Ultrasonic image-assisted diagnosis method and system for thyroid treatment

Abstract

The present application provides an ultrasound image-assisted diagnosis method and system for thyroid treatment, which obtains ultrasound images of different thyroid nodules to obtain an ultrasound image set; then extracts the benign and malignant features of the thyroid nodules, and clusters all the ultrasound images in the ultrasound image set according to the benign and malignant features to obtain multiple ultrasound image clusters; further determines the abnormal distribution of thyroid nodules generated by a thyroid ultrasound-assisted device during the auxiliary diagnosis process, thereby determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound-assisted device when performing ultrasound imaging; acquires ultrasound radio frequency time series signals, and then determines the trend characteristics of the ultrasound echo changing over time during the auxiliary diagnosis of thyroid nodules, thereby determining the auxiliary gain amount of the echo characteristics of the thyroid ultrasound-assisted device when performing ultrasound imaging; when the thyroid ultrasound-assisted device performs diagnostic analysis on the ultrasound image, the auxiliary gain amount performs gain adjustment on the thyroid ultrasound-assisted device.

Description

Technical Field

[0001] The present application relates to the field of auxiliary diagnosis technology, and more specifically, to an ultrasound image-assisted diagnosis method and system for thyroid treatment. Background Art

[0002] Auxiliary diagnosis is a vital part of modern medicine, using advanced scientific and technological tools to help professionals more accurately evaluate and diagnose diseases. Among them, imaging technologies such as X-rays, CT scans and magnetic resonance imaging (MRI) can penetrate into the body structure and detect abnormalities or lesions. Laboratory tests include blood, urine and tissue sample analysis, providing biochemical and cytological information to help determine the type and severity of the disease. Molecular diagnostic technology identifies specific gene mutations or biomarkers by analyzing changes in DNA, RNA and protein levels, and has important applications in cancer and genetic diseases. Biosensors and microelectronic devices are constantly developing and can monitor physiological parameters or specific molecular concentrations in the body, such as blood sugar and blood pressure. Intelligent algorithms and machine learning play a key role in big data analysis and can extract patterns and trends from complex data to assist doctors in making more accurate diagnostic inferences. These technologies are also applicable to thyroid-related diagnoses. For example, ultrasound examinations can evaluate the nature of thyroid nodules, laboratory tests can detect changes in thyroid hormone levels, and molecular diagnostic technology can help identify specific mutations in thyroid cancer. These advances in auxiliary diagnostic technologies have provided more powerful tools for medical practice and facilitated the early detection and accurate diagnosis of a variety of diseases.

[0003] Ultrasound image-assisted diagnosis in thyroid treatment uses an acoustic imaging system to evaluate the thyroid structure and abnormalities in detail. This method uses the principle of high-frequency sound waves propagating and reflecting in human tissues to generate instant images that show the size, shape, structure and blood flow of the thyroid gland. It can accurately measure the volume of the thyroid gland and check whether there are nodules, cysts or other abnormalities. Ultrasound images are very helpful in distinguishing the nature of thyroid masses and can preliminarily determine their benign or malignant characteristics. However, in the diagnosis of thyroid nodules, there is an overlap of echo features between benign and malignant nodules on ultrasound images. Echo feature overlap refers to the fact that benign and malignant nodules show similar echo patterns on ultrasound images. In ultrasound imaging, echo patterns are characteristics that describe the intensity and distribution of sound wave reflections generated by thyroid nodules in ultrasound return signals. For example, both benign and malignant thyroid nodules may show low echo or mixed echo characteristics, which will increase the difficulty of accurately judging the benign and malignant nature of thyroid nodules. Therefore, how to reduce the impact of echo feature overlap in ultrasound image-assisted diagnosis on the identification of benign and malignant thyroid nodules, thereby improving the accuracy of differential diagnosis of benign and malignant thyroid nodules, has become a difficult problem faced by the industry. Summary of the invention

[0004] The present application provides an ultrasound image-assisted diagnosis method and system for thyroid treatment, which can reduce the impact of echo feature overlap in ultrasound image-assisted diagnosis on the identification of benign and malignant thyroid nodules, thereby improving the accuracy of differential diagnosis of benign and malignant thyroid nodules.

[0005] In a first aspect, the present application provides an ultrasound image-assisted diagnosis method for thyroid treatment, comprising the following steps:

[0006] Start the thyroid ultrasound auxiliary equipment to obtain ultrasound images of different thyroid nodules and obtain an ultrasound image set;

[0007] extracting benign and malignant features of thyroid nodules from the ultrasound image set, and clustering all ultrasound images in the ultrasound image set according to the benign and malignant features to obtain multiple ultrasound image clusters;

[0008] Determining the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary device during the auxiliary diagnosis process through each ultrasound image cluster, and determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics;

[0009] Collecting ultrasonic radio frequency time series signals of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules, determining trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signals, and determining the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging through the abnormality recognition coefficient and the trend characteristics;

[0010] When the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image, the gain of the thyroid ultrasound auxiliary device is adjusted according to the auxiliary gain amount.

[0011] In some embodiments, extracting the benign and malignant features of thyroid nodules from the ultrasound image set specifically includes:

[0012] Preprocessing the ultrasound image set to obtain a preprocessed ultrasound image set;

[0013] Initialize the benign and malignant texture feature model of the thyroid gland;

[0014] The benign and malignant texture feature model is used to extract features from each preprocessed ultrasound image in the preprocessed ultrasound image set to obtain benign and malignant features of the thyroid nodule.

