CN-121987247-A - Orthopedic joint rehabilitation intelligent monitoring system and method based on ultrasonic detection

CN121987247ACN 121987247 ACN121987247 ACN 121987247ACN-121987247-A

Abstract

The invention discloses an intelligent monitoring system and method for rehabilitation of an orthopedic joint based on ultrasonic detection, and relates to the technical field of ultrasonic detection.A collection period of ultrasonic detection equipment is set by installing ultrasonic detection equipment, an ultrasonic image of the orthopedic joint is collected in real time based on the set collection period, and meanwhile, the collected ultrasonic image of the orthopedic joint is processed by an image processing method based on the ultrasonic image of the orthopedic joint; after the collected ultrasonic images of the orthopaedics joint are processed, the collected ultrasonic images of the orthopaedics joint are collected for a plurality of periods, training comparison is carried out on the collected ultrasonic images of the orthopaedics joint, the repair degree is judged, finally, based on the repair degree judged in real time, the ultrasonic images collected in real time are subjected to repair prediction in a digital twin mode, and the accuracy of rehabilitation monitoring of the orthopaedics joint is improved.

Inventors

ZHAO LIANG
ZHAO SHIQI
Liao Zheting
WANG ZIQI
CHEN YUFAN
WU DESHENG
Qiu Gengtao

Assignees

广州医科大学附属第一医院（广州呼吸中心）

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. An intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection is characterized by comprising the following steps: S1, installing ultrasonic detection equipment, setting an acquisition period for installing the ultrasonic detection equipment, and acquiring an ultrasonic image of an orthopedic joint in real time based on the set acquisition period; S2, acquiring an ultrasonic image of the orthopedics joint in real time, and processing the acquired ultrasonic image of the orthopedics joint by an image processing method to obtain a processed ultrasonic image of the orthopedics joint; s21, filtering and denoising the acquired ultrasonic image of the orthopaedics joint by an ultrasonic filtering method to obtain the ultrasonic image of the orthopaedics joint after filtering and denoising; S22, performing image enhancement on the obtained filtered and noise-reduced orthopedic joint ultrasonic image to obtain an enhanced orthopedic joint ultrasonic image; s23, performing background separation on the enhanced orthopedic joint ultrasonic image to obtain a processed orthopedic joint ultrasonic image; S3, summarizing the orthopedics joint ultrasonic images processed in a plurality of periods, training and comparing the summarized orthopedics joint ultrasonic images, and judging the repair degree; S4, based on the repair degree judged in real time, performing repair prediction on the ultrasonic image acquired in real time in a digital twin mode.
2. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 1, wherein the steps of installing ultrasonic detection equipment, setting an acquisition period for installing the ultrasonic detection equipment, and acquiring the ultrasonic image of the orthopedic joint in real time based on the set acquisition period comprise the following steps: Presetting the positions of all reference points on the orthopaedics joint, and fixing the relative positions of the ultrasonic detection equipment and the orthopaedics joint, so that the reference points on the orthopaedics joint are uniformly distributed in the field of view of the whole ultrasonic detection equipment; Setting acquisition periods, and acquiring an orthopedic joint appearance image shot by laser ultrasonic detection equipment in each acquisition period, wherein an angle of the ultrasonic detection equipment right facing the orthopedic joint is used as a standard angle, and an image acquired under the standard angle is used as a standard image; And combining the angle of the ultrasonic detection equipment with the corresponding acquired orthopedic joint appearance image, and establishing a training data set.
3. The intelligent monitoring method for rehabilitation of the orthopedic joint based on ultrasonic detection according to claim 1, wherein the step of filtering and denoising the acquired ultrasonic image of the orthopedic joint by an ultrasonic filtering method to obtain the ultrasonic image of the orthopedic joint after filtering and denoising comprises the following steps: the acquired ultrasonic image of the orthopaedics joint is subjected to filtering treatment through a filtering algorithm; Setting that a window formed by the collected ultrasonic images of the orthopaedics joint is , Represent the first Ultrasound data for each window; ; Wherein, the The window size is indicated and, Represent the first Ultrasound data for a window of time, Representing the filtered first Ultrasound data for a window of time, Indicating the intensity of the ultrasonic signal set up, Representing the intensity of the ultrasound signal collected per unit time.
4. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 1, wherein the image enhancement of the obtained filtered noise-reduced ultrasonic image of the orthopedic joint is carried out, and the obtained enhanced ultrasonic image of the orthopedic joint comprises the following steps: image enhancement is carried out on the orthopedic joint ultrasonic image after filtering and noise reduction through an automatic color balance algorithm; The steps of the automatic color equalization algorithm are as follows; S221, performing color difference correction on the filtered and noise-reduced orthopedic joint ultrasonic image to obtain a corrected reconstructed image; The color difference correction formula is as follows: ; Wherein, the Representing pixel points Is used for correcting the correction result of the (a), Representing pixel points And pixel point Is used for the brightness difference of the (c), The brightness is indicated as such, Representing pixel points And pixel point A function of the distance measure between them, The luminance performance function is represented as such, Representing the subset; s222, linearly expanding the corrected reconstructed image to obtain an enhanced orthopedic joint ultrasonic image.
5. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 4, wherein the linear expansion of the corrected reconstructed image to obtain the enhanced ultrasonic image of the orthopedic joint comprises the following steps: The linear expansion formula is as follows: ; Wherein, the Representing linear expansion Is a slope representing a linear expansion function, Representing pixel values; Enhancing the expanded image to obtain an enhanced orthopedic joint ultrasonic image; ; Wherein, the Representing the enhanced ultrasound image of the orthopaedic joint, Representing the maximum and minimum values of the pixel values, respectively.
6. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 1, wherein the background separation is performed on the enhanced ultrasonic image of the orthopedic joint, and the obtained ultrasonic image of the orthopedic joint after processing comprises the following steps: S231, performing binarization processing on the reinforced orthopedic joint ultrasonic image to obtain a binarized orthopedic joint ultrasonic image; selecting an initial gray threshold Dividing all pixels in the determined ultrasonic image of the orthopaedics joint into two types And ; Setting up Is less than or equal to the gray threshold value Is used for the display of the display panel, Is greater than the gray threshold Is a pixel class of (2); Setting pixel class Is the gray level average value of (2) Pixel class Is the gray level average value of (2) The global gray average value is ; Setting the pixels belonging to the pixel category in the determined orthopaedics joint ultrasonic image The probability of (2) is Belongs to the pixel category The probability of (2) is ; The binarization processing formula is as follows: ; ; Wherein, the Representing a binarization threshold; setting a gray value larger than the binarization threshold value to 255 and setting a gray value smaller than or equal to the binarization threshold value to 0; summarizing the orthopedic joint ultrasonic image after binarization treatment; S232, performing contour extraction on the bone joint ultrasonic image subjected to binarization processing to obtain a bone joint ultrasonic image subjected to background separation.
7. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 6, wherein the contour extraction is performed on the binarized orthopedic joint ultrasonic image, and the step of obtaining the orthopedic joint ultrasonic image with separated background comprises the following steps: establishing an array, recording gray values of 8 neighborhoods around each pixel point of the orthopedic joint ultrasonic image after binarization treatment, setting the pixel point to be in the interior of the outline when the gray values of 8 neighborhoods around the pixel point are the same as the gray values of the central point, and deleting the pixel point; When the gray values of 8 neighborhoods around the pixel point are different from the gray value of the central point, setting the pixel point to be at the contour edge, and reserving the pixel point; Traversing and summarizing each pixel point in the orthopedic joint ultrasonic image after binarization processing to obtain an orthopedic joint ultrasonic image with separated background; Setting the orthopedic joint ultrasonic image with separated background as the processed orthopedic joint ultrasonic image.
8. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 1, wherein the step of summarizing the orthopedic joint ultrasonic images processed in a plurality of periods, performing training comparison on the summarized orthopedic joint ultrasonic images, and judging the repair degree comprises the following steps: S31, dividing the collected orthopaedics joint ultrasonic image into Z small area image data blocks, and taking the Z small area image data blocks as input of a convolutional neural network; The convolutional neural network comprises a convolutional layer, a pooling layer and a full-connection layer; S32, setting the size of a convolution kernel based on the received image data blocks, moving the convolution kernel on the input image data blocks according to the set step length, and simultaneously carrying out convolution calculation on the characteristic value of the corresponding area of each step in the moving process so as to realize the characteristic extraction of each image data block; The convolution calculation formula is as follows: ; Wherein, the Representing the block of image data being input, The weight value of the corresponding convolution kernel is represented, b represents the offset value, and F represents the output characteristic; s33, convoluting the characteristics of the input image data block through the continuously stacked convolution layer and pooling layer, and transmitting the characteristics of the convolved image data block into the full connection layer; s34, unfolding and combining the data through the full connection layer to obtain a feature vector, and storing the feature vector; S35, judging the similarity degree of the adjacent periodic feature vectors in a data comparison mode; Setting a similarity fluctuation threshold, wherein when the similarity fluctuation state value of the adjacent period feature vector is larger than the set similarity fluctuation threshold, the similarity fluctuation threshold is set to indicate that the recovery state is good, and otherwise, the recovery state is abnormal.
9. The intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection according to claim 1, wherein the repairing prediction of the ultrasonic image acquired in real time by a digital twin mode based on the repairing degree judged in real time comprises the following steps: S41, setting digital twin prediction weights based on the repair degree judged in real time; The digital twin prediction weights are set as follows: ; Wherein, the The behavior is represented by a pattern of the behavior, The state of the display is indicated and, Representing a set of all of the actions, The weight function is represented by a function of the weight, The execution behavior is represented by a representation of the execution behavior, Is shown in the state Time execution behavior ; S42, calculating a prediction result of the repair degree based on the set digital twin prediction weight; The digital twin predictions are as follows: ; Wherein, the The probability of a digital twin prediction is represented, Representing a state; and carrying out dynamic restoration prediction on the ultrasonic image acquired in real time by predicting the probability obtained by calculation of each behavior.
10. A system for realizing the ultrasonic detection-based intelligent monitoring method for rehabilitation of the orthopedic joint according to any one of claims 1 to 9, which is characterized by comprising a data acquisition module, an image processing module, an image training module, a repair degree judging module and a repair degree predicting module; the data acquisition module is used for acquiring ultrasonic images of the orthopaedics joint in real time; The image processing module is used for processing the ultrasonic image of the orthopaedics joint acquired in real time to obtain a processed ultrasonic image of the orthopaedics joint; The image training module is used for training and comparing the processed orthopedics joint ultrasonic images; The repair degree judging module is used for judging the repair degree according to training contrast of the ultrasonic image of the orthopedics joint; the repair degree prediction module is used for predicting the repair state of the orthopedic joint in a digital twin mode.

