] Furthermore, the pediatric penile development assessment system based on multimodal sensor measurements is characterized by comprising: Initialization and environment verification module: Obtains power-on command and scene selection command, calls the hardware parameter group of the corresponding scene to configure the current scene parameters, collects and verifies that the environmental conditions are qualified, starts the measurement process, and generates a measurement ready status signal. The baseline positioning and fat pad measurement module: Based on the measurement ready state signal, it collects dual-frequency tissue impedance data of the pubic symphysis region, determines the optimal contact surface through peak detection, locks the measurement baseline zero point, and then identifies the interface channel to obtain the fat pad thickness. Constant force stretching and parameter acquisition module: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is latched through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are acquired, and the circumference value without interference is obtained through differential logic. Physiological verification and development assessment module: The module verifies the stretch length and circumference values ​​according to the preset physiological boundaries through multidimensional physiological logic rules and determines the measurement quality level to obtain the final development parameters. Then, based on the thickness of the fat pad, it generates a development assessment conclusion through concealed penis assessment and development grading. Closed-loop calibration iteration module: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

[0014] The technical effects and advantages of this invention, which is a method and system for assessing pediatric penile development based on multimodal sensor measurement, are as follows: This invention focuses on the clinical screening scenario of pediatric penile development and constructs a fully automated system from standardized measurement and multi-dimensional verification to intelligent assessment. It solves the industry pain points of strong subjectivity and poor consistency of manual measurement and realizes the accuracy, standardization and intelligence of pediatric penile development assessment.

[0015] First, by using a three-stage variable speed preloading and dual-limit constant force traction technology, the standard for traction force in clinical measurements was unified; by combining the dual judgment logic of strain rate and force stability to lock the creep equilibrium point, and by using stress relaxation compensation to correct soft tissue deformation errors, the accuracy of stretching length measurement was improved.

[0016] Secondly, the use of dual-frequency tissue impedance technology enables automatic positioning of the bony benchmark of the pubic symphysis and accurate measurement of the thickness of the subcutaneous fat pad, solving the problem of inconsistent measurement benchmarks in traditional methods. Through the layered differential common-mode interference cancellation technology of dual-channel homogeneous strain sensors, hardness cross-interference and temperature drift are eliminated in one go.

[0017] Then, a three-level data quality control system was constructed, consisting of age-group physiological boundary verification, physiological logic consistency verification, and full-process validity backtracking verification. This system automatically filters valid measurement data and determines the measurement quality level. By combining the layered quantification of fat pad thickness with multi-condition progressive identification rules, the system can accurately distinguish between concealed penis and true micropenis.

[0018] Finally, by adopting a graded developmental assessment system combined with circumference-assisted assessment and dynamic analysis of developmental trends, structured assessment conclusions and stratified clinical recommendations with credibility levels were generated, providing clear diagnostic and treatment guidance for clinicians and reducing the rate of missed diagnoses and misdiagnoses.

[0019] This invention is simple to operate, fast to measure, and quick to measure a single case, which significantly reduces the workload of doctors and improves the cooperation of children. It is applicable to various scenarios such as newborn screening, pediatric health clinics, and specialist clinics, and provides reliable technical support for the early screening and intervention of penile developmental abnormalities in children. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the pediatric penile development assessment method based on multimodal sensor measurement according to the present invention; Figure 2 This is a schematic diagram of the dual-sensor hierarchical differential common-mode interference cancellation process in the pediatric penile development assessment method based on multimodal sensor measurement of the present invention; Figure 3 This is a schematic diagram of the pediatric penile development assessment system based on multimodal sensor measurement according to the present invention. Detailed Implementation

[0021] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1 Please see Figure 1 and Figure 2 As shown in this embodiment, the method for assessing pediatric penile development based on multimodal sensor measurement includes: S1: Obtain the power-on command and scene selection command, call the hardware parameter group of the corresponding scene to configure the current scene parameters, collect and verify that the environmental conditions are qualified, start the measurement process, and generate a measurement ready status signal. S2: Based on the measurement ready state signal, collect dual-frequency tissue impedance data of the pubic symphysis region, determine the optimal contact surface through peak detection, lock the measurement reference zero point, and then identify the interface channel to obtain the fat pad thickness. S3: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is locked through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are collected, and the circumference value without interference is obtained through differential logic. S4: By using multidimensional physiological logic rules to verify the stretch length and circumference values ​​according to the preset physiological boundaries and determine the measurement quality level, the final development parameters are obtained. Then, based on the thickness of the fat pad, a development assessment conclusion is generated through the assessment of concealed penis and development grading. S5: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