[0015] In some embodiments, clustering all ultrasound images in the ultrasound image set by using the benign and malignant features to obtain a plurality of ultrasound image clusters specifically includes:

[0016] Determine a plurality of initial cluster centers according to the benign and malignant characteristics;

[0017] Determining a local similarity of each ultrasound image in the ultrasound image set;

[0018] All the ultrasound images in the ultrasound image set are similarly clustered according to all the initial cluster centers and all the local similarities to obtain a plurality of ultrasound image clusters.

[0019] In some embodiments, determining the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary device during the auxiliary diagnosis process through each ultrasound image cluster specifically includes:

[0020] Abnormal features are extracted from each ultrasound image cluster to obtain multiple abnormal feature sequences of thyroid nodules;

[0021] Perform linear fitting on all abnormal characteristic sequences to obtain multiple abnormal characteristic curves;

[0022] The abnormal distribution of thyroid nodules generated by thyroid ultrasound-assisted equipment during the auxiliary diagnosis process was determined based on all abnormal characteristic curves.

[0023] In some embodiments, determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics specifically includes:

[0024] Determining auxiliary feature quantities for a thyroid ultrasound auxiliary device to identify benign and malignant thyroid nodules during auxiliary diagnosis according to the abnormal distribution of the thyroid nodules;

[0025] The abnormal recognition coefficient of the echo feature of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the auxiliary feature quantity and the benign and malignant features.

[0026] In some embodiments, determining the trend characteristics of the ultrasonic echo changing over time in assisting the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal specifically includes:

[0027] Extracting multiple peaks of ultrasound echo changes over time when assisting in diagnosing thyroid nodules from the ultrasound radio frequency time series signal;

[0028] Determining the fuzzy entropy of the ultrasound echo in assisting the diagnosis of thyroid nodules according to the ultrasound radio frequency time series signal;

[0029] The trend characteristics of the ultrasonic echo changing with time in assisting the diagnosis of thyroid nodules are determined according to the fuzzy entropy and all the peak values.

[0030] In some embodiments, determining the auxiliary gain amount of the echo feature of the thyroid ultrasound auxiliary device when performing ultrasound imaging by using the abnormality recognition coefficient and the trend feature specifically includes:

[0031] determining a trend gain value of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the trend characteristics;

[0032] The auxiliary gain amount of the echo characteristic of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the trend gain value and the abnormality recognition coefficient.

[0033] In a second aspect, the present application provides an ultrasound image-assisted diagnosis system for thyroid treatment, comprising:

[0034] An acquisition module, used to acquire ultrasound images of different thyroid nodules after starting the thyroid ultrasound auxiliary device, and obtain an ultrasound image set;

[0035] a processing module, configured to extract benign and malignant features of thyroid nodules from the ultrasound image set, and cluster all ultrasound images in the ultrasound image set according to the benign and malignant features to obtain a plurality of ultrasound image clusters;

[0036] The processing module is further used to determine the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process through each ultrasound image cluster, and determine the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics;

[0037] The processing module is further used to collect the ultrasonic radio frequency time series signal of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules, determine the trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal, and determine the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging through the abnormal recognition coefficient and the trend characteristics;

[0038] The execution module is used to adjust the gain of the thyroid ultrasound auxiliary device according to the auxiliary gain when the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image.

[0039] In a third aspect, the present application provides a computer device, comprising a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned ultrasound image-assisted diagnosis method for thyroid treatment.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the ultrasound image-assisted diagnosis method for thyroid treatment is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is an exemplary flow chart of an ultrasound image-assisted diagnosis method for thyroid treatment according to some embodiments of the present application;

[0042] Figure 2 is a schematic diagram of a process for determining benign and malignant characteristics in some embodiments of the present application;

[0043] Figure 3 is a schematic diagram of a process for determining trend characteristics in some embodiments of the present application;

[0044] Figure 4 is a structural block diagram of an ultrasound image-assisted diagnosis system for thyroid treatment in some embodiments of the present application;

[0045] Figure 5 It is a structural schematic diagram of a computer device for implementing an ultrasound image-assisted diagnosis method for thyroid treatment according to some embodiments of the present application. DETAILED DESCRIPTION

[0046] The core of the present application is to start the thyroid ultrasound auxiliary equipment, obtain ultrasound images of different thyroid nodules, and obtain an ultrasound image set; extract the benign and malignant features of the thyroid nodules from the ultrasound image set, cluster all the ultrasound images in the ultrasound image set according to the benign and malignant features, and obtain multiple ultrasound image clusters; determine the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process through each ultrasound image cluster, and determine the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant features; collect the ultrasound radio frequency time series signal of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules, determine the trend characteristics of the ultrasound echo changes over time when assisting in the diagnosis of thyroid nodules according to the ultrasound radio frequency time series signal, and determine the auxiliary gain of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal recognition coefficient and the trend characteristics; when the thyroid ultrasound auxiliary equipment performs diagnostic analysis on the ultrasound image, the gain of the thyroid ultrasound auxiliary equipment is adjusted according to the auxiliary gain. The solution of the present application can reduce the influence of echo feature overlap in ultrasound image-assisted diagnosis on the identification of benign and malignant thyroid nodules, thereby improving the accuracy of differential diagnosis of benign and malignant thyroid nodules.

