Intelligent cleaning and monitoring system and method for ultrasound imaging probe

Abstract

The present disclosure relates to the technical field of medical devices, in particular, to an intelligent cleaning and monitoring system and method for an ultrasonic imaging probe, aiming to solve the problems of lack of quantitative monitoring, subjective effect and difficult data tracing in the prior art. A probe identifier is obtained before cleaning, and a reference atlas is generated by multi-spectral pre-scanning; cavitation energy index and cleaning liquid parameters are monitored in real time during cleaning; multi-spectral scanning is performed after cleaning, and a residual contaminant index is quantified; a cleaning quality score is calculated by fusing the indexes; a digital cleaning certificate is generated according to the score and is encrypted and stored for tracing. Through the above scheme, the present disclosure realizes objective quantification of effect, real-time monitoring of process, and data tracing of the whole process, and solves the problems of opaque cleaning, unreliable results, and difficult responsibility tracing.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of medical devices, and in particular, relates to an intelligent cleaning and monitoring system and method for an ultrasonic image probe. BACKGROUND

[0002] The technical field of medical devices covers a variety of products such as instruments, equipment, and appliances used for disease diagnosis, prevention, monitoring, treatment, or alleviation. The core content of this technical field is to combine engineering, biology, and medical knowledge to develop high-tech equipment for protecting human health and life safety. As an important part of the modern medical system, the technical level of medical devices is directly related to the accuracy of clinical diagnosis and the effectiveness of treatment. With the continuous progress of material science, electronic technology, and information science, medical devices are developing towards intelligence, miniaturization, and precision, which has a profound impact on many aspects such as clinical medicine, rehabilitation nursing, and public health.

[0003] Among them, the cleaning and disinfection method of the ultrasonic image probe is a key link in the infection control of medical devices. This method aims to completely remove biological tissues, coupling agents, and pathogenic microorganisms remaining on the surface of the probe through physical or chemical means to prevent cross-infection between different patients. In clinical applications, the cleanliness of the ultrasonic probe directly affects medical safety as it directly contacts the patient's skin or mucosa. A complete cleaning and disinfection process usually includes pretreatment, cleaning, disinfection, rinsing, and drying steps to ensure that the probe meets the specified hygiene standards after each use, thereby ensuring the safety and accuracy of subsequent diagnostic work.

[0004] The existing technology adopts a standardized fixed process in the probe cleaning process, lacks real-time evaluation of the initial contamination level of the probe, and cannot dynamically adjust the cleaning parameters according to the actual situation. The verification of the cleaning effect mainly relies on the naked eye observation of the operator or regular sampling detection, lacks objective and quantitative real-time monitoring means, and it is difficult to ensure the uniformity of the quality of each cleaning. Key data such as cleaning liquid concentration, action temperature, and time during the cleaning process are often not systematically recorded and analyzed, leading to the difficulty in tracing the cleaning process and providing data support for process optimization. In addition, the cleaning system and the monitoring system are usually independent of each other, and cannot form a closed-loop feedback control, resulting in the coexistence of resource waste and the risk of incomplete cleaning. These problems are particularly prominent in medical institutions with heavy diagnostic and treatment tasks, directly restricting the improvement of infection control level and equipment turnover efficiency. SUMMARY

[0005] The purpose of the present application is to provide an intelligent cleaning and monitoring system and method for an ultrasonic image probe, which aims to solve the technical problems of lack of objective and quantitative monitoring in the cleaning process, reliance on subjective judgment of cleaning effect, and inability to effectively trace the data of the whole process in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent cleaning and monitoring method for ultrasonic imaging probes, comprising the following steps:

[0007] S1: When the cleaning process is started, the unique identification code of the ultrasonic imaging probe to be cleaned is obtained, and its historical cleaning data and probe physical model parameters are called based on the unique identification code; then the multispectral imaging unit is driven to perform a pre-cleaning surface scan of the probe, collect a reference image dataset covering a preset spectral range, and extract the initial contaminant distribution characteristics of the probe surface through an image processing algorithm to generate a reference contaminant spectrum.

[0008] S2: Based on the aforementioned baseline contaminant spectrum, the cleaning operation execution unit is activated. The cleaning operation execution unit performs multi-stage physical and chemical cleaning operations, including high-pressure spraying, ultrasonic immersion cleaning, and high-temperature disinfection, within a sealed cleaning chamber according to a preset cleaning protocol. During the ultrasonic immersion cleaning stage, the acoustic field signal of the liquid medium within the cleaning chamber is collected in real time through an acoustic sensing unit, and its spectrum is analyzed to calculate the acoustic cavity cavitation energy index, which characterizes the intensity of the cavitation effect. Simultaneously, the key chemical parameters of the cleaning fluid are monitored online through a chemical sensing unit to generate a real-time fluid state vector.

[0009] S3: After the cleaning operation is completed, drive the multispectral imaging unit to perform post-cleaning surface scanning on the probe, and acquire a finished image dataset with the same spectral range and imaging parameters as the reference image dataset; perform pixel-level registration between the finished image dataset and the reference image dataset, and calculate the difference between the two datasets after registration to generate a residual contaminant differential image.

[0010] S4: Perform image segmentation and morphological analysis on the differential image of the residual pollutants to identify and quantify the patch area, grayscale integral value and shape complexity of the residual pollutants. Calculate the surface residual pollutant index based on the preset pollutant quantification model. Fuse the surface residual pollutant index, the acoustic cavity cavitation energy index and the real-time fluid state vector, and calculate a comprehensive cleaning quality assessment score through a multi-dimensional weighted fusion algorithm.

[0011] S5: Compare the comprehensive cleaning quality assessment score with a preset pass threshold. If the score is higher than the threshold, the cleaning is deemed qualified. The system generates a digital cleaning certificate containing a unique identifier, timestamp, monitoring data at each stage, and the final assessment score. The certificate is then encrypted and bound to the unique identifier of the probe and stored in an immutable distributed database. If the score is lower than the threshold, the cleaning is deemed unqualified, and the system automatically triggers a new round of cleaning operations.

[0012] In one embodiment of the present invention, the base contaminant map includes an initial contaminant coordinate matrix, a contaminant spectral response curve, and an initial contaminant load index; the acoustic cavity cavitation energy index includes the ultrasonic dominant frequency energy value, harmonic energy distribution, and broadband noise integral energy value; the real-time fluid state vector includes the conductivity, pH value, temperature, and light absorbance at a specific wavelength of the cleaning fluid; the surface residual contaminant index includes contaminant coverage, an estimated average residual thickness, and contaminant morphology dispersion; the digital cleaning certificate includes the probe RFID / NFC serial number, cleaning start and end timestamps, pre-cleaning and post-cleaning image summaries, acoustic cavity cavitation energy index time series, fluid state vector historical records, the final cleaning quality assessment score, and the operator's digital signature.

[0013] As a further embodiment of the present invention, the specific process of generating the reference pollutant spectrum in step S1 is as follows:

[0014] S111: Using a radio frequency identification (RFID) reader or near field communication (NFC) module, read the electronic tag attached to the probe handle or cable connector to obtain its globally unique probe serial number (US-ID); using this US-ID as an index, retrieve the probe's model, geometric dimensions, material information, and past cleaning and usage records from the device's historical database.