[0023] Methods for generating measurement ready status signals include: After powering on the measuring device according to the power-on command, the operator can generate corresponding scenario selection commands by selecting the measurement scenario, which includes four scenarios: newborn screening, pediatric outpatient clinic, specialist outpatient clinic and home monitoring. Based on the scene selection command, load the corresponding scene preset hardware parameter group and perform one-time parameter configuration for the multimodal monitoring unit; The monitoring units involved in parameter configuration include a dual-frequency crystal oscillator excitation source (output power and switching timing between 1kHz and 50kHz), a force threshold comparator (reference voltage (corresponding to age-appropriate traction force: 30g for newborns, 40g for infants, 50g for school-aged children, etc.)), an RC timer group (reference resistance value (corresponding to the default duration of constant force maintenance, effective acquisition, etc.)), a verification rule comparator (thresholds such as the ratio range of force to displacement, the ratio range of length to circumference, and the upper limit of the proportion of abnormal points) and a reference positioning module (threshold for determining the number of effective channels). While configuring parameters, ambient temperature and humidity data are collected, and environmental conditions are verified through preset temperature windows (default setting is the optimal temperature of 18℃-28℃) and humidity windows (default setting is the optimal humidity of 30%RH-70%RH). If the current ambient temperature and humidity are both within the corresponding window range, the current environmental conditions are considered acceptable; otherwise, they are considered unacceptable. If the environmental conditions are suitable, a measurement ready status signal is generated, the power supply and signal path of all subsequent measurement hardware are turned on, and the indicator light showing that the device is ready is lit. If the current environmental conditions are not up to standard, the measurement function is locked, the hardware cuts off the power supply to all measurement units, and controls the indicator lights that show the abnormal environment to flash until the environmental parameters return to the acceptable range, and then the measurement ready status signal is regenerated.

[0024] Methods for locking the measurement reference zero point include: Based on the measurement ready status signal, the operator smoothly places the measuring device base against the child's pubic symphysis area and moves it slowly. The device simultaneously collects tissue impedance data at each measurement point in the pubic symphysis area at two frequencies: the bony identification excitation frequency (set to 1kHz, low frequencies are most sensitive to the impedance difference between the periosteum and the cortex, used to accurately locate the bony reference plane of the pubic symphysis) and the tissue layering excitation frequency (default set to 50kHz, the frequency range of about 50kHz is most sensitive to the impedance difference between fat and muscle, used to accurately distinguish subcutaneous fat and muscle layers and calculate fat pad thickness). This forms dual-frequency tissue impedance data. Based on the differences in impedance characteristics of different bony tissues (the periosteum and cortical bone have the lowest impedance, while adipose tissue has the highest), the dual-frequency tissue impedance data of each measurement point is compared with the preset bone tissue thresholds (1kHz<400Ω, 50kHz<300Ω). When the tissue impedance values ​​of the two frequencies at any measurement point simultaneously meet the thresholds, the measurement point is determined to be an effective bony channel. The total number of effective bony channels at the current measurement position is counted in real time, forming a sequence of effective channels that changes with the movement of the device. Peak detection is performed on the effective channel sequence. When the number of effective channels is detected to be a local maximum and remains stable for 1 second, the current position is determined to be the optimal contact surface of the pubic symphysis bony surface, i.e., the optimal position. At the optimal contact surface position, immediately lock the current displacement measurement data and use it as the measurement reference zero point for the entire measurement process; After the baseline is locked, the number of effective channels in the bony region is continuously monitored. If the number of effective channels decreases by more than 20% compared to when the baseline was locked, the baseline is determined to have drifted, and the subsequent measurement process is immediately terminated and a repositioning prompt is given.

[0025] Methods for obtaining fat pad thickness include: Each measurement point is pre-numbered in order from the skin surface to the deeper layers, starting from channel 1. For example, channel 1 = skin surface reference channel (depth defined as 0mm), channel 2 = 0.5mm subcutaneously, channel 3 = 1mm subcutaneously, ..., the last channel = the deepest subcutaneous layer; After the measurement reference zero point is locked, the single-frequency depth impedance values ​​of the tissue layer excitation frequency at each measurement point are collected in a fixed order from the skin surface to the deep layer, and integrated to form a single-frequency depth impedance sequence. Based on the difference in impedance characteristics between fat and muscle (fat impedance > 600Ω, muscle impedance < 600Ω), starting from the first channel on the skin side, the single-frequency depth impedance value is compared with the preset tissue boundary threshold channel by channel. The first channel whose impedance value jumps from greater than 600Ω to less than 600Ω is identified and marked as the interface channel between fat and muscle. Based on the pre-calibrated electrode spacing (the physical distance between the center points of two adjacent measuring electrodes fixed at the factory), when the interface is identified in channel N, the number of channels actually covered by the fat layer is N-1 (because channel 1 is on the skin surface and is not included in the fat thickness). Therefore, the difference between the sequence number of the interface channel and the reference measuring channel on the skin surface is multiplied by the electrode spacing to calculate the thickness of the subcutaneous fat pad in the pubic symphysis region, that is, fat pad thickness = (N-1) × electrode spacing. The calculated fat pad thickness is checked for range. If the value exceeds the physiologically reasonable range of 0.5mm-20mm, it is judged as an abnormal measurement, the subsequent process is terminated and a repositioning prompt is given.

