Dynamic urine flow rate intelligent measuring system and method

Abstract

The invention relates to the technical field of physiological parameter monitoring, in particular to a dynamic urine flow rate intelligent measuring system and method. The method comprises the following steps: collecting identity data of a patient; when urine flows into the urine volume signal acquisition device with the weighing sensor, the output voltage of the weighing sensor is changed to form an original signal; carrying out physiological signal filtering on the original signal through a microprocessor; when the total urination amount is smaller than the threshold total amount, prompt information is output, and urination time is recorded; when the total urination amount is larger than or equal to the threshold total amount, urine data are calculated; and recording all urine data of the patient in a preset collection period, and binding the urine data with the patient identity data to form personalized urine data. According to the invention, through fusion of identity recognition, signal processing and data management, high precision, individualization, scene diversification and remote intelligent application of urine flow rate measurement are realized.

Description

Technical Field

[0001] This invention relates to the field of physiological parameter monitoring technology, and in particular to a dynamic urine flow rate intelligent measurement system and method. Background Technology

[0002] Lower urinary tract dysfunction (LUTD) is relatively common in clinical practice, and it relies on non-invasive uroflowmetry. Early uroflowmeters were mostly hospital-specific devices, primarily using weighing sensors or rotary mechanisms to convert urine flow rate into a physical signal. With the development of electronic measurement and computer technology, capacitive measurement and digital recording methods have emerged, capable of plotting uroflow curves and outputting parameters such as maximum and average uroflow rate, providing clinical reference.

[0003] Existing uroflowmetry technologies have certain characteristics and advantages; however, these devices are mostly limited to hospital settings, are bulky, and require patients to complete the measurement in a specific environment, lacking the naturalness and continuity of home settings.

[0004] Meanwhile, existing technologies still have the following shortcomings: First, the accuracy of a single measurement is insufficient. Studies have shown that the error of a single measurement of maximum urinary flow rate can reach 50%, requiring multiple measurements to reduce the error range. Second, traditional instruments lack threshold determination functions, and output results even when urine volume is insufficient, which can easily interfere with doctors' judgment. Third, hospital-use urinary flow rate meters generally lack the function of binding with patient databases, making it difficult to combine with electronic medical records to form personalized urine data. Fourth, portable home devices are not yet mature, and data from multiple measurements by patients cannot be effectively uploaded and integrated, and periodic urine reports cannot be generated. Summary of the Invention

[0005] Therefore, it is necessary for the present invention to provide a dynamic urinary flow rate intelligent measurement system and method to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above objectives, a dynamic intelligent method for measuring urinary flow rate is provided, applied to a urinary flow rate meter, the urinary flow rate meter including a urine volume signal acquisition device and a microprocessor, the method comprising:

[0007] Step S1: Collect patient identification data; Step S2: When urine flows into the urine volume signal acquisition device with a weighing sensor, the output voltage of the weighing sensor changes, forming the original signal; Step S3: The original signal is filtered by the microprocessor to remove interference pulses using threshold discrimination logic, and frequency control is performed using an IIR Butterworth low-pass filter. Step S4: When the total amount of urine output is less than the threshold amount, output a prompt message and record the urination time; when the total amount of urine output is greater than or equal to the threshold amount, calculate the urine data. Step S5: Within the preset collection period, repeat steps S2 to S4 to record all urine data of the patient within the preset collection period and bind it with the patient's identity data to form personalized urine data; Step S6: After the preset collection period ends, the personalized urine data is uploaded to the wireless network connection center server and analyzed to generate urine report data; Step S7: Visualize the urine report data and send it to the user's device.

[0008] Preferably, the present invention also provides a dynamic urinary flow rate intelligent measurement system for performing the above-described dynamic urinary flow rate intelligent measurement method, the dynamic urinary flow rate intelligent measurement system comprising: The patient identification data collection module is used to collect patient identification data; The urine signal acquisition module is used to generate a raw signal by changing the output voltage of the weighing sensor when urine flows into the urine volume signal acquisition device equipped with a weighing sensor. The signal filtering and processing module is used to filter the original signal using a microprocessor, remove interference pulses using threshold discrimination logic, and control the frequency using an IIR Butterworth low-pass filter. The threshold determination and processing module is used to output a prompt message and record the urination time when the total amount of urine is less than the threshold amount; and to calculate the urine data when the total amount of urine is greater than or equal to the threshold amount. The periodic data storage module is used to repeatedly execute steps S2 to S4 within a preset collection period, record all urine data of the patient within the preset collection period, and bind it with the patient's identity data to form personalized urine data. The data upload and analysis module is used to upload personalized urine data to the wireless network connection center server after the preset collection period ends, and analyze and generate urine report data. The report visualization and transmission module is used to visualize urine report data and send it to the user's device.

[0009] This invention significantly enhances the clinical value and user experience of uroflowmetry by organically combining identity recognition, signal acquisition, intelligent filtering, threshold determination, periodic data management, and wireless network transmission. First, this method significantly reduces the error risk associated with single measurements, resulting in higher stability and repeatability of voiding characteristic data, thereby improving the accuracy and reliability of the results. Second, by binding the data to patient identity data, it enables personalized management of urine data, ensuring long-term monitoring data remains consistent with medical records, facilitating continuous tracking and analysis of the patient's disease progression by physicians. Furthermore, this method is suitable for both hospital and home settings, overcoming the limitations of traditional devices confined to fixed locations. Patients can complete the test in a natural voiding environment, improving compliance and measurement accuracy. Through periodic data collection and uploading, a complete dynamic urine database can be formed in the cloud, generating standardized reports that support remote image review and follow-up management by physicians. Finally, the method achieves automation and intelligence throughout the entire data flow, avoiding information omissions or judgment biases caused by manual intervention, providing a reliable tool for efficacy evaluation and rehabilitation tracking of urinary system diseases. Attached Figure Description

[0010] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the steps of a dynamic urine flow rate intelligent measurement method according to the present invention; Figure 2 This is a schematic diagram of the workflow of the present invention; Figure 3 This is a schematic diagram of the technical route of the present invention; Figure 4 This is a schematic diagram of the system architecture of the present invention; Figure 5 This is a schematic diagram of the experimental design process of the present invention; Figure 6 This is a physical image of the urine collector of the present invention. Detailed Implementation

[0011] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0012] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0013] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0014] To achieve the above objectives, please refer to Figures 1 to 6 This invention provides a dynamic intelligent method for measuring urine flow rate, applied to a urine flow meter, wherein the urine flow meter includes a urine volume signal acquisition device and a microprocessor, and the method includes: Step S1: Collect patient identification data; Step S2: When urine flows into the urine volume signal acquisition device with a weighing sensor, the output voltage of the weighing sensor changes, forming the original signal; Step S3: The original signal is filtered by the microprocessor to remove interference pulses using threshold discrimination logic, and frequency control is performed using an IIR Butterworth low-pass filter. Step S4: When the total amount of urine output is less than the threshold amount, output a prompt message and record the urination time; when the total amount of urine output is greater than or equal to the threshold amount, calculate the urine data. Step S5: Within the preset collection period, repeat steps S2 to S4 to record all urine data of the patient within the preset collection period and bind it with the patient's identity data to form personalized urine data; Step S6: After the preset collection period ends, the personalized urine data is uploaded to the wireless network connection center server and analyzed to generate urine report data; Step S7: Visualize the urine report data and send it to the user's device.