[0047] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of an ultrasound image-assisted diagnosis method for thyroid treatment according to some embodiments of the present application, and the ultrasound image-assisted diagnosis method for thyroid treatment mainly includes the following steps:

[0048] In step 101, a thyroid ultrasound auxiliary device is started to obtain ultrasound images of different thyroid nodules to obtain an ultrasound image set.

[0049] In a specific implementation, after the thyroid ultrasound auxiliary device is started, ultrasound images of different thyroid nodules are obtained from a database of the thyroid ultrasound auxiliary device, so that a collection of all ultrasound images is used as an ultrasound image set.

[0050] It should be noted that the thyroid ultrasound auxiliary equipment described in this application is an auxiliary diagnostic medical device used to evaluate thyroid structure and lesions. The thyroid ultrasound auxiliary equipment includes an ultrasound probe, an ultrasound generator, a receiver, and an ultrasound image processing and diagnostic analysis system. The thyroid ultrasound auxiliary equipment uses high-frequency ultrasound technology to generate high-resolution images, which can clearly display the anatomical structure, blood flow and abnormal lesions of the thyroid gland, such as nodules, cysts, etc., and then perform measurement, diagnosis and automated analysis of the ultrasound images.

[0051] In step 102, benign and malignant features of thyroid nodules are extracted from the ultrasound image set, and all ultrasound images in the ultrasound image set are clustered according to the benign and malignant features to obtain a plurality of ultrasound image clusters.

[0052] In some embodiments, reference Figure 2 As shown, this figure is a schematic diagram of the process of determining benign and malignant features in some embodiments of the present application. In this embodiment, the benign and malignant features of thyroid nodules can be extracted from the ultrasound image set by the following steps:

[0053] First, in step 1021, the ultrasound image set is preprocessed to obtain a preprocessed ultrasound image set;

[0054] Secondly, in step 1022, the benign and malignant texture feature model of the thyroid gland is initialized;

[0055] Then, in step 1023, feature extraction is performed on each preprocessed ultrasound image in the preprocessed ultrasound image set using the benign and malignant texture feature model to obtain benign and malignant features of the thyroid nodule.

[0056] It should be noted that the benign and malignant texture feature model of the thyroid described in the present application can be constructed using the U-Net model in deep learning. First, an ultrasound image data set containing benign and malignant thyroid nodules is collected. Secondly, the nodule area in the ultrasound image is accurately marked to generate the corresponding label image. Then, all the label images are subjected to image preprocessing including histogram equalization, bilateral filtering and other operations. Then, the preprocessed ultrasound image and the corresponding label image are input as training data into the U-Net model for training. After the U-Net model training is completed, the model after training is used as the benign and malignant texture feature model of the thyroid.

[0057] In a specific implementation, the ultrasound image set is preprocessed to obtain the preprocessed ultrasound image set, which can be achieved in the following manner: first, each ultrasound image in the ultrasound image set is subjected to histogram equalization processing to enhance image contrast, then, the ultrasound images subjected to the histogram equalization processing are filtered using a bilateral filtering algorithm, and the filtered images obtained after the filtering processing are all used as preprocessed ultrasound images, and then a set consisting of all preprocessed ultrasound images is used as a preprocessed ultrasound image set.

[0058] It should be noted that the histogram equalization processing used in this application to enhance the global contrast of the ultrasound image can clarify the edges of the thyroid nodes, making the grayscale value distribution in the histogram more uniform, which is helpful for the subsequent extraction of feature information in the ultrasound image of the thyroid nodules; the use of the bilateral filtering algorithm can better balance the noise and maintain the boundary contour information.

[0059] In specific implementation, feature extraction is performed on each preprocessed ultrasound image in the preprocessed ultrasound image set through the benign and malignant texture feature model, and the benign and malignant features of the thyroid nodule can be obtained in the following manner, namely: the convolutional layer in the benign and malignant texture feature model is used to extract the texture features of each preprocessed ultrasound image in the preprocessed ultrasound image set, and the set of all extracted texture feature combinations is used as the benign and malignant features of the thyroid nodule.

[0060] It should be noted that the benign and malignant features described in the present application are composed of texture features of multiple preprocessed ultrasound images, wherein the texture features represent a vector composed of multiple features such as the morphology, boundaries, internal structure, etc. of the benign and malignant texture feature model captured by the convolutional layer. The texture features in the ultrasound image can be extracted through the benign and malignant features, which can improve the accuracy and reliability of the diagnosis of thyroid nodules, and each ultrasound image corresponds to a texture feature.

[0061] In some embodiments, clustering all ultrasound images in the ultrasound image set by the benign and malignant features to obtain multiple ultrasound image clusters may be achieved by the following steps:

[0062] Determine a plurality of initial cluster centers according to the benign and malignant characteristics;

[0063] Determining a local similarity of each ultrasound image in the ultrasound image set;

[0064] All the ultrasound images in the ultrasound image set are similarly clustered according to all the initial cluster centers and all the local similarities to obtain a plurality of ultrasound image clusters.

[0065] In specific implementation, multiple initial clustering centers can be determined by the benign and malignant features in the following manner, namely: first, the total number of all texture features in the benign and malignant features is squared and rounded down, and the rounded-down value is used as the initial clustering center; secondly, the variance of all texture features in the benign and malignant features is calculated, and the values ​​obtained by calculating the variance of each texture feature are used as the texture feature variance; then, all texture feature variances are sorted in descending order, and the sequence obtained in descending order is used as the texture feature variance sequence; then, texture feature variances with the same numerical number as the initial clustering center are selected from the texture feature variance sequence in descending order, and the texture features corresponding to the selected texture feature variances are used as the initial clustering centers. In addition, it should be noted that each initial clustering center in the present application corresponds to an ultrasound image.