[0026] Methods for generating draw length include: Based on the measurement reference zero point start clamping, after clamping starts, segmented variable speed traction preloading and constant force standardized traction are performed; Specifically, a three-stage speed change is preloaded first, and then a standardized constant force is used for traction at each speed. In the first stage (rapid approach), the traction is carried out at a rate of 2 mm / s until the force sensor detects that the initial contact force is 5g. The second stage (medium-speed loading) switches to a pulling rate of 0.5mm / s until the force value reaches 0.8 times the age-specific force threshold (default setting is 30g for 0 to 1 month, 40g for 1 month to 3 years, and 50g for 3 to 14 years). The third stage (slow approximation) switches to a stretching rate of 0.1 mm / s until the force value reaches the target value of the force threshold for each age group. Once the traction force reaches the preset age-specific force threshold, it enters a constant force creep maintenance state. By monitoring the creep process in real time, the strain rate of the pediatric penile soft tissue (i.e., the degree of soft tissue deformation per unit time, referring to the speed at which the pediatric penile soft tissue deforms when subjected to force) is calculated, and the creep process is divided into three stages. Among them, strain rate > 0.01 mm / s is the initial creep stage, 0.001 mm / s ≤ strain rate ≤ 0.01 mm / s is the steady-state creep stage, and strain rate < 0.001 mm / s is the creep equilibrium stage; Using a dual judgment logic of strain rate and force stability, when the strain rate is less than 0.001 mm / s for three consecutive sampling periods (the sampling frequency is set to 100 Hz by default), and when the value fluctuation range of the tensile force is less than ±0.1 g for one second, it is determined that the creep equilibrium point has been reached, and the current displacement value corresponding to the current creep equilibrium point is latched. Based on the stress relaxation characteristics of soft tissue (after soft tissue is deformed by force, its internal stress gradually decreases over time while the deformation remains constant, resulting in a smaller locked displacement value), compensation correction is required after locking the displacement. Therefore, stress relaxation compensation is needed to correct the current displacement value. The calculation formula is as follows: in, This is the compensation displacement for stress relaxation. The tissue relaxation coefficient is a dimensionless tissue characteristic parameter used to reflect the amount of tissue relaxation displacement per unit force drop and per unit time. It is preset for different age groups; the younger the age, the better the soft tissue elasticity and the larger the relaxation coefficient. The average traction force value within the first second after the latching displacement (because the stress relaxation of the soft tissue of the pediatric penis mainly occurs within the first second after creep equilibrium, and the traction force value decreases very little after 1 second, which has no significant impact on the measurement results). For 1 second, The target values ​​for force thresholds by age group, The creep time constant is and These are all preset constants for different age groups, and can also be obtained based on biomechanical experiments of pediatric penile soft tissue. Alternatively, during the steady-state creep stage, the creep strain curve at the corresponding moment can be fitted in real time using the least squares method to obtain the value for newborns aged 0 to 1 month. The default recommended value is 0.002. The typical recommended time is 2 seconds for infants aged 1 month to 3 years. It is 0.0015. For 1.5 seconds, school-aged children aged 3 to 14 It is 0.001. It takes 1 second; Add the current displacement to the compensated displacement to obtain the final displacement; Then, the final displacement is subtracted from the measurement reference zero point and the inherent length compensation value of the clamping end (factory calibration constant) to generate the final standardized drawing length.