[0015] S112: Drives a two-dimensional rotation and translation mechanism controlled by a stepper motor, built into the cleaning chamber, to precisely control the movement of the probe within the field of view of the multispectral imaging unit; the multispectral imaging unit includes a complementary metal-oxide-semiconductor (CMOS) image sensor with a resolution of not less than 3840x2160 pixels, and a ring light source system arranged around the probe array; the light source system consists of multiple sets of independently controllable light-emitting diodes (LEDs), whose emission wavelengths cover ultraviolet light (UV-A) of 365 nm, blue light of 450 nm, green light of 525 nm, red light of 630 nm, and infrared light of 940 nm, respectively.

[0016] S113: The system sequentially illuminates LED light sources of different wavelengths and acquires a high-resolution image frame at each wavelength to form a baseline image dataset. Subsequently, a semantic segmentation model based on a convolutional neural network (CNN) is applied to process the image dataset, identify and segment potential contaminant regions with specific spectral response characteristics attached to the probe's acoustic window and the surface of the housing, such as the fluorescence reaction of protein residues under ultraviolet light irradiation, or the phosphorescence effect of specific coupling agents under blue light excitation. Finally, the coordinates, area, and average pixel intensity of all segmented contaminant regions are encoded to generate a structured baseline contaminant map.

[0017] As a further embodiment of the present invention, the specific process of calculating the acoustic cavity cavitation energy index and generating the real-time fluid state vector in step S2 is as follows:

[0018] S211: A broadband hydrophone with a center frequency of 2.5 MHz is symmetrically installed on the inner wall of the cleaning chamber; during the ultrasonic immersion stage, the hydrophone continuously acquires the sound field pressure signal at a sampling rate of not less than 5 megasamples / second (MS / s); the acquired time-domain signal is sent to a signal processing unit implemented by a field programmable gate array (FPGA).

[0019] S212: The signal processing unit performs a 4096-point Short-Time Fourier Transform (STFT) on the acquired time-domain signal, continuously generating a sound field spectrum with a time resolution of 10 milliseconds; for each spectrum frame, the system calculates the cavity cavitation energy index (CEI), whose mathematical expression is defined as:

[0020]

[0021] Where P(f) is the power spectral density function, and the integration interval is [ , The CEI is a broadband noise evaluation band (e.g., from 100 kHz to 800 kHz) that excludes the fundamental frequency and low harmonics of the ultrasonic transducer, while the denominator is the total energy of the entire band. The system monitors the CEI value in real time. If its average value is lower than the preset cavitation efficiency threshold within 1 second, the system automatically adjusts the output power or frequency of the ultrasonic generator to maintain the best cleaning state.

[0022] S213: A microfluidic sensing chip is integrated into the cleaning fluid circulation loop. This chip is etched with a four-electrode sensor for measuring conductivity, a pH sensor based on an ion-selective field-effect transistor (ISFET), a platinum resistance temperature sensor (PT100), and a small optical absorption spectrometer. The system acquires readings from these sensors every 5 seconds, forming a four-dimensional real-time fluid state vector. and the initial state vector By comparing and calculating the fluid degradation index, the chemical efficacy of the cleaning fluid is ensured to be within the effective range. In time Electrical conductivity, measured at specific times, reflects changes in the ion concentration in the cleaning solution. A decrease in conductivity may indicate the consumption or dilution of the active ingredient; an increase may indicate the accumulation of impurities or the formation of byproducts. In time The pH value (negative logarithm of hydrogen ion concentration) measured at any given time reflects changes in the acidity or alkalinity of the cleaning solution. Many cleaning solutions are pH sensitive, and pH shifts may indicate decreased chemical stability or the formation of reaction byproducts. In time The temperature measured at any given time reflects how temperature affects chemical reaction rates and sensor responses. Simultaneously, temperature drift may affect other sensor readings (such as conductivity and pH), therefore it needs to be compensated for or considered as a state variable. In time The absorbance value measured at specific times is a signal from a small optical absorption spectrometer, used to monitor the concentration of specific chemicals in the cleaning solution (such as dyes, oxidants, decomposition products, etc.). Absorption at different wavelengths can reflect the presence and changes of specific components; It is in time The four-dimensional real-time fluid state vector comprehensively reflects the current electrochemical and optical properties of the cleaning fluid.

[0023] As a further embodiment of the present invention, the specific process for calculating the surface residual contaminant index in steps S3 and S4 is as follows:

[0024] S311: After cleaning, the same scanning path and imaging parameters as in step S112 are used to acquire the completed image dataset; an image registration algorithm based on scale-invariant feature transform (SIFT) is applied to calculate the geometric transformation matrix between the reference image dataset and the completed image dataset to achieve precise alignment at the sub-pixel level.

[0025] S411: For each spectral band, the registered completed image is subtracted pixel by pixel from the reference image, and the absolute value is taken to generate a series of difference images; an adaptive threshold segmentation algorithm (such as the Otsu method) is applied to the difference images to extract the regions with pixel values ​​higher than the background noise level, and these regions are identified as residual pollutant patches.

[0026] S412: Calculate the total area of ​​all identified residual contaminant patches. and the area of ​​each patch. ,perimeter and average gray value Surface residual contaminant index ( It is calculated using the following formula:

[0027]

[0028] in It is the total surface area of ​​the probe. It is the total number of plaques. , , These are weighting coefficients corresponding to pollutant coverage, average residual thickness, and morphological dispersion, respectively, and their sum is 1.

[0029] As a further embodiment of the present invention, the specific process for calculating the comprehensive cleaning quality assessment score in step S4 is as follows:

[0030] S421: The calculated time-series integral values ​​of the Residual Surface Contaminant Index (RCI), the Cavity Cavitation Energy Index (CEI) during the ultrasonic immersion stage, and the norm of the fluid state vector deviating from the initial state are normalized to map their numerical ranges to the [0,1] interval. The normalized indices are expressed as follows: , ,and .

[0031] S422: The Comprehensive Cleaning Quality Score (CQS) is calculated using a weighted summation model, and its mathematical expression is as follows:

[0032]

[0033] in, The Residual Contamination Index (RCI) is a normalized surface residual contaminant index. Its source is the concentration of residual contaminants (such as particles, organic matter, ions, etc.) measured by sensors or detection methods. It is a time-series integral value representing the accumulated residual contaminant amount throughout the cleaning process (possibly an integration of continuous measurement data). Normalization involves mapping the original RCI value to the [0,1] interval, where 0 → no residue (ideal cleanliness) and 1 → severe residue (uncleaned or cleaning failed). (1- The item converts the contamination index into a cleanliness score, meaning the lower the RCI, the higher the cleanliness score. The normalized cavitation energy index, also known as the Cavitation Energy Index, is derived from the intensity of the cavitation effect generated by the sound field during the ultrasonic cleaning stage (i.e., the energy of bubble formation and collapse). The time-series integral value reflects the total accumulation of cavitation energy during the entire ultrasonic cleaning stage. Normalization is applied by mapping to [0,1], where 0 → no cavitation (ineffective cleaning) and 1 → maximum cavitation intensity (ideal physical cleaning conditions). The higher the intensity and efficiency values, which directly represent the physical cleaning process, the stronger the physical action and the greater the cleaning potential. The Fluid Degradation Index (FDI) is a normalized fluid state deviation index. It is derived from a state vector constructed using multiple parameters (conductivity, pH, temperature, and light absorption) collected by a microfluidic chip. and the initial state The degree of deviation obtained from the comparison; calculation method: for example, using Euclidean distance, Manhattan distance, or weighted distance:

[0034]

[0035] Then normalize to [0,1], 0 → fluid state unchanged (fresh), 1 → severely degraded (ineffective). The effectiveness score represents the cleaning solution's performance. Fresher fluid results in a lower FDI (Fee Dissolved Intake) and a higher effectiveness score. If the cleaning solution is depleted or deteriorated, even a strong process will not be effective. Weighting coefficients α, β, and γ are preset parameters reflecting the relative importance of surface cleanliness, cleaning process intensity, and cleaning medium quality in the final evaluation, satisfying α + β + γ = 1. For example, in an application scenario where final cleanliness is the highest priority, α could be set to 0.7, β to 0.2, and γ to 0.1.