[0015] Preferably, step S1 includes: Obtain the interactive operation instructions of the QR code interface of the uroflowmeter, parse the patient identification information embedded therein, and extract the patient identity data; Alternatively, facial recognition devices deployed at hospital user terminals can be used to collect patients' facial images in real time. The collected images are then normalized for illumination, aligned with faces, and enhanced for features. The pre-processed image features are then matched with the registered features in a pre-defined patient face database, and patient identity data is extracted based on the feature matching results.

[0016] In this embodiment of the invention, before the patient uses the uroflowmeter, the control terminal of the uroflowmeter loads a QR code interactive interface. This interface displays a two-dimensional barcode graphic in a fixed position on an LCD screen. The barcode uses the QR Code standard version V4, and the module size is set to 0.5 mm. After the patient completes registration at the self-service terminal using an identification card provided by the hospital, the system backend writes the patient's unique identification number, registration serial number, and timestamp information into the QR code.

[0017] During the QR code scanning interaction phase, the patient performs a scan confirmation operation on the operation panel of the uroflowmeter. The built-in scanning camera on the operation panel captures the QR code image at a resolution of 640×480 pixels. The image data is then processed by the embedded image processing chip using a fixed threshold binarization method to parse the QR code matrix. The decoded content is then parsed into patient identification information using UTF-8 encoding. The parsing result includes the patient's unique identification number, registration serial number, and timestamp. The system integrates these three fields into patient identity data and verifies the integrity of the data structure through validation logic. After successful verification, the patient identity data is written to the uroflowmeter's identity cache register.

[0018] In another embodiment of the invention, patient identity is obtained through a facial recognition device deployed at the hospital user terminal. The facial recognition device is an infrared-illuminated high-definition camera with a resolution of 1920×1080 pixels and a frame rate of 30 frames per second. When the patient stays in an area of ​​30 to 50 centimeters in front of the device, the system automatically captures five consecutive frames of facial images.

[0019] In the image preprocessing stage, the illumination normalization method based on histogram equalization is first called to limit the grayscale range to 0 to 255 in order to eliminate illumination differences. Then, the key coordinate points of the corners of the eyes, nose, and mouth are determined by the face feature point detection module, the affine transformation matrix is ​​calculated, and geometric alignment is completed. Finally, the feature enhancement method based on local binary mode is called in the normalized face region to generate a feature vector with a length of 256 dimensions.

[0020] The generated feature vector is input to the comparison module, which calls the Euclidean distance calculation formula. The system traverses all registered features in the face database and selects the identity label corresponding to the minimum Euclidean distance as the matching result. When the minimum distance is less than the preset threshold of 0.35, the match is considered successful, the matched identity label is extracted as the patient's identity data, and written to the uroflowmeter's identity cache register.

[0021] It is worth noting that QR codes and facial recognition devices can be used simultaneously.

[0022] First, the patient's identification number is obtained during the QR code scanning process, followed by secondary confirmation via facial recognition. When the patient identification numbers output by both methods match, the system binds this number to the collection time, forming enhanced identity data. If the two methods do not match, the system triggers an identity verification failure alarm. The alarm message is displayed via an audible notification on the uroflowmeter's control panel, and the abnormal event is recorded in the log file. Log fields include the QR code decoding result, facial comparison result, and comparison failure time. This enhanced identity data is consistently stored as an index field during subsequent urine data collection.

[0023] Preferably, the change in the output voltage of the weighing sensor in step S2 to form the original signal includes: When urine flows into the urine volume signal acquisition device equipped with a weighing sensor, the strain gauge of the weighing sensor deforms and generates a resistance change signal. The resistance change signal is transmitted through the bridge circuit to form a resistance signal; The resistance signal is amplified 128 times by the measurement circuit and then conditioned by a low-pass filter with a cutoff frequency of 10Hz to be converted into a voltage signal. The voltage signal is converted into the original signal through an analog-to-digital converter circuit.

[0024] In this embodiment of the invention, when urine enters the urine volume signal acquisition device, the weighing sensor at the bottom of the device absorbs stress due to the weight of the urine. This weighing sensor employs a strain gauge structure, with each strain gauge having a resistance of 350Ω. The sensitivity coefficient is limited to 2.0 ± 0.05. As the weight of urine increases, the metal wire of the strain gauge undergoes a slight elongation, resulting in a linear change in resistance.

[0025] All strain gauges are connected via a Wheatstone bridge circuit, which is powered by a stable 5V DC supply and configured as a full-bridge structure. Changes in resistance result in a millivolt-level differential voltage signal across the bridge, ranging from 0 mV to 20 mV.

[0026] The differential voltage signal first enters the measurement circuit, which consists of a low-noise instrumentation operational amplifier, model AD620, with a gain set to 128 and an input impedance higher than 10MΩ. The operational amplifier amplifies the millivolt-level voltage and then conditions it with a low-pass filter. The filter cutoff frequency is limited to 10Hz to suppress high-frequency noise interference. The conditioned voltage signal is stabilized in the range of 0V to 3V.

[0027] Finally, the conditioned voltage signal is converted into a raw digital signal through an analog-to-digital converter (ADC). The ADC uses a 12-bit precision ADC chip, model ADS1115, with a sampling rate set to 128 SPS (Samples per second). Each sample outputs a digital value between 0 and 4095, corresponding to a voltage resolution of 0.73 mV. The sampling results are written in a time sequence to the buffer register of the urine flow meter's microprocessor, forming the raw signal sequence.

[0028] In another embodiment of the invention, the urine volume signal acquisition device uses a bridge-type weighing sensor composed of four parallel strain gauges, with a maximum range of 2000 grams and a resolution of 0.1 grams. As urine gradually flows in, the strain gauges experience a slight change in resistance. This change in resistance is transmitted through a precision bridge circuit to output a differential voltage signal, with the bridge output signal amplitude ranging from -15mV to +15mV.

[0029] The differential voltage signal enters a high-precision preamplifier circuit. The amplifier uses an INA333 chip with a gain set to 200, and a hardware limiting circuit restricts the output voltage to 0V to 2.5V. The amplified voltage signal is then sent to a secondary conditioning circuit, which includes a second-order Butterworth low-pass filter with a cutoff frequency of 8Hz. The filter consists of precision capacitors and metal film resistors and is used to remove 50Hz power supply interference and transient spike signals.