[0027] The circumference value can be obtained in the following ways: Based on the measurement start signal, while preloading is being stretched, two sets of flexible strain sensor data with completely identical characteristics are simultaneously acquired, including measurement surface sensor data and reference surface sensor data, as dual-path co-source strain data of the measurement surface and reference surface. Among them, the measurement surface sensor data is a mixed signal that is in direct contact with the tissue being measured, and includes the actual circumference change, hardness change and temperature drift. The reference surface sensor data is attached to the backing layer and does not contact the tissue being measured. It only contains common-mode interference signals of hardness changes and temperature shifts. Then, the raw data from both the measurement surface and the reference surface are preprocessed by sliding window outlier removal: the sliding window size is fixed at 5 consecutive sampling points (one sampling point per sampling period), and the window step size is 1; the average value of the other 4 points in the window is calculated. If the deviation of the current point value from the average value exceeds ±10%, it is determined to be an isolated outlier, and the outlier is replaced by the average value of the above 4 points. If the number of outliers in a single window is ≥2, mark the window as an outlier data segment. Traverse all marked abnormal data segments and perform hierarchical processing according to their length: For samples with a length of ≤3 sampling points: replace the abnormal segment with linear interpolation using the two nearest normal data points before and after the abnormal segment; For data with 4 to 10 sampling points in length and located within the first 2 seconds of measurement: discard this data segment directly and continue processing from the 3rd second onwards; For samples with a length of 4 to 10 sampling points and in a stable measurement phase: issue an early warning signal (local data anomaly), retain the data segment and mark it; If the length exceeds 10 sampling points: the sensor is considered to have poor contact, and the measurement is terminated. After preprocessing, the total length of the valid data segments is calculated: If the total length of the valid data segment is less than 5 seconds, it is determined that the data volume is insufficient and the measurement is terminated. If the total length of the valid data segment is ≥5 seconds, extract the last 3 seconds of stable data segment and proceed to the next process; The two sets of data in the preprocessed stable data segment are aligned with the same timestamp to form a one-to-one measurement and reference data pair. Based on the principle of common-mode interference suppression, the intensity of common-mode interference is determined by differential logic layering, and the hardness and temperature in the strain data are layered to cancel out common-mode interference. Specifically: For the formed data pairs, calculate the real-time difference between the measured surface data and the reference surface data, and determine the common-mode interference intensity according to the size of the difference: a difference ≤ 0.1mm is weak interference, 0.1mm < difference ≤ 0.3mm is medium interference, and a difference > 0.3mm is strong interference; The percentage of each interference level in the data is statistically analyzed. If the percentage of strong interference exceeds 30%, it is determined that the environmental electromagnetic interference is too large. Based on the highest level of interference, differentiated hierarchical common-mode interference cancellation processing is performed: Since the effects of hardness changes and temperature shifts on the two sensors are exactly the same, only the actual circumference change will produce a difference between the two sensors. Therefore, for weak interference, basic differential operations can be performed directly, that is, the reference surface data is subtracted from the measured surface data, eliminating the two common-mode errors of hardness cross interference and temperature drift in one go, and obtaining pure data containing only the true circumference change. For interference, first perform a 3-point moving average filter on the two data streams (each point takes the average of the three adjacent points before and after), and then perform basic difference operations. For strong interference, first perform 5-point moving average filtering on the two data streams respectively, and then perform basic difference operation; Stability verification is performed on the differential circumference data: the last 20 consecutive sampling points of the stable data segment are taken, and the difference between the maximum and minimum values ​​of the data set is calculated. If the difference is ≤0.1mm, the data is considered stable. If the difference is >0.1mm, the measurement time is automatically extended by 2 seconds, and the last 20 sampling points are extracted again for verification. If the stability requirement is still not met after 3 consecutive extended measurements, the data is considered unstable, and the measurement is terminated. Then, an effectiveness verification is performed. Take 10 consecutive sampling points and determine the direction of change (increasing, decreasing, or unchanged) of the two data at each sampling point. If the direction of change of at least 9 sampling points is completely consistent, it is judged as good consistency; if the direction of change of 5 to 8 sampling points is consistent, it is judged as sensor consistency deviation. If fewer than 5 sampling points show the same direction of change, it is determined that the sensor is connected in reverse, and the measurement is terminated. After the calibration is passed, the obtained circumference change is added to the zero-point circumference value calibrated by the sensor at the factory to obtain the final circumference value without hardness interference and temperature drift.