[0036] This invention also provides an intelligent cleaning and monitoring system for ultrasonic imaging probes, used to perform the above method, the system comprising:

[0037] The probe identification and status recognition module integrates an RFID reader and a near-field communication module. It is used to read the probe's unique identification code and communicate with the background database to retrieve the probe's historical data and physical model parameters.

[0038] The multispectral imaging analysis module includes a high-resolution CMOS image sensor, a programmable light source system consisting of a multi-wavelength LED array, and a precision two-dimensional rotation and translation mechanism. It is used to scan the probe surface before and after cleaning, acquire a baseline image dataset and a finished image dataset, and calculate and generate a baseline contaminant spectrum and a surface residual contaminant index through the built-in image processing unit.

[0039] The cleaning operation and process monitoring module includes a sealed cleaning chamber, a high-pressure spray system, an ultrasonic generator and transducer, and a heating and disinfection component. The module further integrates a broadband hydrophone and a microfluidic chemical sensor chip for online monitoring of the acoustic cavity cavitation energy index and the real-time fluid state vector of the cleaning fluid during the cleaning process.

[0040] The data fusion and decision authentication module is an embedded computing unit that runs a multi-dimensional data fusion algorithm. This algorithm integrates various quantitative indicators from the multispectral imaging analysis module and the cleaning operation and process monitoring module to calculate the final cleaning quality assessment score. Based on the comparison of this score with a preset threshold, the algorithm determines whether the cleaning is qualified or unqualified and ultimately generates or rejects the generation of a digital cleaning certificate.

[0041] The data recording and traceability interface module is responsible for encrypting and timestamping the generated digital cleaning certificate, along with all raw data and analysis results from process monitoring, and writing it into a distributed ledger database based on blockchain technology to ensure data integrity and immutability. This module also provides an interface compliant with HL7 or DICOM standards for secure data exchange with hospital information systems (HIS) or medical image archiving and communication systems (PACS).

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] This invention, by introducing multispectral imaging technology and quantitative image analysis algorithms, transforms the detection of contaminants on the probe surface from subjective human judgment to objective and repeatable machine vision quantitative evaluation, greatly improving the accuracy and reliability of cleaning effect verification. Furthermore, this invention not only evaluates the cleaning results but also performs real-time quantitative monitoring and closed-loop feedback control of the cleaning process itself by online monitoring of the acoustic cavity cavitation effect and the chemical composition of the cleaning fluid, ensuring that the cleaning operation is always performed under optimal conditions. This system integrates probe identification, pre-cleaning status assessment, dynamic monitoring of the cleaning process, quantitative verification of cleaning results, and encrypted data storage and traceability into a fully automated closed-loop process. It generates a digital certificate containing physical, chemical, and acoustic evidence for each cleaning activity, fundamentally solving the long-standing core problems in the field of medical device cleaning and disinfection, such as opaque processes, unreliable results, and difficulty in tracing responsibility. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of the overall process of the intelligent cleaning and monitoring method for ultrasonic imaging probes provided in an embodiment of the present invention;

[0045] Figure 2 This is a schematic diagram of the functional modules of the intelligent cleaning and monitoring system for ultrasonic imaging probes provided in an embodiment of the present invention;

[0046] Figure 3 This is a detailed flowchart illustrating the steps of pre-cleaning the probe surface and performing scanning and analysis in an embodiment of the present invention.

[0047] Figure 4This is a flowchart illustrating the steps for online monitoring and data acquisition during the cleaning process in an embodiment of the present invention.

[0048] Figure 5 This is a flowchart illustrating the steps for quantitative evaluation and decision-making verification of cleaning effectiveness in an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0050] Please see Figure 1 The intelligent cleaning and monitoring method for ultrasonic imaging probes provided in this embodiment of the invention includes the following steps:

[0051] Step S1: When the cleaning process is started, the unique identification code of the ultrasonic imaging probe to be cleaned is obtained, and its historical cleaning data and probe physical model parameters are called based on the unique identification code; then the multispectral imaging unit is driven to perform a pre-cleaning surface scan of the probe, acquire a reference image dataset covering a preset spectral range, and extract the initial contaminant distribution characteristics of the probe surface through an image processing algorithm to generate a reference contaminant spectrum.

[0052] Step S2: Based on the reference contaminant spectrum, the cleaning operation execution unit is activated. The cleaning operation execution unit performs multi-stage physical and chemical cleaning operations, including high-pressure spraying, ultrasonic immersion cleaning, and high-temperature disinfection, in a closed cleaning chamber according to a preset cleaning protocol. During the ultrasonic immersion cleaning stage, the acoustic field signal of the liquid medium in the cleaning chamber is collected in real time by the acoustic sensing unit, and its spectrum is analyzed to calculate the acoustic cavity cavitation energy index, which characterizes the intensity of the cavitation effect. At the same time, the key chemical parameters of the cleaning fluid are monitored online by the chemical sensing unit to generate a real-time fluid state vector.

[0053] Step S3: After the cleaning operation is completed, drive the multispectral imaging unit to perform post-cleaning surface scanning on the probe, and acquire a finished image dataset with the same spectral range and imaging parameters as the reference image dataset; perform pixel-level registration between the finished image dataset and the reference image dataset, and calculate the difference between the two datasets after registration to generate a residual contaminant differential image.

[0054] Step S4: Perform image segmentation and morphological analysis on the differential image of residual pollutants to identify and quantify the patch area, grayscale integral value and shape complexity of residual pollutants. Calculate the surface residual pollutant index based on the preset pollutant quantification model. Fuse the surface residual pollutant index, the acoustic cavity cavitation energy index and the real-time fluid state vector, and generate a comprehensive cleaning quality assessment score through a multi-dimensional weighted fusion algorithm.

[0055] Step S5: Compare the comprehensive cleaning quality assessment score with the preset pass threshold, determine the cleaning result, and generate a digital cleaning certificate or trigger a new round of cleaning operations based on the determination result.

[0056] In one specific embodiment, the baseline contaminant map is a structured data object containing fields such as: initial contaminant coordinate matrix, contaminant spectral response curve, and initial contaminant load index. The acoustic cavity cavitation energy index is a scalar value, calculated based on ultrasonic dominant frequency energy, harmonic energy distribution, and broadband noise integral energy. The real-time fluid state vector is a multi-dimensional vector, with dimensions including the conductivity, pH value, temperature, and light absorbance at a specific wavelength of the cleaning fluid. The surface residual contaminant index is a comprehensive indicator, composed of contaminant coverage, estimated average residual thickness, and contaminant morphology dispersion. The digital cleaning certificate is an encrypted data packet containing the probe RFID / NFC serial number, cleaning start and end timestamps, pre-cleaning and post-cleaning image summaries, acoustic cavity cavitation energy index time series, fluid state vector historical records, final cleaning quality assessment score, and operator digital signature.