[0030] The voltage signal, after secondary conditioning, is input to the analog-to-digital converter (ADC). The ADC uses a 16-bit resolution, a reference voltage of 2.5V, and a sampling interval of 10 milliseconds. Each sample yields a digital value between 0 and 65535, corresponding to a voltage resolution of 38. V. The digitized result is stored in the data acquisition buffer of the uroflowmeter, forming a raw signal sequence that can be used for subsequent filtering processing.

[0031] In another embodiment of the invention, the weighing sensor in the urine volume signal acquisition device adopts a half-bridge strain gauge configuration, and the maximum weighing capacity of the sensor is limited to 1000 grams. After urine enters the container, the resistance difference output by the sensor is introduced into the reference resistor through the half-bridge circuit, forming an unbalanced bridge, and the output signal fluctuates within the range of ±10mV.

[0032] The signal first passes through a differential amplifier, using an LT1167 device with a gain limited to 100. The amplified signal then enters a multi-stage filtering and conditioning module. The first stage is an RC low-pass filter with a cutoff frequency of 12Hz; the second stage is a passive notch filter to suppress 50Hz power frequency interference. After multi-stage filtering, the voltage signal amplitude is stabilized between 0V and 1.8V.

[0033] Subsequently, the voltage signal is fed into an analog-to-digital converter (ADC) integrated within the microprocessor. This ADC has a 10-bit resolution, a sampling frequency of 100Hz, and a quantization step size of 1.76mV per sample. The digitized signal is written to a FIFO buffer queue in real time, with a buffer depth of 256 points. As sampling progresses, the original signal accumulates point by point in the buffer.

[0034] Preferably, step S3 includes: Call the artifact detection logic to remove abrupt changes in the original signal where the velocity increment exceeds the threshold and the duration is less than 0.2 seconds; The original signal after rejection is input to the microprocessor cache via the data interface; The microprocessor calls the zeroing program to correct the initial deviation of the weighing sensor based on the original signal and outputs a correction signal. The microprocessor uses anti-interference circuitry and threshold discrimination logic to remove interference pulses from the correction signal and output a clean signal. The IIR Butterworth low-pass filter in the microprocessor is loaded to control the frequency of the purified signal and form a filtered signal.

[0035] In this embodiment of the invention, the sampling frequency of the raw signal output by the urine volume signal acquisition device is set to 100Hz.

[0036] First, the microprocessor invokes the artifact detection logic to calculate the volume difference between two consecutive sampling points. When the difference exceeds 20 mL / s and the duration of the change is less than 0.2 seconds, it is identified as an artifact abrupt change and removed. The removed signals are then reassembled into a continuous data stream.

[0037] Subsequently, the data is transmitted to the microprocessor cache via the SPI interface. The cache capacity is set to 1024 points and stored using a FIFO queue.

[0038] Within the buffer, the microprocessor invokes the zeroing procedure. This procedure acquires 100 raw signals under no-load conditions and calculates the average value as the zero-point reference. Subsequent signals are then subtracted from this reference value point by point to form the correction signal.

[0039] The corrected signal enters the anti-interference circuit and threshold discrimination logic. The anti-interference circuit is an RC low-pass circuit with a cutoff frequency of 20 Hz. The threshold discrimination logic compares the corrected signal with a threshold of ±5 sampling steps. Anomalies exceeding the threshold are discarded and the signal is padded with the average of two adjacent points. After processing, a purified signal is output.

[0040] Finally, the purified signal is input to the IIR Butterworth low-pass filter built into the microprocessor. The filter is of order 4, with a cutoff frequency of 5 Hz, and uses fixed-point arithmetic to recursively generate the filtered signal point by point.

[0041] In another embodiment of the invention, the sampling frequency of the original signal is set to 50Hz.

[0042] The artifact detection logic uses a sliding window approach with a window length of 10 points. If the maximum flow rate increment exceeds 30 mL / s within the window and its duration is less than 0.2 seconds, it is marked as an artifact segment and removed. The removed data is then rearranged according to timestamps.

[0043] The purified raw signal is transmitted to the microprocessor cache at a baud rate of 115200 via the UART interface, with a cache depth of 2048 points.

[0044] During the zero-calibration process, the microprocessor records a reference signal for 2 seconds before the test begins, collecting a total of 100 points. The arithmetic mean of all sampled values ​​is taken as the zero-point reference value. During the test, each sampled value is subtracted from this reference value, and a correction signal is output.

[0045] The anti-interference mechanism is implemented jointly by a hardware notch filter and logical discrimination. The notch filter operates at a frequency of 50 Hz to remove power frequency interference; the logical discrimination threshold is set to ±10 sampling steps, and points exceeding the range are replaced by the average of the preceding and following points. The purified signal is thus generated.

[0046] The purified signal is fed into an IIR Butterworth low-pass filter. This filter is a third-order filter with a cutoff frequency of 8 Hz. Each point is calculated using a recursive formula, and the final output is a filtered signal sequence, which is then written to a storage unit.

[0047] In another embodiment of the present invention, the sampling frequency is set to 200Hz and the single-point time interval is 5 milliseconds.

[0048] The artifact detection logic sets a threshold of 50 mL / s. If the difference between any point and the previous point exceeds this threshold and the duration of the anomaly is less than 0.2 seconds, the signal segment is discarded. The discarded sequences are immediately spliced ​​together to maintain the continuity of the time series.

[0049] The processed signal passes through The interface writes to the microprocessor cache, which has a capacity of 4096 points.

[0050] The zeroing procedure runs before urination begins, collecting 500 baseline signals and using the median as the zero baseline. The corrected signal is obtained by subtracting the zero baseline from all sampled values.

[0051] The corrected signal is processed by an anti-interference circuit. This circuit consists of a two-stage RC low-pass filter, with a first-stage cutoff frequency of 30Hz and a second-stage cutoff frequency of 15Hz. The threshold discrimination logic sets the threshold to ±7 sampling steps. Points exceeding the threshold are discarded, and the signal is replaced with the weighted average of the three most recent points to generate a purified signal.

[0052] The purified signal is processed by a second-order IIR Butterworth low-pass filter with a cutoff frequency of 6Hz, using 16-bit fixed-point arithmetic. The output is a filtered signal sequence, with each point accompanied by a timestamp.

[0053] Preferably, in step S4, when the total amount of urine output is less than the threshold amount, outputting a prompt message and recording the urination time includes: When the total amount of urine output is less than 150 ml, the microprocessor sends a control signal to the buzzer. The buzzer receives the signal and emits a buzzing sound; The start and end times of urination are written into the microprocessor's storage unit to form time data.