[0066] In specific implementation, determining the local similarity of each ultrasonic image in the ultrasonic image set can be achieved in the following manner, namely: using the local binary pattern (Local Binary Pattern, LBP) in the prior art to calculate the local binary pattern value of the pixel points in each ultrasonic image in the ultrasonic image set, and the local binary pattern values ​​of all the pixel points of each ultrasonic image are formed into vectors in the order of the pixel points from left to right and from top to bottom, and each obtained vector is used as the local binary vector of the corresponding ultrasonic image, then, an ultrasonic image is selected from the ultrasonic image set, and the local binary vector of the ultrasonic image and the local binary vectors of the remaining ultrasonic images are calculated by using the cosine similarity, and the maximum cosine similarity is selected from all the calculated cosine similarities as the local similarity of the ultrasonic image, and the above steps are repeated to obtain the local similarities of the remaining ultrasonic images.

[0067] It should be noted that the local similarity described in the present application measures the degree of similarity between an ultrasound image and other ultrasound images in local structure or texture. The greater the local similarity, the higher the degree of similarity between the ultrasound image and other ultrasound images in local structure or texture. Conversely, the lower the degree of similarity between the ultrasound image and other ultrasound images in local structure or texture.

[0068] In specific implementation, all the ultrasonic images in the ultrasonic image set are similarly clustered according to all the initial clustering centers and all the local similarities to obtain multiple ultrasonic image clusters, which can be achieved in the following manner, namely: the average of all the local similarities is calculated and divided by the initial clustering center value, the value obtained by the division is used as the clustering similarity threshold, an initial clustering center is selected, and then the local similarities of all the ultrasonic images in the ultrasonic image set are traversed, if the absolute value of the difference between the local similarity of the ultrasonic image corresponding to the initial clustering center and the local similarities of the remaining ultrasonic images in the ultrasonic image set is less than or equal to the clustering similarity threshold, the ultrasonic image corresponding to the initial clustering center and the ultrasonic images that meet the judgment condition are divided into the same category, and the set of ultrasonic images of the same category is used as the ultrasonic image cluster of the initial clustering center, and the above steps are repeated to obtain the ultrasonic image clusters of the remaining initial clustering centers.

[0069] It should be noted that, in the present application, after similarity clustering is completed, if there are unclassified ultrasound images, all unclassified ultrasound images are classified into the same category to form an ultrasound image cluster.

[0070] In step 103, the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process is determined through each ultrasound image cluster, and the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment during ultrasound imaging is determined based on the abnormal distribution of the thyroid nodules and the benign and malignant characteristics.

[0071] In some embodiments, determining the abnormal distribution of thyroid nodules generated by a thyroid ultrasound auxiliary device during auxiliary diagnosis through each ultrasound image cluster may be achieved by the following steps:

[0072] Abnormal features are extracted from each ultrasound image cluster to obtain multiple abnormal feature sequences of thyroid nodules;

[0073] Perform linear fitting on all abnormal characteristic sequences to obtain multiple abnormal characteristic curves;

[0074] The abnormal distribution of thyroid nodules generated by thyroid ultrasound-assisted equipment during the auxiliary diagnosis process was determined based on all abnormal characteristic curves.

[0075] In specific implementation, abnormal feature extraction is performed on each ultrasound image cluster to obtain multiple abnormal feature sequences of thyroid nodules, which can be achieved in the following manner, namely: select an ultrasound image cluster, calculate the energy value of the gray level co-occurrence matrix for each ultrasound image in the ultrasound image cluster, and use the energy value of the gray level co-occurrence matrix of each ultrasound image as the abnormal feature value, then sort all the abnormal feature values ​​in ascending order, and use the sequence obtained after sorting as the abnormal feature sequence of the thyroid nodule corresponding to the ultrasound image cluster, repeat the above steps to obtain the abnormal feature sequences of the thyroid nodules corresponding to the remaining ultrasound image clusters.

[0076] It should be noted that the abnormal feature value in the abnormal feature sequence described in the present application represents the parameter value of the abnormal texture uniformity of the ultrasound image, and the abnormal distribution of gray levels between pixel pairs in the ultrasound image can be evaluated through the abnormal feature value.

[0077] In the specific implementation, linear fitting is performed on all abnormal feature sequences to obtain multiple abnormal feature curves, which can be achieved in the following manner, namely: each abnormal feature sequence is fitted using the least squares support vector machine algorithm in the prior art, and each curve obtained by fitting is used as an abnormal feature curve. In other embodiments, other fitting algorithms can also be used, which is not limited here.

[0078] In specific implementation, the abnormal distribution of thyroid nodules generated by the thyroid ultrasound-assisted device during the auxiliary diagnosis process can be determined based on all abnormal feature curves in the following manner, namely: taking the derivative of the position where each abnormal feature value is located on each abnormal feature curve, and then selecting the abnormal feature value corresponding to the position with the largest derivative on each abnormal feature curve, arranging all the selected abnormal feature values ​​into a vector in order of the local similarity of the abnormal feature values ​​corresponding to the ultrasound images, and using the vector as the abnormal distribution of thyroid nodules generated by the thyroid ultrasound-assisted device during the auxiliary diagnosis process.