[0057] Please see Figure 3 This illustrates the specific process of generating the baseline pollutant map in step S1. In one embodiment, this process can be further broken down into the following sub-steps:

[0058] Step S111: Using a radio frequency identification (RFID) reader or near field communication (NFC) module, read the electronic tag attached to the probe handle or cable connector to obtain its globally unique probe serial number (US-ID). The US-ID serves as the primary key for indexing and querying in the device history database. The device history database is a relational database, and its table structure includes, but is not limited to, the following fields: us_id (VARCHAR, PRIMARYKEY), probe_model (VARCHAR), geometry_data (JSONB), material_spec (TEXT), last_cleaned_timestamp (TIMESTAMP), total_usage_cycles (INT), and last_cleaning_certificate_hash (VARCHAR).

[0059] Specifically, when an abdominal probe of model C5-2 (US-ID: PROBE-C52-SN789012) is placed into the cleaning equipment, the RFID reader operates at a frequency of 13.56 MHz and reads the US-ID stored in its electronic tag. The system uses this ID to query the database and retrieves the geometry_data field. This field stores a JSON object that details the probe's physical dimensions and scanning path planning parameters, for example: {"model":"C5-2","type":"convex","face_radius_mm":60,"body_length_mm":180,"scan_profile":{"rotation_axis":"longitudinal","rotation_total_angle_deg":360,"rotation_step_deg":5,"translation_axis":"longitudinal","translation_distance_mm":180,"translation_speed_mm_s":15}}. Simultaneously, the system retrieves the probe's historical cleaning records for subsequent trend analysis.

[0060] As an alternative, probe identification can be achieved using an industrial-grade QR code reader integrated inside the cleaning chamber. A DataMatrix QR code, encoding the US-ID, is etched onto the probe handle. Upon process startup, an auxiliary light source illuminates the QR code area, a camera captures the image, and the US-ID is deciphered using a software module based on an open-source decoding library (such as libdmtx). The limitations of this approach are that the QR code surface must be clean and unobstructed, and high positioning accuracy is required.

[0061] As an alternative, in case of RFID or QR code reading failure, the system can switch to manual input mode. In this mode, the touchscreen interface prompts the operator to select from a preset list of probe models and manually enter the serial number. To reduce input errors, the system incorporates a checksum algorithm specific to the manufacturer's serial number format. For example, after entering PROBE-C52-SN789012, the system will verify the format and checksum of the US-ID according to preset rules. If this US-ID does not exist in the database, the system will trigger a "new probe registration" process, guiding the operator to confirm the probe model and create a new database entry.

[0062] Step S112 involves driving a two-dimensional rotation and translation mechanism, built into the cleaning chamber and controlled by a stepper motor, to precisely control the movement of the probe within the field of view of the multispectral imaging unit. The two-dimensional rotation and translation mechanism includes a rotation axis and a linear translation axis. The rotation axis is controlled by a stepper motor with a 1.8-degree step angle and a 16x microstepping driver, achieving an angular resolution of 360 / (1.8 / 16) = 3200 microsteps / revolution. The translation axis is driven by another stepper motor via a ball screw, achieving a linear resolution of 0.01 mm / step. The multispectral imaging unit includes a back-illuminated CMOS image sensor with a resolution of 3840x2160 pixels and a ring light source system arranged around the probe array. The light source system consists of five independently controllable high-power LEDs, with center emission wavelengths of 365 nm (UV-A), 450 nm (blue light), 525 nm (green light), 630 nm (red light), and 940 nm (near-infrared light), respectively.

[0063] Specifically, based on the scan_profile data obtained in step S111, the system generates a motion command sequence for the C5-2 probe. The command sequence drives the rotary motor to rotate the probe in 5-degree increments, executing 72 steps to complete a 360-degree rotation. Between every two rotation steps, the translation axis uniformly translates the probe by 2.5 mm (180 mm / 72 steps) at a speed of 15 mm / s. This coordinated motion ensures that the entire cylindrical surface of the probe is completely covered by the imaging unit's helical scanning trajectory.

[0064] In step S113, at each scanning location, the system sequentially illuminates LEDs of different wavelengths and acquires a high-resolution image frame at each wavelength, forming a reference image dataset. The parameters of the acquisition sequence, such as the driving current of the LEDs at each wavelength and the exposure time of the CMOS sensor, are retrieved from a preset "imaging profile" based on the probe model and common contaminant types. For example, a profile for detecting biofilms and protein residues might be set as follows: a 365nm LED driven at 100% duty cycle with an exposure time of 200ms to maximize the signal-to-noise ratio of the fluorescence signal; while a 630nm red LED driven at 50% duty cycle with an exposure time of 20ms, used to acquire a reference image of the surface morphology. In a complete scanning cycle, if there are 72 scanning locations, the system will acquire 72*5=360 raw images, constituting the reference image dataset.

[0065] Subsequently, a semantic segmentation model based on a convolutional neural network (CNN) was applied to process the image dataset. The CNN model adopted a U-Net architecture, with its input layer receiving a 5-channel image tensor (each channel corresponding to a spectral band), and the spatial resolution was unified to 1024x1024 pixels after preprocessing. The network passed through an encoder-decoder structure, using skip connections to fuse multi-scale features, and finally output a single-channel probability map. The value of each pixel in the map (between 0 and 1) represents the probability that the pixel belongs to a contaminant. This model was trained offline on a dataset containing tens of thousands of probe images manually labeled by experts with various contaminants (such as coupling agent residue, bloodstains, gel, protein membranes, etc.). The loss function used during training was a weighted sum of Dice loss and binary cross-entropy loss to balance the requirements for small target detection and pixel-level classification accuracy.

[0066] Specifically, when processing a five-channel image frame, the probability map output by the CNN model is binarized using a fixed threshold (e.g., 0.5) to generate a contaminant mask image. The system then stitches together the mask images generated from all scan locations into a complete contaminant unfolded map of the probe surface through coordinate transformation. Finally, each independent connected component (i.e., contaminant patch) in the unfolded map is analyzed to extract its centroid coordinates, bounding rectangle, pixel area, and average pixel intensity value across the five original spectral channels. This structured information is encoded to generate a basic contaminant map.

[0067] The following is an example of the data structure for a baseline pollutant map:

[0068] JSON

[0069] 1{

[0070] 2"us_id":"PROBE-C52-SN789012",

[0071] 3"timestamp":"2024-05-21T09:30:15Z",

[0072] 4"initial_contaminant_load_index":0.76,

[0073] 5"contaminant_blobs":[

[0074] 6{

[0075] 7"blob_id":1,

[0076] 8"bounding_box":[1024,512,150,80],

[0077] 9"pixel_area":9870,

[0078] 10"spectral_signature":{

[0079] 11"uv_365nm_intensity":225.4,

[0080] 12"blue_450nm_intensity":98.1,

[0081] 13"green_525nm_intensity":110.5,

[0082] 14"red_630nm_intensity":105.2,

[0083] 15"ir_940nm_intensity":70.8

[0084] 16},

[0085] 17"classification_probability":{

[0086] 18"protein":0.92,

[0087] 19"gel":0.05,

[0088] 20"other":0.03

[0089] 21}

[0090] 22} 23]

[0092] twenty four}

[0093] Among them, initial_contaminant_load_index is a comprehensive pollution load index, which is calculated by weighting the total pollutant coverage area and the spectral characteristics of each patch.