[0054] It should be noted that this invention includes a urine volume threshold determination function, setting the effective voiding threshold to 150 ml. Specifically, this is based on common knowledge in the field of clinical urinary flow rate testing: under natural urination conditions, when the urine volume is less than 150 ml, the bladder is insufficiently full, and urethral resistance and detrusor muscle contraction do not reach their normal physiological state. Consequently, key parameters such as the measured maximum and average urine flow rates are prone to significant deviations and cannot accurately reflect lower urinary tract function. Therefore, this invention uses a 150 ml threshold for automatic determination, which can promptly alert and mark invalid voiding events when urine volume is insufficient, avoiding the inclusion of invalid data in the analysis and thus ensuring the accuracy and reliability of the test results.

[0055] In this embodiment of the invention, after urine flows into the urine volume signal acquisition device, it is output by a weighing sensor and converted into a raw signal by an analog-to-digital converter. The microprocessor then filters the raw signal to obtain urine volume data. The microprocessor accumulates the urine volume data in each sampling cycle and calculates the total amount of urine excreted in real time.

[0056] When the total urine output is less than 150 ml, the microprocessor outputs a set of control level signals to the buzzer driver. The buzzer is an active piezoelectric buzzer with a rated voltage of 3.3V, an operating current of less than 20 mA, and a control level of 3.3V (high level). When the buzzer receives a high-level signal, it immediately emits a continuous buzzing sound to indicate that the urine output has not reached the threshold.

[0057] Simultaneously, the microprocessor records a start timestamp when it detects the start of urination and an end timestamp when it detects the end of urination. Both timestamps are based on the time provided by the real-time clock (RTC) module, with an accuracy limited to ±1 millisecond. The microprocessor writes the start and end timestamps into a time data table in its internal storage unit. The time data table fields include urination number, start time, and end time.

[0058] In another embodiment of the invention, the urine volume calculation unit calculates the total urine volume by summing the instantaneous volume every 100 milliseconds as a sampling period. If the total volume is less than 150 ml at the end of urination, the microprocessor outputs a pulse control signal to the buzzer via the GPIO interface. The buzzer has a rated frequency of 2.4 kHz, and the drive circuit maintains a 10-millisecond pulse width when outputting a high level, repeating three times at a 1-second cycle. After receiving three pulse signals, the buzzer produces an intermittent humming sound to alert the operator.

[0059] During the urination time recording process, the microprocessor calls an internal high-precision timer with a counting frequency of 32.768kHz. The timer value is captured and stored as the start time code at the moment urination begins, and captured and stored as the end time code at the moment urination ends. The microprocessor converts the time codes into a standard time format and stores them in the EEPROM memory, forming complete urination time data.

[0060] In another embodiment of the invention, the total urine output is calculated based on the cumulative result at the end of the urine volume curve. When the cumulative result is less than 150 ml, the microprocessor invokes the buzzer driver module to emit a 5-second warning sound. The buzzer driver module includes a power MOSFET switch with a switching threshold voltage of 2.5V, driven by the microprocessor's GPIO port. The buzzer is a 12 mm disc type with a rated sound pressure level greater than 85 dB, and the warning sound is clearly audible in a quiet environment.

[0061] For recording voiding time, the microprocessor uses a DS3231 real-time clock chip with a clock error limited to ±2 ppm. When the first effective volume increment occurs at the start of voiding, the system writes this moment into the storage unit as the start time; if no new volume increment occurs for 10 consecutive seconds, the system automatically determines the end of voiding and writes this moment into the storage unit as the end time. The storage unit is a flash memory module, and data is written in a fixed-length structured record format. Each record contains the patient number, start time, and end time.

[0062] Preferably, in step S4, when the total urine output is greater than or equal to the threshold total output, the calculation of urine data includes: When the total urine output is greater than or equal to 150 ml, the instantaneous volume corresponding to the current filtered signal is accumulated point by point to determine the total urine output data and the urine volume data. Instantaneous flow velocity data were calculated based on urine volume data and sampling time intervals, and a flow velocity data sequence was constructed. Identify the maximum urinary flow rate data in a flow rate data sequence; The instantaneous flow rate data were averaged over the urination period to calculate the average flow rate data; The total urine volume, urine volume, maximum urine flow rate, and average flow velocity data are integrated into urine data.

[0063] In this embodiment of the invention, the filtered signal output by the urine volume signal acquisition device is accumulated by a microprocessor to obtain instantaneous volume data. The sampling period is set to 100 milliseconds, and each filtered signal point is converted into a corresponding urine volume value. The microprocessor accumulates the volume value point by point until urination ends. If the final accumulated result is greater than or equal to 150 ml, it is determined that the threshold has been reached. At this time, the accumulated result is written to the total volume register as the total urine volume data, and at the same time, the point-by-point volume values ​​recorded in time series are stored in the volume data table as the urination volume data.

[0064] In the flow velocity calculation stage, the sampling time interval is fixed at 0.1 seconds. The instantaneous flow velocity is calculated as follows: ;in, For the first Cumulative volume at time step The sampling interval is 0.1 seconds. The calculated instantaneous flow velocity data is stored in array form, forming a flow velocity data sequence.

[0065] The microprocessor then compares the flow rate data point by point in the sequence, identifies the maximum value, and marks this value as the maximum urinary flow rate. Further, the arithmetic mean of all instantaneous flow rate data is calculated over the entire voiding period: ;in, This represents the total number of sampling points during the urination period. The average value is the average flow rate data. Finally, the total urine volume data, urine volume data, maximum urine flow rate data, and average flow rate data are integrated into a complete urine data record and written to the storage unit.

[0066] In another embodiment of the invention, the sampling frequency of the filtered signal is set to 50 Hz, i.e., sampling once every 20 milliseconds. After the volume increment of each sampling point is calculated, the points are superimposed to form a urine volume curve. If the value at the end of the urine volume curve is greater than or equal to 150 ml, the urine data is calculated.

[0067] In flow rate calculation, the difference between two adjacent volume points is divided by 0.02 seconds to obtain instantaneous flow rate data, which is then continuously stored in a buffer to form a flow rate sequence. The length of this sequence depends on the duration of urination and is generally between several hundred and several thousand points.

[0068] In maximum urinary flow rate identification, the microprocessor calls point-by-point comparison logic to scan the flow velocity sequence sequentially and stores the maximum value as the maximum urinary flow rate data. For average flow velocity calculation, all instantaneous flow velocities are summed and then divided by the number of points to obtain the average flow velocity data.

[0069] Finally, total urine volume data is stored in milliliters, urine volume data is recorded in time series, and maximum urine flow rate data and average flow velocity data are integrated into urine data and written to the flash memory module in binary structured format for subsequent urine report analysis.