[0079] It should be noted that the abnormal distribution described in the present application represents the vector of abnormal changes in benign and malignant thyroid nodules in the ultrasound image generated during the auxiliary diagnosis process of the thyroid ultrasound auxiliary equipment. The abnormal distribution can reflect the internal tissue structure and blood flow of the thyroid nodule, which is helpful to evaluate the benign and malignant nature of the nodule, the degree of lesions and possible growth trends.

[0080] In some embodiments, determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics can be achieved by the following steps:

[0081] Determining auxiliary feature quantities for a thyroid ultrasound auxiliary device to identify benign and malignant thyroid nodules during auxiliary diagnosis according to the abnormal distribution of the thyroid nodules;

[0082] The abnormal recognition coefficient of the echo feature of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the auxiliary feature quantity and the benign and malignant features.

[0083] In specific implementation, determining the auxiliary feature quantity of the thyroid ultrasound auxiliary equipment for identifying benign and malignant thyroid nodules in the auxiliary diagnosis process according to the abnormal distribution of the thyroid nodules can be achieved in the following manner, namely: performing differential calculation on the abnormal distribution of the thyroid nodules, and using the sequence obtained after the differential calculation as the abnormal distribution differential sequence, then calculating the mean of all values ​​in the abnormal distribution differential sequence, and using the obtained mean as the auxiliary feature quantity of the thyroid ultrasound auxiliary equipment for identifying benign and malignant thyroid nodules in the auxiliary diagnosis process.

[0084] In addition, it should be noted that the auxiliary feature quantity described in the present application represents the feature quantity of the thyroid ultrasound auxiliary equipment for identifying benign and malignant thyroid nodules under the influence of the echo characteristics of the ultrasound image. The auxiliary feature quantity can be used to distinguish between benign and malignant lesions, thereby improving the accuracy of auxiliary diagnosis.

[0085] It should be noted that the abnormal recognition coefficient described in the present application represents the parameter value of the abnormal characteristics between benign thyroid nodules and malignant thyroid nodules identified by the thyroid ultrasound auxiliary equipment when generating echo features during ultrasound imaging. As a preferred embodiment, the abnormal recognition coefficient of the echo features of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the auxiliary feature quantity and the benign and malignant features. It can be achieved in the following way, namely: each value in the benign and malignant features is multiplied by the auxiliary feature quantity, and then the average of all the multiplied values ​​is calculated, and the value obtained after taking the average is used as the abnormal recognition coefficient of the echo features of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging.

[0086] In step 104, the ultrasonic radio frequency time series signal of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules is collected, and the trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules are determined based on the ultrasonic radio frequency time series signal, and the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging is determined by the abnormality recognition coefficient and the trend characteristics.

[0087] In specific implementation, the position of the ultrasound probe in the thyroid ultrasound auxiliary equipment is kept unchanged, the ultrasound device is switched to the radio frequency function state, and the radio frequency ultrasound echo signal is continuously collected for 120 seconds according to the set parameters. The sequence of radio frequency ultrasound echo signals obtained by continuous collection is used as the ultrasound radio frequency time series signal for auxiliary diagnosis of thyroid nodules.

[0088] It should be noted that the ultrasonic radio frequency time series signal described in the present application is composed of a plurality of time series points and radio frequency ultrasonic echo signal values ​​corresponding to each time series point.

[0089] In some embodiments, reference Figure 3 As shown, this figure is a schematic diagram of the process of determining trend characteristics in some embodiments of the present application. In this embodiment, the trend characteristics of the change of ultrasound echo over time in assisting the diagnosis of thyroid nodules can be determined according to the ultrasound radio frequency time series signal, which can be achieved by the following steps:

[0090] First, in step 1041, multiple peaks of ultrasound echoes changing with time when assisting in diagnosing thyroid nodules are extracted from the ultrasound radio frequency time series signal;

[0091] Secondly, in step 1042, the fuzzy entropy of the ultrasound echo in assisting the diagnosis of thyroid nodules is determined according to the ultrasound radio frequency time series signal;

[0092] Then, in step 1043, the trend characteristics of the ultrasonic echo changing with time in assisting the diagnosis of thyroid nodules are determined according to the fuzzy entropy and all the peak values.

[0093] In specific implementation, extracting multiple fluctuation peaks of ultrasonic echoes changing with time to assist in diagnosing thyroid nodules from the ultrasonic radio frequency time series signal can be achieved in the following way, namely: using wavelet transform to extract multiple peaks of ultrasonic echoes changing with time from the ultrasonic radio frequency time series signal.

[0094] It should be noted that the wavelet transform used in this application can provide time-frequency localization information of the signal, which is very useful for analyzing the transient features and peaks in the thyroid ultrasound radio frequency time series signal. The ultrasound signals of thyroid nodules often have complex time variation patterns, and the wavelet transform can help capture the peaks in these signals and their time domain positions.