[0094] Please see Figure 4 This illustrates the specific process of online monitoring and data acquisition for the cleaning process in step S2. In one embodiment, this process can be further broken down into the following sub-steps:

[0095] Step S211: One or more broadband hydrophones are symmetrically installed on the inner wall of the cleaning chamber. In a specific embodiment, two piezoelectric ceramic hydrophones with a center frequency of 2.5 MHz and a bandwidth of 3 MHz are used, respectively installed on two opposite side walls of the chamber to obtain more comprehensive sound field information. During the ultrasonic immersion stage, an ultrasonic generator drives multiple piezoelectric transducer arrays attached to the bottom of the chamber to operate at a fundamental frequency of 40 kHz. The hydrophones continuously acquire sound field pressure signals at a sampling rate of not less than 5 megasamples / second (MS / s). This sampling rate is selected based on the Nyquist sampling theorem to ensure that all signal components within the hydrophone's operating bandwidth can be acquired without distortion. The acquired digital signal stream is sent to a dedicated signal processing unit based on a field-programmable gate array (FPGA). The FPGA is chosen because its inherent parallel processing capability and low latency characteristics can meet the requirements for real-time, high-throughput spectral analysis of the sound field signal. The FPGA's internal logic is divided into a data acquisition interface module, a dual-port RAM cache module, and a hardware-implemented Fast Fourier Transform (FFT) core.

[0096] In step S212, the signal processing unit performs continuous short-time Fourier transform (STFT) on the acquired time-domain signal stream. Specifically, the FPGA divides the signal stream into frames, each containing 4096 sampling points, with 50% overlap between frames (i.e., 2048 sampling points). A Hamming window function is applied to each frame to suppress spectral leakage, and then the signal is fed into a 4096-point FFT core for transformation, continuously generating the sound field power spectrum with a time resolution of approximately 10 milliseconds. For each calculated power spectral density function P(f), the system calculates the cavity cavitation energy index (CEI). This index is defined as the ratio of energy within a specific broadband noise band to the total energy across the entire frequency band, and its mathematical expression is:

[0097]

[0098] The source and selection of parameters have clear physical meaning. Integration interval [ , The choice of frequency band [150kHz, 800kHz] aims to capture broadband noise generated by cavitation bubble collapse while avoiding direct interference from the ultrasonic transducer's fundamental frequency (40kHz) and its higher harmonics (80kHz, 120kHz, ...). In a typical configuration, this band is set to [150kHz, 800kHz]. The denominator... This corresponds to the upper limit of the effective bandwidth of the hydrophone, for example, 1.5 MHz.

[0099] The following simulation example illustrates the calculation of CEI: At a certain time t, the integrated energy of the power spectrum calculated by the FPGA is 1.52 watts across the entire frequency band [0, 1.5MHz]. The integrated energy within the broadband noise evaluation band [150kHz, 800kHz] is 0.58 watts. Therefore, the cavity cavitation energy index CEI at this moment is 0.58 / 1.52 ≈ 0.382. The system monitors the CEI value in real time. If the sliding average value of CEI within one second is lower than the preset cavitation efficiency threshold (e.g., 0.30), closed-loop feedback control is triggered. This threshold was determined experimentally during the system development phase and is associated with the optimal cleaning effect under a specific combination of cleaning agent and contaminants. The feedback control logic first attempts to increase the output power of the ultrasonic generator by 5% and monitors the change in CEI. If the CEI value does not increase significantly, the system will start a frequency scanning mode, sweeping the frequency in steps of 0.1kHz within a range of 40kHz ± 2kHz, aiming to break the standing waves in the sound field and activate the cavitation effect in inert regions. If the CEI remains below the threshold after three consecutive adjustments, the system will record a "low cavitation performance" warning and notify the upper control system.

[0100] Step S213: In the circulation loop of the cleaning fluid, a multi-parameter online monitoring module is integrated. In one embodiment, this module is a flow cell integrating a microfluidic sensor chip. Multiple sensors are etched onto this chip using microelectromechanical systems (MEMS) technology: a four-electrode platinum sensor for measuring conductivity, a pH sensor based on an ion-selective field-effect transistor (ISFET), a high-precision platinum resistance temperature sensor (PT100), and a small optical absorption spectrometer integrating a micro-grating and a linear photodiode array. The system polls and collects readings from these sensors at 5-second intervals, forming a four-dimensional real-time fluid state vector. ,in The absorbance is the light at a specific wavelength (e.g., 280 nm, used to detect protein dissolution).

[0101] Before the cleaning cycle begins, the system acquires the initial state vector V_0 of the cleaning solution. For example, the initial vector of a newly prepared alkaline cleaning solution might be... =[ =12.5 mS / cm, pH=11.2, T=25.1°C, =0.02AU]. At the 10th minute of the cleaning process, the collected real-time vector was... =[ =16.8 mS / cm, pH=10.6, T=55.3°C, =0.45AU]. The change in the vector reflects the chemical dynamics of the cleaning process: conductivity. The increase and light absorption The significant increase in pH indicates that contaminants such as salts and proteins have dissolved from the probe surface into the cleaning solution; the decrease in pH indicates the consumption of alkaline active ingredients.

[0102] The system calculates the Fluid Degradation Index (FDI) based on the difference between the real-time vector and the initial vector to quantify the performance degradation and contamination level of the cleaning fluid. FDI can be defined as the normalized vector change norm:

[0103]

[0104] in This is the normalized weight vector for each parameter, and its value is set according to the importance of different chemical parameters to the cleaning efficiency. If the FDI exceeds the preset degradation threshold (e.g., 0.85), the system will determine that the current cleaning solution has expired, and will forcibly lock it after the current cleaning task is completed, prohibiting its use in new cleaning cycles, and at the same time generate a maintenance alarm "Please replace the cleaning solution".

[0105] As an alternative, the online monitoring module can consist of discrete industrial-grade sensors. The cleaning fluid is pumped into a series-connected measuring line containing, in sequence, a ring-type inductive conductivity meter, a temperature-compensated glass electrode pH meter, a sheathed PT100 thermometer, and an external UV-Vis spectrophotometer with a quartz flow cell. This approach offers more robust and durable sensors that are easier to calibrate and replace individually; however, its drawbacks include higher system integration complexity, a larger dead volume in the tubing, and a relatively longer response time.

[0106] Please see Figure 5 This illustrates a detailed process for quantitative evaluation and decision verification of cleaning effectiveness in an embodiment of the present invention, covering the core operations of steps S3, S4, and S5. In a specific embodiment, the specific process of calculating the surface residual contaminant index in steps S3 and S4 can be further decomposed into the following sub-steps:

[0107] In step S311, after the cleaning operation is completed and the surface has been initially dried, the two-dimensional rotation and translation mechanism and the multispectral imaging unit are driven using the same scanning path, motion command sequence, light source parameters, and imaging parameters as in step S112 to perform a post-cleaning surface scan on the probe, acquiring a complete dataset of finished images. This strict parameter reproducibility is the foundation for ensuring the effectiveness of subsequent differential analysis. Subsequently, the system applies an image registration algorithm based on scale-invariant feature transform (SIFT) to precisely align each frame in the finished image dataset with its corresponding frame in the reference image dataset.