[0070] In another embodiment of the invention, the sampling frequency of the filtered signal is set to 200Hz, i.e., the sampling period is 5 milliseconds. The microprocessor converts each sampling point into an increase in urine volume and accumulates them point by point to generate a volume curve. The total urine output is obtained at the end of the volume curve, and when the total output is greater than or equal to 150 ml, the urine data calculation stage begins.

[0071] The instantaneous flow velocity is calculated by dividing the volume difference between the i-th and (i-1)-th sampling points by the sampling interval of 0.005 seconds at any i-th sampling point. This method provides higher flow velocity resolution and can capture subtle flow velocity fluctuations.

[0072] During the identification of maximum urinary flow rate data, the microprocessor uses a continuous comparison method to update the maximum value register sequentially, ultimately retaining the maximum flow rate value in the entire sequence. For average flow rate calculation, the sum of all instantaneous flow rates is divided by the total number of sampling points to obtain the average flow rate data, expressed in milliliters per second.

[0073] Finally, the total urine volume data, point-by-point volume data, maximum urine flow rate data, and average flow velocity data are uniformly encapsulated into a urine data structure, with fields including total volume field, volume sequence field, maximum flow velocity field, and average flow velocity field.

[0074] Most importantly, by progressively accumulating the instantaneous volume corresponding to the current filtered signal, the total urine volume and urine output volume data are determined, including: Based on the detection of abrupt changes in the initiation and interruption phases of urine flow using filtered signals, motion artifacts are suppressed, and a smooth signal is generated. Identify the peaks and troughs corresponding to changes in urine weight in the smooth signal, extract stable segments, and determine the peak urine data; The preset urine density parameter is used to convert the peak urine data into urine volume data; The urine volume data are accumulated over time to form a urine volume curve; The cumulative results are truncated at the end of the urine volume curve to determine the total urine volume data.

[0075] In this embodiment of the invention, the filtered signal is output by a microprocessor, each sampling point corresponds to a weight change value, and the sampling frequency is 100Hz.

[0076] First, in the initial stage when urine enters the collection device, the microprocessor invokes the mutation detection logic. This logic compares the differences between adjacent sampling points; when the difference exceeds 0.5 grams, it determines the start of the urine flow; when there is no weight increase for 5 consecutive seconds and the signal fluctuation is less than 0.2 grams, it determines the interruption of the urine flow. In this way, the start and end points of urination are identified.

[0077] Between the start and end points, the microprocessor uses a moving average method to process the signal with a window width of 10 sampling points to eliminate motion artifacts and generate a smooth signal.

[0078] In a smooth signal, the microprocessor identifies local peaks and troughs in the weight curve. A peak is defined as the maximum value within 10 adjacent sampling points, and a trough is defined as the minimum value within 10 adjacent sampling points. The system extracts these stable segments, treating the peak values ​​as the peak urine weight data.

[0079] Next, the preset urine density parameters were used, with the density fixed at 1.02 g / mL. The conversion formula was then applied. ;in, This represents peak urine weight data. The urine density is 1.02 g / mL, and the urine volume data is calculated accordingly.

[0080] All volume data are accumulated point by point over time to form a urine volume curve. The cumulative value at the end of the curve is the total urine volume data. The total urine volume data and the point-by-point urine volume data are stored together in a data register.

[0081] In another embodiment of the present invention, the sampling frequency of the filtered signal is set to 50Hz, and the interval between each point is 20 milliseconds.

[0082] In the urine flow initiation detection process, the microprocessor monitors the average increment of three consecutive sampling points. When the average increment exceeds 0.3 grams, it is marked as the urine flow initiation point. When there is no increment for 200 consecutive sampling points in subsequent sampling and the fluctuation is less than 0.1 grams, it is marked as the urine flow interruption point.

[0083] Within the urination range, the microprocessor uses the smoothed signal output by the hardware low-pass filter to further remove interference spikes and obtain a stable curve.

[0084] In the stable curve, the positions of each peak and trough are identified, and the difference between adjacent peaks and troughs is taken as the weight change amplitude. The microprocessor defines these weight change amplitudes as urine weight peak data and stores them in a weight data table.

[0085] Subsequently, the peak weight data is converted into volume values ​​using a formula. All converted volume values ​​are accumulated in the sampling order to form a volume curve. The microprocessor reads the accumulated value at the end of the volume curve as the total urine volume data. The final result is stored in binary format, with fields including a total volume field and a point-by-point volume sequence field.

[0086] In another embodiment of the present invention, the sampling frequency of the filtered signal is 200Hz and the single-point time interval is 5 milliseconds.

[0087] The microprocessor first uses a mutation detection mechanism to identify the start and end points: when the weight change at a single point is greater than 0.4 grams, it is defined as the start point; when the weight does not increase for 1000 consecutive points and the fluctuation range does not exceed 0.1 grams, it is defined as the interrupt point.

[0088] Within this interval, the filtered signal is processed by a moving average with a window length of 20 points to form a smooth signal.

[0089] In the smoothed signal curve, the system identifies the weight increase and decrease regions segment by segment and extracts the weight peaks. Each peak represents the weight of a stable urination segment.

[0090] Using the urine density parameter of 1.02 g / mL, the peak weights were converted into volume data one by one. All volume data were added point by point in chronological order to form a complete urine volume curve.

[0091] The microprocessor reads the cumulative result at the last point of the volume curve and stores it as total urine volume data. The total urine volume data is written to the flash memory along with the volume curve.

[0092] Of particular importance is the recording of all urine data from patients within a preset collection period, and the binding of this data with patient identification information to create personalized urine data, including: All urine data of the patient within the preset collection period are timestamped; All labeled urine data are matched and bound with patient identity data, and encrypted and written to the built-in storage module of the urine flow meter; At the end of the preset collection period, all stored data is structured and organized to form personalized urine data.

[0093] In this embodiment of the invention, the microprocessor of the uroflowmeter immediately calls the real-time clock module after each urine output data is generated. The real-time clock module is a DS3231 with a timing accuracy limited to ±2ppm. The microprocessor reads the time information at that moment and generates a standardized timestamp in the format "YYYY-MM-DD HH:MM:SS.sss", where "sss" represents the millisecond time. Each urine data point is appended with a unique timestamp.

[0094] After timestamp annotation, the microprocessor binds the urine data to the patient's identity data. The patient's identity data includes three fields: a unique patient identification number, a registration serial number, and the registration time. The binding process is achieved by establishing an index relationship in the data structure: the urine data is stored in a sequential list, with each record appended with an identity index field.

[0095] During data writing, the urine flow meter's built-in storage module is an encrypted flash memory chip with a capacity of 64MB, supporting AES-128 hardware encryption. The microprocessor invokes the encryption circuit during writing to encrypt urine data and identity data in 128-bit blocks before writing them to the flash memory storage unit.