[0028] The methods for obtaining the final developmental parameters include: The circumference value and draft length without interference are uniformly standardized and aligned. The inherent length compensation value of the clamping device is subtracted from the draft length, and the zero-position circumference value of the sensor is subtracted from the circumference value without interference to obtain the basic development parameters (including the aligned draft length and circumference value). Then, based on the pre-set multidimensional physiological logic rules, the basic developmental parameters are sequentially checked for age-group physiological boundaries, physiological logic consistency, and full-process validity backtracking. Specifically, the verification logic for the multidimensional physiological logic rules is as follows: First, perform physiological boundary rule verification by age group: Preset physiological boundaries for different age groups. According to the publicly available expert consensus on the assessment of penile development in Chinese children, the initial default physiological boundary for penile stretch length is 20 to 38 mm for children aged 0 to 1 month and 12 to 22 mm for circumference. For children aged 1 month to 3 years, the initial stretch length is 28 to 55 mm and the initial circumference is 15 to 28 mm. For children aged 3 to 14 years, the initial stretch length is 38 to 110 mm and the initial circumference is 18 to 45 mm. Then, three-layer boundary checks were performed on the draft length and circumference values ​​respectively: In the first layer, if the value exceeds the upper or lower limit of the physiological boundary of the corresponding age group, it is directly marked as an invalid measurement; The second level is marked as Level 1 suspicious if the value is within the critical range of ±5% of the physiological boundary. The third layer, if the value is completely within a reasonable range, proceeds to the next step of physiological logic consistency verification. If the boundary verification results of two parameter values ​​are inconsistent, cross-judgment is performed: if one parameter is absolutely invalid and the other is normal, the overall measurement is determined to be invalid; if one parameter is critically suspicious and the other is normal, a level 1 suspicious mark is retained; if both parameters are level 1 suspicious, a total of 2 level 1 suspicious marks are accumulated. Physiological consistency verification is based on clinically recognized physiological patterns of pediatric penile development and involves three consistency rules: Rule 1: The draw length must be greater than half of the circumference; Rule 2: The ratio of circumference to draft length must be between 0.4 and 0.8; Rule 3: If the fat pad thickness is greater than 5mm, the stretching length shall not be less than 80% of the lower limit for the same age group; Rule 4: If the stretch length is greater than or equal to 120% of the upper limit of the corresponding age group, then the circumference value shall not be less than the lower limit of the corresponding age group. Rule 5: If there is a previous valid measurement for the same patient, the difference between the current stretch length and the previous measurement shall not exceed 2 mm; If the patient's age is <6 months, the difference threshold of Rule 5 is relaxed to 3 mm; if the fat pad thickness is ≥10 mm, the lower limit threshold of Rule 3 is relaxed to 70% of the lower limit of the corresponding age group. If any rule is not met, it is marked as a Level 2 suspicious; if all rules are met, proceed to the next step of full-process validity backtracking verification. All marked suspicious items are weighted and statistically analyzed, with each item of Level 1 suspicion receiving a weight of 1 point and each item of Level 2 suspicion receiving a weight of 2 points. The preliminary measurement quality level is determined based on the cumulative weighted total score. If the cumulative weight = 0 points, it is judged as preliminary valid. If the cumulative weight = 1 to 2 points, it is judged as a questionable measurement. If the cumulative weight ≥ 3 points, it is an invalid measurement. The end-to-end measurement validity backtracking verification is a process that connects the validity indicators of all preceding steps and performs a full-process logical verification: If the ambient temperature calibration fails, the measurement reference zero point lock fails, the stretching process is invalid, or the dual sensor calibration fails, it is directly judged as an invalid measurement; if the fat pad thickness measurement fails, it is marked as fat pad data missing, and the concealed penis identification cannot be performed, but it does not affect the basic development assessment. If all preliminary steps are deemed valid, the measurement is ultimately determined to be valid. Based on the verification results, the final measurement quality level is determined and priorities are assigned: Priority 1: If any invalid measurement marker exists, the entire measurement is deemed invalid. Priority 2: If there are no invalid markers but there is any suspicious measurement marker, the whole thing is judged as suspicious; Priority 3: All checks pass, and the overall result is deemed valid; The final valid measurement results are used as the final development parameters, including the final draw length and circumference parameters; for doubtful and invalid results, corresponding abnormality cause indicators are generated. If multiple valid measurements are taken from the same patient within 30 minutes, a consistency check is performed. The average of all valid stretch length measurements is calculated, and abnormal measurements with a deviation of more than ±5% from the average are removed. The average of the remaining measurements is taken as the final baseline developmental parameter.