[0108] Specifically, for each image pair in the spectral band (e.g., a pre-cleaned 365 nm UV image and a post-cleaned 365 nm UV image), the SIFT algorithm first independently detects scale-space extrema as keypoints in both images. Then, a 128-dimensional local image descriptor is computed for each keypoint, which is invariant to illumination changes, rotation, and scale scaling. The system finds initial matching point pairs by calculating the Euclidean distance between the keypoint descriptors of the two images using a nearest neighbor matching strategy. To eliminate false matches due to surface changes (contaminant removal) or noise, the system employs the Random Sample Consensus (RANSAC) algorithm to estimate a robust geometric transformation model. The RANSAC algorithm iteratively selects a minimum subset (e.g., 4 pairs) from the matching point pairs, computes a perspective transformation matrix, and then uses this matrix to check all other matching point pairs, counting the number of inliers. After hundreds of iterations, the transformation matrix with the most inliers is selected as the final transformation model. Finally, the system applies this transformation matrix to the completed image, generating a resampled new image that is precisely aligned to the reference image at the sub-pixel level using bilinear or bicubic interpolation. This process is performed independently for each of the five spectral bands.

[0109] In step S411, for each spectral band, the grayscale value of the completed image registered in step S311 is subtracted pixel by pixel from the corresponding reference image, and the absolute value is taken to generate a series (five images in this embodiment) of difference images. The pixel values ​​in the difference images directly reflect the changes in surface reflectance or fluorescence characteristics before and after cleaning, with high-value areas corresponding to successfully removed contaminants. Subsequently, in order to accurately separate the signal of residual contaminants from the background noise, the system applies an adaptive threshold segmentation algorithm to each difference image. In a preferred embodiment, Otsu's method is used. This algorithm automatically finds an optimal global threshold by maximizing the inter-class variance, dividing image pixels into foreground (residual contaminants) and background categories.

[0110] Specifically, for differential images in the ultraviolet band, their gray-level histograms may exhibit a bimodal distribution. One peak corresponds to background noise and unchanged areas, while the other peak corresponds to areas where the fluorescence signal is significantly weakened after contaminant removal. The Otsu algorithm calculates the gray-level value that maximizes the variance between these two categories, for example, T=42. All regions with pixel values ​​higher than this threshold T are initially identified as areas with removed contaminants. Conversely, if the original post-cleaned image is analyzed, the identification logic for residual contaminant regions is reversed. To further improve segmentation accuracy, the system can perform a morphological post-processing operation. For example, a morphological opening operation (erosion followed by dilation) is applied to a 3x3 pixel structuring element to eliminate isolated noise pixels, followed by a morphological closing operation (dilation followed by erosion) to fill any small pores that may exist within the residual contaminant patches.

[0111] In step S412, after obtaining the binary mask images of residual pollutants for all spectral bands, the system performs a logical OR operation on these mask images to fuse them into a final residual pollutant distribution map. Subsequently, all independent connected components (i.e., residual pollutant patches) in this distribution map are quantitatively analyzed. The system calculates the geometric and photometric properties of each patch, including its pixel area. ,perimeter and the corresponding average gray value in each difference image. Based on these measurements, the Residual Contaminant Index (RCI) is calculated. This index is a dimensionless comprehensive score, and its calculation formula is defined as follows:

[0112]

[0113] The three components of this formula each have a definite physical meaning. The first term,

[0114]

[0115] Represents pollutant coverage, of which It is the total area of ​​all remaining patches. This is the total surface area of ​​the probe (obtained from the physical model parameters acquired in step S111). This item reflects the extent of the contamination.

[0116] The second item,

[0117]

[0118] The area-weighted average gray value representing a contaminant patch can be considered an estimate of the average thickness or density of residual contaminants. Patches with higher gray values ​​in the difference image indicate more severe initial contamination before removal, and their remaining components may be more hazardous. This reflects the depth of contamination.

[0119] The third item,

[0120]

[0121] This represents the dispersion or irregularity of pollutant morphology. Among them,

[0122]

[0123] This is the isoperimetric quotient of a single plaque, with a value in the range (0,1], where 1 represents a perfect circle. Irregular plaques with complex edges (such as biofilm colonies) have lower isoperimetric quotient values. This term, in the form of 1-..., assigns a higher penalty value to plaques with more irregular shapes, because such residues are more difficult to remove and more likely to harbor microorganisms. N is the total number of plaques.

[0124] , , These are weighted coefficients corresponding to biofilm coverage, average residual thickness, and morphological dispersion, respectively, and their sum is 1. These weights are set based on risk assessment; for example, for scenarios where biofilm residue is of particular concern, It can be assigned a higher value (such as 0.5).

[0125] In a specific calculation example, let's assume the analysis yields the following: total probe surface area =12000mm², a total of N=3 residual patches were found.

[0126] Patch 1: =10mm², P1=15mm, I_avg1=150.

[0127] Patch 2: =5mm², P2=10mm, I_avg2=120.

[0128] Patch 3: =2mm², P3=8mm, I_avg3=200.

[0129] Let the weight be =0.4, =0.4, =0.2.

[0130] Total residual area =10+5+2=17mm².

[0131] Coverage rate = 0.4 * (17 / 12000) ≈ 0.00057.

[0132] Thickness term = 0.4 * ((10150 + 5120 + 2200) / 17) = 0.4 * (2500 / 17) ≈ 58.82. For ease of fusion, this term needs to be normalized. Assuming its maximum possible value is 1000, the normalized value is 58.82 / 1000 = 0.05882.

[0133] Calculation of roundness of shape:

[0134] =4π10 / 15²≈0.558

[0135] =4π5 / 10²≈0.628

[0136] =4π² / 8²≈0.393

[0137] Average roundness = (0.558 + 0.628 + 0.393) / 3 ≈ 0.526

[0138] Morphological term = 0.2 * (1 - 0.526) = 0.0948.

[0139] The final RCI (sum of all components before normalization) = 0.00057 + 0.05882 + 0.0948 ≈ 0.154. This value will be used for the next step of normalization and fusion calculation.

[0140] As a follow-up to step S4, the specific process of calculating the comprehensive cleaning quality assessment score can be further broken down into the following sub-steps:

[0141] Step S421 involves normalizing the surface residual contaminant index (RCI) calculated in the previous step, the time-series integral value of the cavity cavitation energy index (CEI) recorded in step S212 throughout the ultrasonic immersion stage, and the final fluid degradation index (FDI) calculated in step S213. The purpose of normalization is to map these indices from different sources and with different dimensions to a unified [0,1] interval for fair weighted fusion.

[0142] Specifically, normalization employs the min-max scaling method. For RCI, the normalization formula is as follows: =(RCI-RCI_min) / (RCI_max-RCI_min). Where RCI_min is theoretically 0 (completely clean), and RCI_max is a preset upper limit representing an unacceptable level of contamination, set through experimental data statistics or based on clinical risk standards. The time-series integral value of CEI (denoted as CEI_int) represents the total acoustic energy input during the entire cleaning process, and its normalization formula is: =(CEI_int-CEI_min_int) / (CEI_max_int-CEI_min_int). CEI_min_int and CEI_max_int represent the integral value ranges corresponding to ineffective and optimal cleaning processes, respectively. For the fluid degradation index (FDI), which is already partially normalized, it can be used directly or scaled again to obtain... For example, if the effective range of FDI is [0, 1.2], then =FDI / 1.2.