[0096] At the end of the preset collection period (set to 24 hours), the microprocessor calls the data processing program to rearrange all stored data in timestamp order, remove redundant indexes, and output a structured data table. This table includes fields such as the patient's unique identifier, timestamp, total urine volume, voiding volume sequence, maximum urine flow rate, and average flow velocity, ultimately forming personalized urine data.

[0097] In another embodiment of the invention, a microprocessor assigns a sequence number to each urine data entry upon generation, with the sequence number incrementing by a natural number. A real-time clock module provides a timestamp, and each urine data entry contains both a sequence number and a timestamp as dual identifiers.

[0098] The microprocessor binds the timestamp to the patient's identity data and stores it in the built-in EEPROM as record blocks. The EEPROM storage unit has a capacity of 8MB, and each record block is 512 bytes in size, including an identity information field (32 bytes), a timestamp field (16 bytes), a urine data field (400 bytes), and an encryption verification field (64 bytes). The encryption method uses a SHA-256 hash checksum to ensure the integrity of the data during storage.

[0099] At the end of the collection cycle (set to 7 days), the microprocessor sequentially reads the record blocks in the EEPROM and merges them into a complete cycle dataset in ascending order of sequence number. This cycle dataset is then formatted into a JSON structure, with fields including "patient_id", "timestamp", "urine_volume", "max_flow_rate", and "avg_flow_rate". This completes the processing and forms personalized urine data.

[0100] In another embodiment of the invention, the timestamp annotation of urine data is performed by a timer built into the microprocessor, with a timer frequency of 1MHz, and the count value is directly converted into a millisecond timestamp. All urine data is written to a buffer immediately after generation, and a patient identity data field is added to the buffer.

[0101] Identity data binding is achieved by establishing a chain structure: each urine data node contains urine parameters and an identity information pointer, and all nodes are linked into a data linked list in chronological order.

[0102] In terms of data security, the storage module uses an eMMC chip with a hardware security engine, has a capacity of 128MB, and supports AES-256 encryption. The microprocessor uses a randomly generated key during writing, and the key is refreshed after each write to prevent the leakage of repeatedly encrypted data.

[0103] When the preset collection period (set to 72 hours) ends, the microprocessor calls a sorting function to export the data list into a two-dimensional table structure. Each row in the table represents a urination event, and each column represents specific parameters, including patient ID, start time, end time, total urine volume, flow rate sequence, and average flow rate. This two-dimensional table, as personalized urine data, is stored in the file system and prepared for uploading to the server.

[0104] Preferably, uploading personalized urine data to the wireless network connection center server in step S6 includes: When the preset acquisition cycle ends, the microprocessor in the urine flow meter sends a cycle end signal. Personalized urine data is packaged using a microprocessor and uploaded to a central server connected to a wireless network. The wireless network connection center server receives personalized urine data and writes it into the database.

[0105] In this embodiment of the invention, when the preset acquisition cycle ends, the microprocessor of the urine flow meter reads the time data from the RTC real-time clock module and determines whether the difference between the current time and the start time of the acquisition cycle reaches 24 hours. If it reaches 24 hours, the microprocessor sends a cycle end signal on the control bus. The signal is a high level of 3.3V and lasts for 100 milliseconds.

[0106] The end-of-cycle signal triggers the data packaging process. The microprocessor retrieves the personalized urine data table from memory, writes each data table field (including patient ID, timestamp, total urine volume, volume sequence, maximum flow rate, and average flow rate) into the buffer, and generates a data packet header. The header fields include the data packet length (in bytes), a CRC32 checksum, and the collection cycle number.

[0107] After the data is packaged, the microprocessor uploads the data via the wireless communication module. The wireless communication module is a 4G LTE communication unit, supporting an uplink speed of 5Mbps. The upload uses the TCP / IP protocol, with a fixed port number of 8080. During transmission, a checksum signal is added after each 1KB data packet is sent to confirm successful server reception before continuing transmission.

[0108] The wireless network connection center server runs a Linux system on the receiving end, with the receiving process listening on port 8080. When a complete data packet is detected, the server calls a verification program to verify the CRC32 checksum. After successful verification, the data is unpacked. The unpacked personalized urine data is written to a MySQL database, with database table fields including patient ID, cycle number, total volume, flow rate sequence, maximum flow rate, and average flow rate.

[0109] Preferably, the analysis and generation of urine report data in step S6 includes: Remove artifacts and invalid urination events from personalized urine data and output purified data; Convert the purification data into key numerical parameters; Urine flow rate sequences were constructed using key numerical parameters, and urine flow curves were plotted. Compare the urine flow curve with historical database samples, and output the verification data. Key numerical parameters, urine flow curves, and calibration data are integrated into urine report data.

[0110] In this embodiment of the invention, after personalized urine data is uploaded to the wireless network connection center server, the server invokes a data processing program to first remove artifact signals and invalid urination events. The criteria for determining artifact signals are: a flow rate increase greater than 50 mL / s and a duration less than 0.2 seconds within three consecutive sampling points; the system marks such abrupt changes as artifacts and removes them. The criteria for determining invalid urination events are: urination events with a total volume less than 20 ml or a duration less than 3 seconds are directly deleted. After the removal operation is completed, the purification data is output.

[0111] The purification data was converted into key numerical parameters, including: total urine volume (mL), urination duration (s), maximum urine flow rate (mL / s), average urine flow rate (mL / s), and the start and end points of urination time (timestamp format).

[0112] Based on key numerical parameters, the server constructs a urine flow rate sequence. This sequence is formed by calculating the instantaneous flow rate from volume data and a sampling interval (0.1 seconds), with each point in the sequence expressed in milliliters per second. The graphics module is then invoked to plot the urine flow curve, with time on the horizontal axis and instantaneous flow rate on the vertical axis.

[0113] After the curve is plotted, it is compared with historical database samples. The historical database samples include standard urine flow curves stored over the past five years, grouped by age and gender. The system compares each point and calculates the mean squared error. When the mean squared error is less than 5%, the curve is considered to conform to the reference sample, and "normal" verification data is generated; otherwise, "deviation" verification data is generated.

[0114] Ultimately, key numerical parameters, urine flow curves, and validation data are integrated into urine report data. The report data uses a PDF file structure, with fields including a title page, numerical tables, graphs, and validation conclusions, and is stored uniformly in the server database.

[0115] In another embodiment of the invention, after receiving personalized urine data, the server first preprocesses the data. The rules for judging artifact signals are: sampling points with an instantaneous flow rate greater than 200 mL / s are directly rejected; the criteria for invalid urination events are a total urine volume of less than 15 mL or a total duration of less than 2 seconds. After rejection, purified data is obtained.