[0029] The methods for generating developmental assessment conclusions include: Based on the final developmental parameters, a layering determination with boundary hysteresis is performed (a 0.2mm buffer zone is reserved at the upper and lower boundaries to avoid frequent switching caused by boundary fluctuations), and the fat pad thickness is quantified into different thickness levels. Specifically, it was divided into four layers: normal fat layer (thickness < 3.0 mm, identification weight 0), mild fat accumulation (3.0 mm ≤ thickness < 5.0 mm, identification weight 1), moderate fat accumulation (5.0 mm ≤ thickness < 10.0 mm, identification weight 2), and severe fat accumulation (thickness ≥ 10.0 mm, identification weight 3). Special processing is performed on fat pad thickness data that exceeds the physiological range: If the thickness of the fat pad is <1.0mm, it is considered to be abnormally low and should be treated as a normal fat layer. If the thickness of the fat pad is >20.0mm, it is judged as an abnormally high fat pad thickness and should be treated as severe fat accumulation. Furthermore, by combining the stretching length and fat pad thickness, the pediatric penis was assessed for concealment and its reliability was classified using a pre-defined multi-condition stratification rule for concealed penis. Specifically, the logic behind the multi-condition stratification rule for concealed penis is as follows: If both severe fat accumulation and stretching length are ≤80% of the normal lower limit for the corresponding age group, it is judged as highly suspected of being a concealed penis, with an initial confidence level of medium. If both moderate fat accumulation and stretching length are ≤80% of the lower limit of normal for the corresponding age group, it is judged as moderate suspected concealed penis, with an initial confidence level of low. If the patient simultaneously meets the criteria of mild fat accumulation and stretching length ≤ 70% of the normal lower limit for the corresponding age group, the patient is judged to have mild suspected concealed penis, with an initial confidence level of low. If both severe fat accumulation and stretching length are ≤70% of the normal lower limit for the corresponding age group, the highly suspected concealed penis will be upgraded to extremely highly suspected concealed penis, and it is recommended to seek immediate specialist consultation. If the patient has a normal fat layer and the stretching length is ≤70% of the lower limit of normal for the corresponding age group, the patient is suspected of having micropenis. Further examination is recommended to rule out fat burial factors. If any of the above conditions are not met, it is determined to be a non-concealed penis; If the patient is less than 6 months old, all conclusions of concealed penis identification will be automatically marked with the possibility of physiological burial during the neonatal period, and a follow-up examination is recommended after 3 months of age; If the patient is older than 14 years, the concealed penis identification will not be performed, and the patient will be considered to be outside the age range for identification. At the same time, based on the stretch length, penile development is divided into four different developmental levels according to age group; The age groups are divided into newborns (0-1 month), infants (1 month to 3 years), preschool children (3 to 6 years), school-age children (6 to 10 years), and pre-adolescent children (10 to 14 years). For each age group, three fixed thresholds are preset: the lower limit of normal stretching length, the median, and the median plus one standard deviation. The standard deviation refers to the standard deviation of the stretching length data within the corresponding age group. Including: well-developed (stretch length ≥ median of the corresponding age group + 1 standard deviation), normal-developed (median of the corresponding age group ≤ stretch length < median of the corresponding age group + 1 standard deviation), short-developed (lower limit of normal for the corresponding age group ≤ stretch length < median of the corresponding age group), and underdeveloped (stretch length < lower limit of normal for the corresponding age group). Based on the developmental grade, auxiliary developmental assessment is conducted in conjunction with circumference to obtain the final basic developmental grade result. Specifically: if the stretch length is normal but the circumference value is less than the lower limit of the corresponding age group, the circumference is marked as being too thin. If the stretch length is shorter than expected but the circumference is greater than or equal to the median of the corresponding age group, add a note indicating that the circumference development is normal. If both the stretch length and circumference are less than the lower limit of the corresponding age group, an additional note should be added indicating overall developmental delay. If the stretch length is well developed but the circumference value is greater than or equal to the upper limit of the corresponding age group, add a note that the circumference is too large. If valid historical measurement data are available for the same patient at an interval of ≥3 months, continue with the auxiliary developmental assessment using developmental trend analysis: Calculate the time interval between the two measurements (calculated in months) and the increase in stretch length. If the monthly increase is less than 50% of the normal monthly increase for the corresponding age group, add a note indicating developmental delay. If the monthly growth value is greater than 150% of the normal monthly growth value for the corresponding age group, an additional marker of accelerated development is added. If the monthly growth value is within the normal range, add a note indicating a normal development trend; Finally, combining the measurement quality level and identification conditions, the final reliability level calibration is performed: If the measurement is valid and meets the criteria for extremely high or highly suspected concealed penis, or if the developmental grade is within the normal range, the final calibration is high confidence: If the measurement is valid and meets the criteria for moderate or mild suspected concealed penis, or if the developmental grade is in the critical range, the final calibration will be medium confidence. If all evaluation conclusions generated from questionable or partially valid measurements are ultimately calibrated to low confidence; By integrating the results of concealed penis identification (e.g., moderately suspected concealed penis), confidence level, basic developmental grading results (e.g., 3-year-old boy with shorter penis development) and auxiliary assessments (e.g., normal circumference development and normal developmental trend), a final developmental assessment conclusion is generated.

[0030] Methods for dynamically updating physiological boundaries include: Based on the developmental assessment conclusions and the effective measurement results of the final developmental parameters, the update and iteration process is only initiated when the clinician actively triggers the one-click calibration command, thereby receiving the calibration command triggered by the doctor and the gold standard measurement value; among them, the gold standard measurement value refers to the penile stretch length and circumference measurement value independently completed by the clinician using the standardized manual measurement method recommended by the Chinese Medical Association Pediatrics Branch, which is currently the recognized reference benchmark in the field of pediatric penile development assessment. The valid measurement results, the corresponding confidence markers in the developmental assessment conclusions, and the gold standard measurement values ​​are paired and stored in the calibration sample library. The calibration sample library has a maximum retention limit for valid paired data. If the number of data to be stored exceeds the limit, the earliest pair is deleted. The number of high-confidence samples is counted by age group. When a certain age group accumulates 10 valid high-confidence samples, a threshold update is triggered. Based on the calibration sample library, if more than 80% of the effective measurements of stretch length and circumference of all samples in this age group are greater than the gold standard measurement, it is determined that there is a positive deviation direction. If more than 80% of the values ​​are less than the gold standard measurement, it is determined that there is a negative deviation direction. If neither of the above two conditions is met, the deviation direction is confirmed to be dispersed, and the deviation direction is determined to be unbiased. Dynamic updates will not be performed, and the sample count for that age group will be cleared and re-accumulated. Only when a deviation direction is determined, the physiological boundary threshold correction amount for each age group is dynamically updated and bounded with a fixed step size. Specifically, if the deviation is positive (which indicates that the current effective measurement values ​​are generally too large), the correction amount for the upper and lower limits of the stretching length for this age group will be reduced by 0.1 mm, and the correction amount for the upper and lower limits of the circumference will be reduced by 0.05 mm. If the deviation is negative (which indicates that the effective measured values ​​are generally too small), the correction amount for the upper and lower limits of the stretching length for this age group should be increased by 0.1 mm, and the correction amount for the upper and lower limits of the circumference should be increased by 0.05 mm. The cumulative correction record for this age group will be updated synchronously. In addition, absolute correction boundary limits are set to prevent over-correction: the cumulative correction amount of the upper and lower limits of the stretching length for a single age group shall not exceed ±0.5mm; the cumulative correction amount of the upper and lower limits of the circumference for a single age group shall not exceed ±0.3mm. If the threshold is exceeded, the dynamic update for that age group will be terminated, and a notification will be sent indicating that the threshold has reached the maximum adjustment range. In the subsequent age-group physiological boundary verification process, the dynamically updated new boundaries are used as the actual verification thresholds. By continuously updating the calibration sample library, a complete closed-loop iterative calibration is formed, consisting of measurement, correction, verification, calibration sample library updates, and further correction.