[0143] Step S422: The final Comprehensive Quality Score (CQS) is calculated using a multidimensional weighted summation model. This model integrates all information from outcome assessment (surface cleanliness) and process monitoring (intensity of physical action, quality of chemical media). Its mathematical expression is:

[0144]

[0145] In this expression, (1- The item converts the contamination index into a cleanliness score, meaning the lower the RCI, the higher the cleanliness score. This directly represents the quality of the physical cleaning process. (1- The FDI (Failure to Degrade) item converts the degree of degradation of the cleaning fluid into its effectiveness score; that is, the higher the FDI, the lower the effectiveness score. The weighting coefficients α, β, and γ are preset parameters that can be adjusted through the system configuration interface, and their sum is 1 (α + β + γ = 1). These weights reflect the relative importance of different dimensions in the final quality assessment.

[0146] For example, in a cleaning scenario for high-risk interventional ultrasound probes (such as transesophageal echocardiography probes), the final surface cleanliness is of primary concern; therefore, the weights can be set as α=0.7, β=0.2, γ=0.1. Assuming that after normalization, we obtain... =0.08, =0.85, =0.20. Therefore, the final CQS calculation is as follows:

[0147] CQS=0.7*(1-0.08)+0.2*0.85+0.1*(1-0.20)

[0148] CQS=0.7*0.92+0.2*0.85+0.1*0.80

[0149] CQS = 0.644 + 0.170 + 0.080 = 0.894

[0150] This score (89.4%) will be used for the next step of qualification determination.

[0151] Step S5: The calculated Comprehensive Cleaning Quality Assessment Score (CQS) is compared with a preset Qualification Threshold. This threshold is set according to relevant medical device cleaning and disinfection specifications, manufacturer guidelines, and hospital infection control department requirements; for example, it can be set to 0.90.

[0152] If the CQS is higher than or equal to the passing threshold (e.g., a calculated CQS of 0.91 ≥ 0.90), the system determines that the cleaning is qualified. Subsequently, the system automatically generates a digital cleaning certificate containing comprehensive traceability information. This certificate is a structured data file in JSON or XML format, and its content includes at least: the probe's unique identification code (US-ID), precise timestamps of the start and end of the cleaning (compliant with ISO 8601 standards), a summary of the pre-cleaning and post-cleaning multispectral image sets (e.g., thumbnails of key areas or SHA-256 hashes of the entire dataset), complete time-series data of the cavity cavitation energy index (CEI), historical records of the fluid state vector (V_t), and scores of each component constituting the CQS (…). , , The final CQS total score, and the system ID and digital signature (if any) of the operator who performed the cleanup.

[0153] To ensure the immutability and trustworthiness of the certificate, the system encrypts the generated certificate data and binds its hash value to the probe's US-ID, writing it as a new entry into a database based on a private blockchain or distributed ledger technology (DLT). Each scrubbing event constitutes a block containing the certificate hash and is linked to the previous block (the probe's last scrubbing record), forming an immutable and auditable traceability chain.

[0154] If the CQS is below the acceptable threshold (e.g., a calculated CQS of 0.894 < 0.90), the system determines that the cleaning is unacceptable. In this case, the system will automatically trigger a warning and display an analysis of the reason for the failure on the user interface, indicating whether it is due to "excessive residual contaminants on the surface (low RCI score)" or "insufficient ultrasonic cavitation efficiency (low CEI score)." Based on this analysis, the system automatically enters re-cleaning mode. In re-cleaning mode, the system can intelligently adjust the cleaning protocol. For example, if the failure is due to excessively high RCI, the system may extend the high-pressure spray stage or increase the concentration of the cleaning agent; if the failure is due to insufficient CEI, the system will use stronger ultrasonic power or a wider sweep frequency range. After re-cleaning is completed, the system will repeat steps S3 to S5 for a new round of evaluation. All unacceptable cleaning attempts and their related monitoring data will be recorded in detail in the log for long-term quality control and process optimization analysis.

[0155] Please see Figure 2 The present invention also provides an intelligent cleaning and monitoring system for ultrasonic imaging probes, which serves as the physical carrier for implementing the above-mentioned method and includes the following functional modules:

[0156] The probe identification and status recognition module physically integrates a 13.56MHz RFID / NFC reader and a high-resolution industrial camera for QR code reading, and is installed at the inlet of the cleaning chamber. The embedded controller of this module executes the logic of step S111, automatically acquiring the probe's identity and its historical data, physical model parameters (such as geometry_data), through network communication (e.g., via TCP / IP protocol) with the background device history database. (Instantaneous invocation of )

[0157] The multispectral imaging analysis module serves as the system's "eyes." At its core are a precision two-dimensional rotation and translation mechanism, a high-resolution CMOS image sensor, and a multi-wavelength LED ring light source system. The module incorporates a high-performance embedded computing unit, such as a system-on-a-chip (SoC) with a GPU. This computing unit is responsible for performing all image processing-related tasks, including controlling scanning motion (step S112), image acquisition sequence (step S113), CNN-based contaminant segmentation (step S113), SIFT-based image registration (step S311), differential image calculation, and the final calculation of the residual contaminant index (RCI) (steps S411 and S412).

[0158] The cleaning operation and process monitoring module serves as the system's "actuator" and "nervous system." This module includes a sealed cleaning chamber, a water pump and precision nozzle array for high-pressure spraying, an ultrasonic generator and multiple piezoelectric transducers, and a heating assembly for high-temperature sterilization. Furthermore, this module integrates process monitoring sensors, namely a broadband hydrophone and a microfluidic chemical sensor chip (or a discrete sensor array). A dedicated FPGA board is connected to the hydrophone, responsible for performing real-time FFT and calculating the cavity cavitation energy index (CEI) (steps S211 and S212). A microcontroller (MCU) polls the chemical sensor readings, generates a real-time fluid state vector (V_t), and calculates the fluid degradation index (FDI) (step S213). This module also performs closed-loop feedback control, such as adjusting the power and frequency of the ultrasonic generator based on the CEI value.

[0159] The data fusion and decision authentication module is the "brain" of the system. This module is typically an industrial-grade computer running a real-time operating system (RTOS). As a central coordinator, it receives RCI values ​​from the multispectral imaging analysis module and the final CEI and FDI values ​​from the cleaning operation and process monitoring module. The core software of this module is the data fusion and decision algorithm, responsible for executing the normalization process in step S421, the CQS weighted calculation in step S422, and the conformity determination in step S5. Based on the determination result, it issues instructions to other modules, such as commands to generate certificates or to start a re-cleaning cycle.

[0160] The data recording and traceability interface module acts as the system's "archive" and "diplomat." This module is responsible for generating, encrypting, and securely storing the digital certificates described in step S5. Internally, it runs lightweight blockchain node software that packages the hash value of the cleaning certificate into a transaction and submits it to the hospital's private distributed ledger network. Furthermore, this module provides standardized data exchange interfaces. For example, an interface compliant with the HL7 (HealthLevelSeven) standard can push cleaning-qualified event messages (containing probe ID, timestamp, and certificate hash) to the Hospital Information System (HIS), or an interface compliant with the DICOM (Digital Imaging and Communications in Medicine) standard can associate the cleaning certificate as a non-image object with the probe and store it in the Picture Archiving and Communication System (PACS). This allows clinicians to simultaneously view the latest, reliable cleaning and disinfection records of the probes used when reviewing patient images, forming a complete closed-loop traceability.