[0116] The purification data is converted into key parameters, including: maximum flow rate, average flow rate, time of peak flow rate, and slope of the cumulative volume curve. All parameters are represented as floating-point numbers and stored in a parameter table.

[0117] Using parameter table data, the system calculates instantaneous flow velocity values ​​point by point and constructs a urine flow rate sequence. The plotting program uses SVG vector graphics to generate urine flow curves, with an output resolution of 600 dpi.

[0118] When comparing the urine flow curve with historical database samples, the system compares it with reference samples of the same age group and gender. The comparison method is as follows: the test curves are aligned by time, and the differences in key point positions are calculated (including the peak time point, the slope of the rising segment of the curve, and the slope of the falling segment). If all three differences are less than 10%, the output verification data is "compliant"; otherwise, it is "incompatible".

[0119] Finally, the system packages the key numerical parameters, urine flow curve SVG file, and verification results into a structured JSON file and stores it in a PostgreSQL database to form urine report data.

[0120] In another embodiment of the invention, when the server performs artifact removal on personalized urine data, a continuous stationary detection method is used: the velocity variance is greater than a threshold (the threshold is set to 25) within 10 sampling points. When a segment is identified as a pseudo-video segment, it is marked and deleted. For invalid urination events, the standard is set as a total urine volume of less than 10 mL or a urination duration of less than 1 second. After removal, purification data is generated.

[0121] The purification data is calculated and converted into key numerical parameters, including: total volume, maximum flow rate, average flow rate, start time, end time, and peak occurrence time. The parameters are stored in binary floating-point format with a length of 32 bits.

[0122] Using the above parameters, the system recalculates the instantaneous flow rate according to the sampling interval (5 milliseconds) and generates a high-resolution urine flow rate sequence. The urine flow curve is plotted as a PNG bitmap with an image size of 800×600 pixels and a grayscale precision of 8 bits.

[0123] During the verification phase, the urine flow curve is compared point by point with historical database samples, and the correlation coefficient is calculated. When the correlation coefficient is greater than 0.9, the verification data output is "normal"; when the correlation coefficient is between 0.7 and 0.9, the output is "borderline"; and when it is less than 0.7, the output is "abnormal".

[0124] Finally, key numerical parameters, urine flow curves, and verification data are integrated into a single report file, presented as numerical tables, PNG images, and text fields. The report file is stored in XML format, with fields including report number, patient number, key parameters, curve file path, and verification conclusion, and is archived as urine report data.

[0125] Preferably, step S7 includes: Transform urine report data into a structured format that combines charts, graphs, and numerical values. The structured data is plotted into urine flow curves, total volume curves, and parameter tables, and the rendered data is output. Extract key numerical parameters from the structured data, compare the key numerical parameters and urine flow curves with historical database samples grouped by gender and age point by point, and output personalized verification data based on the comparison results of mean square error and curve slope. The rendering data and personalized verification data are encapsulated into a standardized file format to form encapsulated data. The encapsulated data is transmitted to the user terminal, which includes both the doctor's and the patient's terminals.

[0126] In this embodiment of the invention, after receiving the urine report data, the server first converts the report data into a structured format. The structured format uses JSON, and the fields include "total_volume", "max_flow", "avg_flow", "duration", "timestamp", "flow_curve", and "volume_curve". The urine flow curve and total volume curve are stored in the form of a time series array, and the key numerical parameters are stored as floating-point numbers with three decimal places.

[0127] After conversion, the server calls the plotting module to draw the structured data into three parts: a urine flow curve (600 dpi resolution, x-axis for time, y-axis for flow velocity), a total volume curve (600 dpi resolution, x-axis for time, y-axis for cumulative volume), and a parameter table (800 pixels wide, Arial 10 font, header fields for total volume, maximum flow velocity, average flow velocity, and urination duration). The rendered data is then output.

[0128] Subsequently, the system extracts key numerical parameters from the structural format and compares them point-by-point with historical database samples. The historical database samples are divided into male and female groups by gender, and each group is further divided into five age ranges (<20 years, 20–40 years, 41–60 years, 61–75 years, >75 years). During the comparison, the mean square error between the urine flow curve and the database sample curve is first calculated.

[0129] Simultaneously calculate the slope difference between the two curves: calculate the slope separately for the rising and falling segments; if the difference is less than 0.2... If the slopes are consistent, then it is determined that the slopes are the same. The results of the mean square error and slope comparison together form personalized verification data.

[0130] The rendered data and personalized verification data are packaged into a PDF file, with a cover page, curve images, parameter tables, and a verification conclusion page. The final file, as packaged data, is transmitted to the user's end via HTTPS, including a browser management interface for doctors and a mobile application for patients.

[0131] In another embodiment of the invention, the urine report data is converted using XML format, with the root node being "Report" and child nodes including "Patient", "Parameters", "FlowCurve", and "VolumeCurve". Key numerical parameters include maximum flow rate, average flow rate, urination time, and total urine volume, all represented as 32-bit floating-point numbers.

[0132] The plotting process utilizes the ECharts visualization library to render the XML structure into interactive urine flow and total volume curves. The curve resolution is 1920×1080 pixels, and the parameter table is generated as an HTML table. The rendering results are then merged into an HTML file as the rendering data.

[0133] In the verification phase, the system extracts key numerical parameters and urine flow curve sequences, comparing them with database samples. The database is divided by gender and age, with at least 100 reference curves stored for each group. The comparison method is as follows: the difference is calculated at each sampling point, and the average is taken to obtain the standard deviation (SD). Simultaneously, the difference between the peak position of the urine flow curve and the peak position of the database samples is calculated; if the difference is less than 5%, it is considered a time match. The SD and peak position difference results are combined to generate personalized verification data.

[0134] Finally, the rendered data and personalized verification data are packaged in a ZIP archive with a compression ratio controlled at 2:1. The packaged data is pushed to the doctor's end in real time via WebSocket, allowing the doctor to directly view the curves interactively in a browser interface. The patient's end loads an HTML file via a mobile application and displays the same content.

[0135] In another embodiment of the invention, the urine report data is converted into a CSV file with fields including timestamp, instantaneous flow rate, cumulative volume, maximum flow rate, average flow rate, and total urine output.

[0136] After the server reads the CSV file, it calls the Matplotlib plotting library to generate three types of images: a PNG image of the urine flow curve (1280×720 resolution), a PNG image of the total flow curve (1280×720 resolution), and an SVG image of the parameter table (containing four columns: total flow, maximum flow rate, average flow rate, and duration). These images constitute the rendering data.

[0137] During the verification phase, the server compares the urine flow curve point by point with the database samples, calculates the root mean square error, and simultaneously extracts the average slope of the descending segment of the curve, comparing it with the average slope of the corresponding age group samples. If the root mean square error is less than 10%, the slope difference is less than 0.3. If the value exceeds the threshold, the output verification data will be "normal"; if it exceeds the threshold, the output verification data will be "abnormal".