[0095] In a specific implementation, determining the fuzzy entropy of the ultrasonic echo in the auxiliary diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal can be achieved in the following manner, namely: dividing the ultrasonic radio frequency time series signal into two parts from the middle position of the ultrasonic radio frequency time series signal, namely an upper ultrasonic radio frequency time series signal and a lower ultrasonic radio frequency time series signal, and then selecting the first time series point from the upper ultrasonic radio frequency time series signal, selecting the first time series point from the lower ultrasonic radio frequency time series signal, calculating the Euclidean distance between the two selected time series points, and using the obtained values ​​as the spatial distance between the two time series points, and then selecting the second time series point from the upper ultrasonic radio frequency time series signal. time series points, select the second time series point from the ultrasound radio frequency time series signal, calculate the Euclidean distance between the two selected time series points, and use the obtained values ​​as the spatial distance between the two time series points, repeat the above steps, and obtain the spatial distances of all remaining time series points in turn. For each time series point, divide the radio frequency ultrasound echo signal value of the time series point by the spatial distance of the time series point, and use the divided values ​​as the spatial fluctuation values ​​of each time series point. Finally, all the spatial fluctuation values ​​are described by entropy characteristics through the information entropy in the prior art, and the value obtained after the entropy characteristic description can be used as the fuzzy entropy of the ultrasound echo in the auxiliary diagnosis of thyroid nodules.

[0096] It should be noted that the fuzzy entropy described in the present application can reflect the degree of spatial fluctuation of the ultrasonic echo signal, and a higher fuzzy entropy indicates that the spatial variation of the signal is more complex or irregular.

[0097] In specific implementation, the trend characteristics of the ultrasonic echo changing over time in assisting the diagnosis of thyroid nodules can be determined based on the fuzzy entropy and all peak values, which can be achieved in the following way, namely: for each peak value, the fuzzy entropy is multiplied by the peak value, and the multiplied values ​​are all used as trend characteristic values, and then, each peak value is sorted according to the time sequence in which it is located in the ultrasonic radio frequency time series signal, and the arranged sequence is used as the trend characteristic of the ultrasonic echo changing over time in assisting the diagnosis of thyroid nodules.

[0098] It should be noted that the trend characteristics described in the present application are composed of multiple trend characteristic values, wherein the trend characteristic values ​​can reflect the trend of ultrasound echo changes over time, and the dynamic characteristics of benign and malignant thyroid nodules can be evaluated through the trend characteristics.

[0099] In some embodiments, determining the auxiliary gain amount of the echo feature of the thyroid ultrasound auxiliary device when performing ultrasound imaging by using the abnormality identification coefficient and the trend feature can be achieved by using the following steps:

[0100] determining a trend gain value of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the trend characteristics;

[0101] The auxiliary gain amount of the echo characteristic of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the trend gain value and the abnormality recognition coefficient.

[0102] In specific implementation, determining the trend gain value of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the trend feature can be achieved in the following manner, namely: first, create a sliding window, set the length of the sliding window to the value obtained by taking the square root of the total number of trend feature values ​​in the trend feature and rounding it down, then align the sliding window with the first trend feature value in the trend feature, select the largest trend feature value from the trend feature values ​​covered by the sliding window, and use the largest trend feature value as the maximum trend feature value, then slide the sliding window, align the sliding window with the second trend feature value in the trend feature, select the largest trend feature value from the trend feature values ​​covered by the sliding window, and use the largest trend feature value as the maximum trend feature value, repeat the above steps, slide the sliding window in sequence until the last position of the sliding window is aligned with the last trend feature value in the trend feature and stop sliding, finally, average all the obtained maximum trend feature values, and use the value obtained by averaging as the trend gain value of the thyroid ultrasound auxiliary device when performing ultrasound imaging.

[0103] In addition, it should be noted that the trend gain value described in the present application represents the characteristic value of the abnormal change trend of benign and malignant thyroid nodules assisted by thyroid ultrasound auxiliary equipment for auxiliary diagnosis.

[0104] It should be noted that the auxiliary gain amount described in the present application represents a quantitative value of the sensitivity of the thyroid ultrasound auxiliary equipment to the echo feature recognition. By processing and enhancing the features extracted from the ultrasonic radio frequency signal, the detection and classification capabilities of the thyroid ultrasound auxiliary equipment for thyroid nodules are improved. As a preferred embodiment, the auxiliary gain amount of the echo feature of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging is determined by the trend gain value and the abnormal recognition coefficient. This can be achieved in the following manner, namely: the value obtained by multiplying the trend gain value and the abnormal recognition coefficient is used as the auxiliary gain amount of the echo feature of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging.

[0105] In step 105, when the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image, the gain of the thyroid ultrasound auxiliary device is adjusted according to the auxiliary gain amount.

[0106] In some embodiments, when the thyroid ultrasound auxiliary device performs diagnostic analysis on an ultrasound image, gain adjustment of the thyroid ultrasound auxiliary device according to the auxiliary gain amount may be implemented by the following steps:

[0107] When the thyroid ultrasound auxiliary equipment performs diagnostic analysis on the ultrasound image, different depth gain areas of the thyroid nodule lesion area are determined;

[0108] The time gain compensation slider in the thyroid ultrasound auxiliary device is adjusted in a zoned manner according to the auxiliary gain amount and all the depth gain areas.

[0109] In specific implementation, different depth gain areas of the thyroid nodule lesion area can be determined in the following manner, namely: first, use the high-frequency probe of the thyroid ultrasound auxiliary equipment and select the thyroid-specific scanning mode, and set three ranges according to the depth of the thyroid nodule lesion area, namely shallow, medium-deep and deep, where the shallow range is 0-1 cm, the medium-deep range is 1-2 cm, and the deep range is more than 2 cm, and the shallow areas, medium-deep areas and deep areas of different depths are all used as depth gain areas.