[0161] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may use the above-disclosed technical content to make changes or modifications to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An intelligent cleaning and monitoring method for ultrasonic imaging probes, characterized in that, Includes the following steps: S1: Obtain the unique identification code of the ultrasonic imaging probe to be cleaned, call its historical cleaning data and physical model parameters, drive the multispectral imaging unit to perform pre-cleaning surface scanning of the probe, collect the reference image dataset, extract the initial contaminant distribution characteristics on the probe surface, and generate the reference contaminant spectrum. S2: Based on the benchmark contaminant spectrum, activate the cleaning operation execution unit to perform multi-stage cleaning operations. During the multi-stage cleaning operations, the acoustic field signal of the liquid medium is collected and analyzed by the acoustic sensing unit, the acoustic cavity cavitation energy index is calculated, and the key chemical parameters of the cleaning fluid are monitored by the chemical sensing unit to generate a cleaning process status dataset. S3: After the cleaning operation is completed, drive the multispectral imaging unit to scan the probe surface after cleaning, collect the completed image dataset with the same parameters as the reference image dataset, perform pixel-level registration and calculate the difference to generate a residual contaminant differential image. S4: Based on the differential image of the residual contaminants, perform image segmentation and morphological analysis, quantify the features of the residual contaminants to calculate the surface residual contaminant index, and fuse the surface residual contaminant index with the cleaning process status dataset. Calculate the comprehensive cleaning quality assessment score through a multi-dimensional weighted algorithm. The baseline contaminant map includes an initial contaminant coordinate matrix, a contaminant spectral response curve, and an initial contaminant load index. The cleaning process state dataset includes the acoustic cavity cavitation energy index and a real-time fluid state vector. The surface residual contaminant index includes contaminant coverage, an estimated average residual thickness, and contaminant morphology dispersion. It also includes step S5: S5: Compare the comprehensive cleaning quality assessment score with the preset pass threshold. If the score is higher than the threshold, the cleaning is deemed qualified, a digital cleaning certificate is generated and it is bound to the probe's unique identification code for storage. If the score is below the threshold, the cleaning is deemed unqualified, triggering a new round of cleaning operations; The digital cleaning certificate includes a unique identifier for the probe, cleaning start and end timestamps, pre-cleaning and post-cleaning image summaries, a time series of acoustic cavity cavitation energy index, a historical record of fluid state vectors, a final cleaning quality assessment score, and the operator's digital signature.

2. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 1, characterized in that, The specific steps for generating the baseline pollutant map are as follows: S111: Read the electronic tag of the probe through an RFID reader or near-field communication module to obtain its unique identification code, and use this as an index to retrieve the probe's model, geometric dimension data and historical records in the device's historical database; S112: Drives a two-dimensional rotation and translation mechanism controlled by a stepper motor to precisely control the probe to move along a preset trajectory within the field of view of the multispectral imaging unit. The multispectral imaging unit includes a high-resolution image sensor and a ring light source system composed of multi-wavelength light-emitting diodes. S113: Light sources of different wavelengths are lit sequentially and high-resolution images are acquired at each wavelength to form a benchmark image dataset. The dataset is processed by a semantic segmentation model based on a convolutional neural network to identify and segment pollutant regions, encode their coordinates, area and spectral intensity, and generate a structured benchmark pollutant map.

3. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 2, characterized in that, The specific steps for generating the cleaning process status dataset are as follows: S211: A broadband hydrophone installed in the cleaning chamber continuously acquires the acoustic pressure time-domain signal during the ultrasonic immersion stage at a preset sampling rate, and sends it to a signal processing unit implemented by a field-programmable gate array. S212: The signal processing unit performs a short-time Fourier transform on the time-domain signal to generate a sound field spectrum. By calculating the ratio of energy in the broadband noise evaluation band to the total energy of the entire band, the cavitation energy index of the acoustic cavity is obtained. The conductivity, pH value, temperature and light absorbance readings are collected by a microfluidic sensor chip integrated in the cleaning fluid circulation loop to form the real-time fluid state vector.

4. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 3, characterized in that, The specific steps for generating the differential image of the residual pollutants are as follows: S311: Using the same scanning path and imaging parameters as the pre-cleaning scan, the completed image dataset is acquired, and an image registration algorithm based on scale-invariant feature transformation is applied to calculate the geometric transformation matrix between the reference image dataset and the completed image dataset to achieve sub-pixel level alignment; S312: For each spectral band, the gray values ​​of the registered completed image and the reference image are subtracted pixel by pixel and the absolute value is taken. An adaptive threshold segmentation algorithm is applied to extract the region with a noise level higher than the background noise level to generate the residual pollutant differential image.

5. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 4, characterized in that, The specific steps for calculating the surface residual contaminant index are as follows: S411: For all residual pollutant patches identified in the residual pollutant differential image, calculate their total area, as well as the area, perimeter, and average gray value of each patch. S412: The surface residual pollutant index is obtained by weighted summation based on the pollutant coverage, the average residual thickness estimated from the average gray value, and the morphological dispersion calculated from the area-to-perimeter ratio.

6. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 5, characterized in that, The calculation steps for the comprehensive cleaning quality assessment score are as follows: S421: Normalize the calculated time series integral values ​​of the surface residual contaminant index, the acoustic cavity cavitation energy index, and the norm of the real-time fluid state vector deviating from the initial state, so that their numerical ranges are mapped to a unified interval. S422: The comprehensive cleaning quality assessment score is calculated by weighting and fusing the normalized surface residual contaminant index, the integral value of the acoustic cavity cavitation energy index, and the fluid state vector deviation norm through a weighted summation model.

7. The intelligent cleaning and monitoring method for ultrasonic imaging probes according to claim 1, characterized in that, The S5 step is specifically as follows: S511: Call the comprehensive cleaning quality assessment score and compare its value with the qualified score threshold stored in the system configuration; S512: When the evaluation score is greater than or equal to the qualified threshold, the cleaning is deemed qualified, a digital cleaning certificate containing full-process monitoring data is generated, and the certificate is encrypted and signed, associated with the unique identification code, and written into an immutable distributed database. S513: When the evaluation score is less than the qualified threshold, the cleaning is deemed unqualified, and an instruction is sent to the cleaning operation execution unit to automatically start a new round of cleaning operation process.

8. An intelligent cleaning and monitoring system for ultrasonic imaging probes, characterized in that, The system is used to implement the intelligent cleaning and monitoring method for ultrasonic imaging probes according to any one of claims 1-7, and the system includes: The initial state assessment module is used to obtain the unique identification code of the ultrasonic imaging probe to be cleaned, drive the multispectral imaging unit to perform pre-cleaning scanning, collect and analyze the benchmark image dataset, and generate a benchmark contaminant spectrum. The cleaning process monitoring module is used to perform multi-stage physical and chemical cleaning operations, and to collect and analyze sound field signals and cleaning fluid parameters online through acoustic and chemical sensing units to generate a cleaning process status dataset. The cleaning effect imaging module is used to drive the multispectral imaging unit to perform post-cleaning scanning after the cleaning operation is completed, acquire the completed image dataset, and perform registration and difference calculation with the reference image dataset to generate a differential image of residual pollutants. The cleaning quality assessment module is used to analyze the differential image of the residual contaminants to calculate the surface residual contaminant index, and to fuse the index with the cleaning process status dataset to calculate and generate a comprehensive cleaning quality assessment score. The decision authentication and evidence storage module is used to compare the comprehensive cleaning quality assessment score with the pass threshold, and determine whether it is qualified to generate and store a digital cleaning certificate, or determine whether it is unqualified to trigger a new round of cleaning operations.

Images