[0031] Example 2 Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A pediatric penile development assessment system based on multimodal sensor measurements is provided, including: Initialization and environment verification module: Obtains power-on command and scene selection command, calls the hardware parameter group of the corresponding scene to configure the current scene parameters, collects and verifies that the environmental conditions are qualified, starts the measurement process, and generates a measurement ready status signal. The baseline positioning and fat pad measurement module: Based on the measurement ready state signal, it collects dual-frequency tissue impedance data of the pubic symphysis region, determines the optimal contact surface through peak detection, locks the measurement baseline zero point, and then identifies the interface channel to obtain the fat pad thickness. Constant force stretching and parameter acquisition module: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is latched through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are acquired, and the circumference value without interference is obtained through differential logic. Physiological verification and development assessment module: The module verifies the stretch length and circumference values ​​according to the preset physiological boundaries through multidimensional physiological logic rules and determines the measurement quality level to obtain the final development parameters. Then, based on the thickness of the fat pad, it generates a development assessment conclusion through concealed penis assessment and development grading. Closed-loop calibration iteration module: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

[0032] Example 3 This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the pediatric penile development assessment system based on multimodal sensor measurement described above.

[0033] Since the electronic device described in this embodiment is the electronic device used to implement the pediatric penile development assessment method based on multimodal sensor measurement in the embodiments of this application, those skilled in the art can understand the specific implementation and various variations of the electronic device in this embodiment based on the pediatric penile development assessment method based on multimodal sensor measurement described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art to implement the pediatric penile development assessment method based on multimodal sensor measurement in the embodiments of this application falls within the scope of protection of this application.

[0034] 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.

[0035] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for assessing pediatric penile development based on multimodal sensor measurements, characterized in that, include: S1: Obtain the power-on command and scene selection command, call the hardware parameter group of the corresponding scene to configure the current scene parameters, collect and verify that the environmental conditions are qualified, start the measurement process, and generate a measurement ready status signal. S2: Based on the measurement ready state signal, collect dual-frequency tissue impedance data of the pubic symphysis region, determine the optimal contact surface through peak detection, lock the measurement reference zero point, and then identify the interface channel to obtain the fat pad thickness. S3: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is locked through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are collected, and the circumference value without interference is obtained through differential logic. S4: By using multidimensional physiological logic rules to verify the stretch length and circumference values ​​according to the preset physiological boundaries and determine the measurement quality level, the final development parameters are obtained. Then, based on the thickness of the fat pad, a development assessment conclusion is generated through the assessment of concealed penis and development grading. S5: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

2. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 1, characterized in that, The methods for generating the measurement ready state signal include: Based on the power-on command of the device and the scene selection command of the operator, the corresponding scene preset hardware parameter group is loaded to perform one-time parameter configuration of the multimodal monitoring unit; Simultaneously, ambient temperature and humidity data are collected, and environmental conditions are verified through preset temperature and humidity windows. If the environmental conditions are qualified, a measurement ready status signal is generated; if the environmental conditions are not qualified, the measurement function is locked.

3. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 2, characterized in that, The methods for locking the measurement reference zero point include: Based on the measurement ready state signal, the bony identification excitation frequency and tissue layering excitation frequency of each measurement point in the pubic symphysis region were collected as dual-frequency tissue impedance data. For any measurement point, a comparison is made using a preset bone tissue threshold. When the tissue impedance values ​​at two frequencies simultaneously meet the threshold, the measurement point is determined to be an effective bone channel. The total number of effective bony channels at each measurement location is counted to form an effective channel sequence that changes with the movement of the device. Peak detection is used to mark the location where the number of effective channels is at a local maximum and stable as the optimal contact surface of the pubic symphysis. Lock the current displacement measurement data at the optimal contact surface position as the measurement reference zero point for the entire measurement process.

4. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 3, characterized in that, The methods for obtaining the thickness of the fat pad include: After locking the measurement reference zero point, single-frequency depth impedance sequences of tissue layer excitation frequencies at each measurement point are collected in sequence from the skin surface to the deeper layers. Based on the difference in impedance characteristics between fat and muscle, the interface channel between fat and muscle is identified by using a preset tissue boundary threshold. The thickness of the subcutaneous fat pad in the pubic symphysis region is calculated by multiplying the difference in the serial number between the interface channel and the skin surface reference measurement channel by the electrode spacing, based on the pre-calibrated electrode spacing.

5. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 3, characterized in that, The methods for generating the draw length include: Based on the measurement reference zero point, a measurement start signal is generated and triggered to start clamping and execute segmented variable speed traction preloading and constant force standardized traction. Once the tensile force reaches the preset age-specific force threshold, it enters a constant force creep state, and the creep process is monitored in real time in stages. By using a dual judgment logic of strain rate and force stability, the creep equilibrium point is determined and locked, and the corresponding current displacement value is stored. Based on the stress relaxation characteristics of soft tissue, the current displacement value is corrected through stress relaxation compensation, and the inherent length compensation value of the clamping device is superimposed to generate the final standardized drawing length.

6. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 5, characterized in that, The circumference value is obtained in the following ways: Based on the measurement start signal, while preloading is being stretched, dual-source strain data from the measurement surface and the reference surface are simultaneously acquired and aligned by timestamp; Based on the principle of common-mode interference suppression, the intensity of common-mode interference is determined by differential logic layering, and the hardness and temperature in the strain data are layered to cancel out common-mode interference. Then, through stability and effectiveness verification, the interference-free circumference value is obtained.

7. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 6, characterized in that, The methods for obtaining the final developmental parameters include: The circumference value and the draw length are standardized and aligned to generate basic development parameters; Then, based on the pre-set multidimensional physiological logic rules, the basic developmental parameters are sequentially checked for age-group physiological boundaries, physiological logic consistency, and full-process validity backtracking. The measurement quality level is determined based on the verification results, and the final developmental parameters are generated from the valid measurement results.

8. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 7, characterized in that, The methods for generating developmental assessment conclusions include: Based on the final developmental parameters, the thickness of the fat pad is stratified into different thickness levels. Then, by combining the stretching length and the thickness of the fat pad, the pediatric penis is assessed for concealed penis and classified into confidence levels through a pre-set multi-condition stratification rule for concealed penis. At the same time, based on the stretch length, the penile development is divided into different developmental levels according to age group, and auxiliary developmental assessment is carried out in combination with the circumference to obtain the final basic developmental grading results; By integrating the results of concealed penis identification, confidence level, and basic developmental grading, a final developmental assessment conclusion is generated.

9. The method for assessing pediatric penile development based on multimodal sensor measurement according to claim 8, characterized in that, The methods for dynamically updating physiological boundaries include: Based on the developmental assessment conclusions and final developmental parameters, the system receives calibration instructions triggered by the doctor and gold standard measurement values. The final developmental parameters are used as valid measurement results and paired with the corresponding confidence markers and gold standard measurement values ​​to be stored in the calibration sample library. Based on the calibration sample library, the direction of measurement deviation is determined by age group according to high-confidence samples, and the physiological boundary threshold correction amount for each age group is dynamically updated and bounded by a fixed step size; then, through the continuous updating of the calibration sample library, a closed-loop iterative calibration is formed.

10. A pediatric penile development assessment system based on multimodal sensor measurement, implemented based on the pediatric penile development assessment method based on multimodal sensor measurement as described in claims 1 to 9, characterized in that... include: Initialization and environment verification module: Obtains power-on command and scene selection command, calls the hardware parameter group of the corresponding scene to configure the current scene parameters, collects and verifies that the environmental conditions are qualified, starts the measurement process, and generates a measurement ready status signal. The baseline positioning and fat pad measurement module: Based on the measurement ready state signal, it collects dual-frequency tissue impedance data of the pubic symphysis region, determines the optimal contact surface through peak detection, locks the measurement baseline zero point, and then identifies the interface channel to obtain the fat pad thickness. Constant force stretching and parameter acquisition module: Based on the measurement reference zero point, constant force standardized stretching is started. After reaching the force threshold of each age group, constant force creep is maintained. The current displacement is latched through creep equilibrium determination to generate the stretching length. At the same time, dual-source strain data are acquired, and the circumference value without interference is obtained through differential logic. Physiological verification and development assessment module: The module verifies the stretch length and circumference values ​​according to the preset physiological boundaries through multidimensional physiological logic rules and determines the measurement quality level to obtain the final development parameters. Then, based on the thickness of the fat pad, it generates a development assessment conclusion through concealed penis assessment and development grading. Closed-loop calibration iteration module: Receives calibration instructions from doctors and dynamically updates physiological boundaries based on developmental assessment conclusions and final developmental parameters.

Images