[0110] It should be noted that the depth gain region described in the present application refers to the region where the gain of thyroid nodules at different depths is controlled using thyroid ultrasound-assisted equipment.

[0111] In addition, it should be noted that the time gain compensation (TGC) slider in the present application is a control device in the thyroid ultrasound auxiliary equipment, which is used to adjust the ultrasound signal gain at different depths. The TGC slider can compensate for the ultrasound signal attenuation caused by the increase in depth, thereby ensuring uniform brightness of the entire ultrasound image.

[0112] In specific implementation, the time gain compensation slider in the thyroid ultrasound auxiliary device is adjusted in a zoned manner according to the auxiliary gain amount and all the depth gain areas, which can be achieved in the following manner, namely: for the shallow depth gain area, the ultrasonic echo signal is strong, and the time gain compensation slider is reduced by a value equal to the auxiliary gain amount. By reducing the gain, the image brightness of the shallow area is controlled to avoid overexposure, making the thyroid tissue and details clearer; for the medium-deep depth gain area, the gain level of the medium-deep layer is kept unchanged, and the medium-deep thyroid tissue structure can be clearly displayed with rich details and a high signal-to-noise ratio, which can be accurately diagnosed; for the deep depth gain area, the ultrasonic signal is significantly attenuated, and the time gain compensation slider is increased by a value equal to the auxiliary gain amount, so that the deep signal is effectively enhanced. By increasing the gain, the ultrasonic signal in the deep area is compensated, which increases the clarity and details of the thyroid structure in the image, making it easier to find potential lesions. Other methods can also be used in other embodiments, which will not be repeated here.

[0113] In addition, in another aspect of the present application, in some embodiments, the present application provides an ultrasound image-assisted diagnosis system for thyroid treatment, referring to Figure 4, which is a schematic diagram of exemplary hardware and / or software of an ultrasound image-assisted diagnosis system for thyroid treatment according to some embodiments of the present application, the ultrasound image-assisted diagnosis system 400 for thyroid treatment includes: an acquisition module 401, a processing module 402 and an execution module 403, which are described as follows:

[0114] Acquisition module 401, in the present application, the acquisition module 401 is mainly used to acquire ultrasound images of different thyroid nodules after starting the thyroid ultrasound auxiliary device to obtain an ultrasound image set;

[0115] Processing module 402, in the present application, is used to extract benign and malignant features of thyroid nodules from the ultrasound image set, and cluster all ultrasound images in the ultrasound image set according to the benign and malignant features to obtain multiple ultrasound image clusters;

[0116] It should be noted that the processing module 402 in the present application is also used to determine the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process through each ultrasound image cluster, and determine the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics;

[0117] In addition, the processing module 402 in the present application is also used to collect the ultrasonic radio frequency time series signal of the thyroid ultrasound auxiliary device when assisting in the diagnosis of thyroid nodules, determine the trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal, and determine the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary device when performing ultrasonic imaging through the abnormal recognition coefficient and the trend characteristics;

[0118] The execution module 403 in the present application is mainly used to adjust the gain of the thyroid ultrasound auxiliary device according to the auxiliary gain when the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image.

[0119] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores codes, and the processor is configured to obtain the codes and execute the above-mentioned ultrasound image-assisted diagnosis method for thyroid treatment.

[0120] In some embodiments, reference Figure 5 , which is a schematic diagram of the structure of a computer device according to some embodiments of the present application using an ultrasound image-assisted diagnosis method for thyroid treatment. The ultrasound image-assisted diagnosis method for thyroid treatment in the above embodiment can be performed by Figure 5The computer device 500 shown in the figure is implemented, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503 and at least one communication interface 504.

[0121] The processor 501 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more components for controlling the execution of the ultrasound image-assisted diagnosis method for thyroid treatment in the present application.

[0122] The communication bus 502 may be used to transmit information between the above-mentioned components.

[0123] The memory 503 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 503 may exist independently and be connected to the processor 501 via the communication bus 502. The memory 503 may also be integrated with the processor 501.

[0124] The memory 503 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 501. The processor 501 is used to execute the program code stored in the memory 503. The program code may include one or more software modules. The method described in the above method embodiment can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0125] The communication interface 504 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0126] In a specific implementation, as an embodiment, a computer device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0127] The above-mentioned computer device may be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer device.

[0128] In addition, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned ultrasound image-assisted diagnosis method for thyroid treatment.

[0129] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0130] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. An ultrasound image-assisted diagnosis method for thyroid treatment, characterized in that: The steps include: Start the thyroid ultrasound auxiliary equipment to obtain ultrasound images of different thyroid nodules and obtain an ultrasound image set; extracting benign and malignant features of thyroid nodules from the ultrasound image set, and clustering all ultrasound images in the ultrasound image set according to the benign and malignant features to obtain multiple ultrasound image clusters; Determining the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary device during the auxiliary diagnosis process through each ultrasound image cluster, and determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics; Collecting ultrasonic radio frequency time series signals of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules, determining trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signals, and determining the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging through the abnormality recognition coefficient and the trend characteristics; When the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image, the gain of the thyroid ultrasound auxiliary device is adjusted according to the auxiliary gain amount; The abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process is determined by each ultrasound image cluster, including: Abnormal features are extracted from each ultrasound image cluster to obtain multiple abnormal feature sequences of thyroid nodules; Perform linear fitting on all abnormal characteristic sequences to obtain multiple abnormal characteristic curves; Determine the abnormal distribution of thyroid nodules generated by thyroid ultrasound-assisted equipment during the auxiliary diagnosis process based on all abnormal characteristic curves; The derivative of each abnormal characteristic value on each abnormal characteristic curve is calculated, and then the abnormal characteristic value corresponding to the position with the largest derivative on each abnormal characteristic curve is selected. All the selected abnormal characteristic values ​​are arranged into a vector in the order of the local similarity of the abnormal characteristic values ​​corresponding to the ultrasound images, and the vector is used as the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment in the auxiliary diagnosis process.