Description

Orthopedic joint rehabilitation intelligent monitoring system and method based on ultrasonic detection Technical Field The invention relates to the technical field of ultrasonic detection, in particular to an intelligent monitoring system and method for rehabilitation of an orthopedic joint based on ultrasonic detection. Background At present, the identification, evaluation and analysis of the ultrasonic image are carried out by subjective judgment of technicians, but the subjective judgment of the technicians often has the problems of strong subjectivity, low efficiency and the like. And with the increase of ultrasonic images, subjective judgment of technicians is gradually unable to meet actual demands. The ultrasonic image detection method suitable for the internal injury of the high-speed track is disclosed in the patent application with the publication number of CN109212033A in the prior art, and comprises an ultrasonic image processing step and an ultrasonic image recognition step, wherein the ultrasonic image processing step comprises a sub-channel image normalization mode for carrying out standardized processing on images, and meanwhile, the ultrasonic image of the internal injury of the high-speed track is divided and divided in a mode of solving the number and the positions of auxiliary lines and dividing the ROI region into blocks, and the ultrasonic image recognition step comprises the steps of judging the type of the injury, positioning the position of the injury and calculating the characteristic parameters of the injury. According to the scheme, the existing ultrasonic detection method is mainly aimed at the identification and analysis of the damaged area, and lacks of real-time monitoring and prediction analysis of the damaged area and real-time judgment and processing. Disclosure of Invention The invention aims to provide an intelligent orthopedic joint rehabilitation monitoring system and method based on ultrasonic detection, which solve the problem that technical staff subjective judgment cannot meet actual demands in the background art. In order to solve the technical problems, the invention adopts the following technical scheme that the invention provides an intelligent monitoring method for rehabilitation of an orthopedic joint based on ultrasonic detection, which specifically comprises the following steps: S1, installing ultrasonic detection equipment, setting an acquisition period for installing the ultrasonic detection equipment, and acquiring an ultrasonic image of an orthopedic joint in real time based on the set acquisition period; S2, acquiring an ultrasonic image of the orthopedics joint in real time, and processing the acquired ultrasonic image of the orthopedics joint by an image processing method to obtain a processed ultrasonic image of the orthopedics joint; s21, filtering and denoising the acquired ultrasonic image of the orthopaedics joint by an ultrasonic filtering method to obtain the ultrasonic image of the orthopaedics joint after filtering and denoising; S22, performing image enhancement on the obtained filtered and noise-reduced orthopedic joint ultrasonic image to obtain an enhanced orthopedic joint ultrasonic image; s23, performing background separation on the enhanced orthopedic joint ultrasonic image to obtain a processed orthopedic joint ultrasonic image; S3, summarizing the orthopedics joint ultrasonic images processed in a plurality of periods, training and comparing the summarized orthopedics joint ultrasonic images, and judging the repair degree; S4, based on the repair degree judged in real time, performing repair prediction on the ultrasonic image acquired in real time in a digital twin mode. The method has the advantages that firstly, ultrasonic detection equipment is installed, the acquisition period of the ultrasonic detection equipment is set, ultrasonic images of the orthopedics joints are acquired in real time based on the set acquisition period, meanwhile, the acquired ultrasonic images of the orthopedics joints are processed through an image processing method based on the ultrasonic images of the orthopedics joints acquired in real time, after the processing of the ultrasonic images of the orthopedics joints is completed, the ultrasonic images of the orthopedics joints processed in a plurality of periods are summarized, training comparison is conducted on the ultrasonic images of the orthopedics joints obtained through the summarization, the repair degree is judged, and finally, the real-time acquired ultrasonic images are subjected to repair prediction in a digital twin mode based on the repair degree judged in real time, so that the accuracy of rehabilitation monitoring of the orthopedics joints is improved. Preferably, the step of installing the ultrasonic detection device and setting an acquisition period of installing the ultrasonic detection device, and acquiring the ultrasonic image of the orthopaedic joint in real time based on the set ac