[0138] The rendered data and validation data are merged into an XML report file, with a file size not exceeding 5MB. This file is transmitted to the user's device via the MQTT protocol. Doctors open the XML file in the desktop application to view the curves and validation conclusions, while patients decode the XML in the mobile application to display parameter tables and images.

[0139] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is not limited by the foregoing description. Thus, all changes falling within the meaning and scope of the equivalents of the application are intended to be included within the scope of the invention.

[0140] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for intelligent dynamic urinary flow rate measurement, characterized in that, Applied to a uroflowmeter, the uroflowmeter including a urine volume signal acquisition device and a microprocessor, the method includes: Step S1: Collect patient identification data; Step S2: When urine flows into the urine volume signal acquisition device with a weighing sensor, the output voltage of the weighing sensor changes, forming the original signal; Step S3: The original signal is filtered by the microprocessor to remove interference pulses using threshold discrimination logic, and frequency control is performed using an IIR Butterworth low-pass filter. Step S4: When the total amount of urine output is less than the threshold amount, output a prompt message and record the urination time; when the total amount of urine output is greater than or equal to the threshold amount, calculate the urine data. Step S5: Within the preset collection period, repeat steps S2 to S4 to record all urine data of the patient within the preset collection period and bind it with the patient's identity data to form personalized urine data; Step S6: After the preset collection period ends, the personalized urine data is uploaded to the wireless network connection center server and analyzed to generate urine report data; Step S7: Visualize the urine report data and send it to the user's device.

2. The intelligent dynamic urinary flow rate measurement method according to claim 1, characterized in that, Step S1 includes: Obtain the interactive operation instructions of the QR code interface of the uroflowmeter, parse the patient identification information embedded therein, and extract the patient identity data; Alternatively, facial recognition devices deployed at hospital user terminals can be used to collect patients' facial images in real time. The collected images are then normalized for illumination, aligned with faces, and enhanced for features. The pre-processed image features are then matched with the registered features in a pre-defined patient face database, and patient identity data is extracted based on the feature matching results.

3. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, The change in the output voltage of the weighing sensor in step S2, forming the original signal, includes: When urine flows into the urine volume signal acquisition device equipped with a weighing sensor, the strain gauge of the weighing sensor deforms and generates a resistance change signal. The resistance change signal is transmitted through the bridge circuit to form a resistance signal; The resistance signal is amplified 128 times by the measurement circuit and then conditioned by a low-pass filter with a cutoff frequency of 10Hz to be converted into a voltage signal. The voltage signal is converted into the original signal through an analog-to-digital converter circuit.

4. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, Step S3 includes: Call the artifact detection logic to remove abrupt changes in the original signal where the velocity increment exceeds the threshold and the duration is less than 0.2 seconds; The original signal after rejection is input to the microprocessor cache via the data interface; The microprocessor calls the zeroing program to correct the initial deviation of the weighing sensor based on the original signal and outputs a correction signal. The microprocessor uses anti-interference circuitry and threshold discrimination logic to remove interference pulses from the correction signal and output a clean signal. The IIR Butterworth low-pass filter in the microprocessor is loaded to control the frequency of the purified signal and form a filtered signal.

5. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, In step S4, when the total urine output is less than the threshold total, a prompt message is output and the urination time is recorded, including: When the total amount of urine output is less than 150 ml, the microprocessor sends a control signal to the buzzer. The buzzer receives the signal and emits a buzzing sound; The start and end times of urination are written into the microprocessor's storage unit to form time data.

6. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, In step S4, when the total urine output is greater than or equal to the threshold total output, the urine data calculation includes: When the total urine output is greater than or equal to 150 ml, the instantaneous volume corresponding to the current filtered signal is accumulated point by point to determine the total urine output data and the urine volume data. Instantaneous flow velocity data were calculated based on urine volume data and sampling time intervals, and a flow velocity data sequence was constructed. Identify the maximum urinary flow rate data in a flow rate data sequence; The instantaneous flow rate data were averaged over the urination period to calculate the average flow rate data; The total urine volume, urine volume, maximum urine flow rate, and average flow velocity data are integrated into urine data.

7. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, Step S6, which involves uploading personalized urine data to the wireless network connection center server, includes: When the preset acquisition cycle ends, the microprocessor in the urine flow meter sends a cycle end signal. Personalized urine data is packaged using a microprocessor and uploaded to a central server connected to a wireless network. The wireless network connection center server receives personalized urine data and writes it into the database.

8. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, Step S6 involves analyzing and generating urine report data, including: Remove artifacts and invalid urination events from personalized urine data and output purified data; Convert the purification data into key numerical parameters; Urine flow rate sequences were constructed using key numerical parameters, and urine flow curves were plotted. Compare the urine flow curve with historical database samples, and output the verification data. Key numerical parameters, urine flow curves, and calibration data are integrated into urine report data.

9. The intelligent method for dynamic urinary flow rate measurement according to claim 1, characterized in that, Step S7 includes: Transform urine report data into a structured format that combines charts, graphs, and numerical values. The structured data is plotted into urine flow curves, total volume curves, and parameter tables, and the rendered data is output. Extract key numerical parameters from the structured data, compare the key numerical parameters and urine flow curves with historical database samples grouped by gender and age point by point, and output personalized verification data based on the comparison results of mean square error and curve slope. The rendering data and personalized verification data are encapsulated into a standardized file format to form encapsulated data. The encapsulated data is transmitted to the user terminal, which includes both the doctor's and the patient's terminals.

10. A dynamic urinary flow rate intelligent measurement system, characterized in that, For performing the intelligent dynamic urinary flow rate measurement method as described in claim 1, the intelligent dynamic urinary flow rate measurement system comprises: The patient identification data collection module is used to collect patient identification data; The urine signal acquisition module is used to generate a raw signal by changing the output voltage of the weighing sensor when urine flows into the urine volume signal acquisition device equipped with a weighing sensor. The signal filtering and processing module is used to filter the original signal using a microprocessor, remove interference pulses using threshold discrimination logic, and control the frequency using an IIR Butterworth low-pass filter. The threshold determination and processing module is used to output a prompt message and record the urination time when the total amount of urine is less than the threshold amount; and to calculate the urine data when the total amount of urine is greater than or equal to the threshold amount. The periodic data storage module is used to repeatedly execute steps S2 to S4 within a preset collection period, record all urine data of the patient within the preset collection period, and bind it with the patient's identity data to form personalized urine data. The data upload and analysis module is used to upload personalized urine data to the wireless network connection center server after the preset collection period ends, and analyze and generate urine report data. The report visualization and transmission module is used to visualize urine report data and send it to the user's device.

Images