2. The method according to claim 1, characterized in that The benign and malignant features of thyroid nodules extracted from the ultrasound image set specifically include: Preprocessing the ultrasound image set to obtain a preprocessed ultrasound image set; Initialize the benign and malignant texture feature model of the thyroid gland; The benign and malignant texture feature model is used to extract features from each preprocessed ultrasound image in the preprocessed ultrasound image set to obtain benign and malignant features of the thyroid nodule.

3. The method according to claim 1, characterized in that Clustering all the ultrasound images in the ultrasound image set by using the benign and malignant features to obtain multiple ultrasound image clusters specifically includes: Determine a plurality of initial cluster centers according to the benign and malignant characteristics; Determining a local similarity of each ultrasound image in the ultrasound image set; All the ultrasound images in the ultrasound image set are similarly clustered according to all the initial cluster centers and all the local similarities to obtain a plurality of ultrasound image clusters.

4. The method according to claim 1, characterized in that Determining the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics specifically includes: Determining auxiliary feature quantities for a thyroid ultrasound auxiliary device to identify benign and malignant thyroid nodules during auxiliary diagnosis according to the abnormal distribution of the thyroid nodules; The abnormal recognition coefficient of the echo feature of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the auxiliary feature quantity and the benign and malignant features.

5. The method according to claim 1, characterized in that Determining the trend characteristics of the ultrasonic echo changing over time in assisting the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal specifically includes: Extracting multiple peaks of ultrasound echo changes over time when assisting in diagnosing thyroid nodules from the ultrasound radio frequency time series signal; Determining the fuzzy entropy of the ultrasound echo in assisting the diagnosis of thyroid nodules according to the ultrasound radio frequency time series signal; The trend characteristics of the ultrasonic echo changing with time in assisting the diagnosis of thyroid nodules are determined according to the fuzzy entropy and all the peak values.

6. The method according to claim 1, characterized in that Determining the auxiliary gain amount of the echo feature of the thyroid ultrasound auxiliary device when performing ultrasound imaging by using the abnormal recognition coefficient and the trend feature specifically includes: determining a trend gain value of the thyroid ultrasound auxiliary device when performing ultrasound imaging according to the trend characteristics; The auxiliary gain amount of the echo characteristic of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging is determined by the trend gain value and the abnormality recognition coefficient.

7. An ultrasound image-assisted diagnosis system for thyroid treatment, characterized in that: include: An acquisition module, used to acquire ultrasound images of different thyroid nodules after starting the thyroid ultrasound auxiliary device, and obtain an ultrasound image set; a processing module, configured to extract benign and malignant features of thyroid nodules from the ultrasound image set, and cluster all ultrasound images in the ultrasound image set according to the benign and malignant features to obtain a plurality of ultrasound image clusters; The processing module is further used to determine the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process through each ultrasound image cluster, and determine the abnormal recognition coefficient of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasound imaging according to the abnormal distribution of the thyroid nodules and the benign and malignant characteristics; The processing module is further used to collect the ultrasonic radio frequency time series signal of the thyroid ultrasound auxiliary equipment when assisting in the diagnosis of thyroid nodules, determine the trend characteristics of the ultrasonic echo changing over time when assisting in the diagnosis of thyroid nodules according to the ultrasonic radio frequency time series signal, and determine the auxiliary gain amount of the echo characteristics of the thyroid ultrasound auxiliary equipment when performing ultrasonic imaging through the abnormal recognition coefficient and the trend characteristics; an execution module, configured to adjust the gain of the thyroid ultrasound auxiliary device according to the auxiliary gain amount when the thyroid ultrasound auxiliary device performs diagnostic analysis on the ultrasound image; The abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment during the auxiliary diagnosis process is determined by each ultrasound image cluster, including: Abnormal features are extracted from each ultrasound image cluster to obtain multiple abnormal feature sequences of thyroid nodules; Perform linear fitting on all abnormal characteristic sequences to obtain multiple abnormal characteristic curves; Determine the abnormal distribution of thyroid nodules generated by thyroid ultrasound-assisted equipment during the auxiliary diagnosis process based on all abnormal characteristic curves; The derivative of each abnormal characteristic value on each abnormal characteristic curve is calculated, and then the abnormal characteristic value corresponding to the position with the largest derivative on each abnormal characteristic curve is selected. All the selected abnormal characteristic values ​​are arranged into a vector in the order of the local similarity of the abnormal characteristic values ​​corresponding to the ultrasound images, and the vector is used as the abnormal distribution of thyroid nodules generated by the thyroid ultrasound auxiliary equipment in the auxiliary diagnosis process.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the ultrasound image-assisted diagnosis method for thyroid treatment according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the ultrasound image-assisted diagnosis method for thyroid treatment according to any one of claims 1 to 6 is implemented.

Images