Joint ultrasound image visual enhancement system and method based on neural network

Abstract

The invention relates to the technical field of joint ultrasound images, and discloses a joint ultrasound image vision enhancement system based on a neural network, and the system comprises an information collection module, an analysis processing module, and an evaluation output module. The information acquisition module is used for acquiring basic information, joint lesion related disease information and historical joint ultrasonic image information of a patient, forming a data set from acquired data and sending the data set to the analysis processing module, and the analysis processing module is used for evaluating a patient group data set and related disease indexes; the analysis processing module sends the patient group data set and the related disease index obtained through analysis to the evaluation output module, and the evaluation output module carries out grading and color distinguishing according to the evaluation result of the patient group data set and the evaluation result of the related disease index. The current joint ultrasonic image is comprehensively analyzed by fully combining the personal information of the patient, the joint lesion related disease information and the historical joint ultrasonic image, and misdiagnosis and missed diagnosis are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of joint ultrasonic images, and in particular to a joint ultrasonic image visual enhancement system and method based on a neural network. Background Art

[0002] Joint ultrasound is a method of examining the joints using ultrasonic imaging technology. It can help doctors diagnose various joint diseases, such as arthritis, tendinitis, bursitis and ligament injuries. Joint ultrasound examination has the advantages of being non-invasive, radiation-free, real-time and portable. Ultrasound transmits and receives high-frequency sound waves through the probe. These sound waves will be reflected when passing through tissues of different densities, thus forming an image. Ultrasound is non-invasive and radiation-free, and is suitable for evaluating soft tissues.

[0003] Although existing neural network-based systems can assist doctors in interpreting joint ultrasound images, there is still no way to combine the patient's own clinical information for a comprehensive interpretation, which may lead to misdiagnosis or missed diagnosis. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a neural network-based joint ultrasound image visual enhancement system and method, which has the advantages of fully combining the patient's personal information, joint disease-related disease information, and historical joint ultrasound images to comprehensively analyze the current joint ultrasound image, thereby reducing misdiagnosis and missed diagnosis.

[0006] (II) Technical solution

[0007] To achieve the above-mentioned object, the present invention provides the following technical solutions: a visual enhancement system for joint ultrasound images based on neural network, comprising an information acquisition module, an analysis and processing module, and an evaluation and output module;

[0008] The information acquisition module includes a patient information acquisition unit, a related disease information acquisition unit and a historical joint ultrasound image information acquisition unit. The patient information acquisition unit is connected to the hospital central system through a network to collect the patient's basic information Hx and form a patient basic information data set. The related disease information acquisition unit is connected to the hospital central system through a network to collect joint lesion related disease information Bx and form a joint lesion related disease information data set. The historical joint ultrasound image information acquisition unit is connected to the hospital central system through a network to collect historical joint ultrasound image information Cx and form a historical joint ultrasound image information data set. The information acquisition module sends the collected data set to the analysis and processing module.

[0009] The analysis and processing module includes a patient information analysis unit and a related disease information analysis unit. The patient information analysis unit evaluates the patient group data group HZqt through the patient basic information data set and the historical joint ultrasound image information data set. The patient group data group HZqt is extracted according to the patient basic information data set, including age groups of patients of different genders. The related disease information analysis unit evaluates the related disease index XGzs according to the joint lesion related disease information data set and the historical joint ultrasound image information data set. The analysis and processing module sends the analyzed patient group data group HZqt and the related disease index XGzs to the evaluation output module;

[0010] The evaluation output module performs a light red color marking on the current joint ultrasound image according to the evaluation result of the patient group data group HZqt;

[0011] The evaluation output module divides the evaluation results of the relevant disease index XGzs into primary, intermediate and advanced levels, and marks the levels in color according to light yellow, medium yellow and dark yellow.

[0012] Preferably, the expression of the patient basic information data set is: {Hx1 a 、Hx2 a 、Hx3 a 、…、Hxn a}, where Hx1 a Represents the first patient basic information in the patient basic information dataset, Hxn a represents the last patient's basic information in the patient basic information dataset, n represents the number of patients for whom basic information is collected, and a represents the patient's basic information, including the patient's age, gender, lifestyle, and family medical history.

[0013] Preferably, the expression of the joint disease-related disease information data set is:

[0014] {Bx1, Bx2, Bx3, …, Bxn}, where Bx1 represents the first joint disease related disease information in the joint disease related disease information data set, Bxn represents the last joint disease related disease information in the joint disease related disease information data set, and n represents that there are n pieces of collected joint disease related disease information.

[0015] Preferably, the expression of the historical joint ultrasound image information data set is:

[0016] {Cx1, Cx2, Cx3, …, Cxn}, where Cx1 represents the first historical joint ultrasound image information in the historical joint ultrasound image information data set, Cxn represents the last historical joint ultrasound image information in the historical joint ultrasound image information data set, and n represents that there are n pieces of historical joint ultrasound image information collected.

[0017] Preferably, the patient group data group HZqt is divided into age groups of patients of different genders according to the patient basic information data set, including female juvenile group, female middle-aged group, female elderly group, male juvenile group, male middle-aged group, and male elderly group, wherein the female juvenile group includes female juveniles aged 0 to 16 years old, the female middle-aged group includes middle-aged people aged 17 to 55 years old, the female elderly group is elderly people over 55 years old, the male juvenile group includes male juveniles aged 0 to 16 years old, the male middle-aged group includes middle-aged people aged 17 to 60 years old, and the male elderly group is elderly people over 60 years old, and the labels of the female juvenile group, female middle-aged group, female elderly group, male juvenile group, male middle-aged group, and male elderly group are: Fs, Fz, Fl, Ms, Mz, Ml.

[0018] Preferably, the calculation formula of the patient population data set HZqt is as follows:

[0019]

[0020] In the above formula, Represents the sum of the patient basic information data set, represents the sum of historical joint ultrasound image information datasets, Represents the basic information data of x female teenagers in the patient basic information data set, Represents the basic information data of y middle-aged women in the patient basic information data set, Represents the basic information data of o elderly women in the patient basic information data set, Represents the basic information data of p male teenagers in the patient basic information data set, Represents the basic information data of q middle-aged males in the patient basic information dataset, Represents the basic information data of r elderly males in the patient basic information dataset.

[0021] Preferably, when the patient group data group HZqt is compared with the current joint ultrasound image, group correspondence is performed according to age groups of patients of different genders, and a light red color mark is generated for the corresponding group.

[0022] Preferably, the relevant disease index XGzs is calculated as follows:

[0023]

[0024] In the above formula, w represents the number of disease information data related to specific joint diseases.

[0025] Preferably, when the relevant disease index XGzs is less than 10%, it is judged as primary, and light yellow is used for grade color identification; when the relevant disease index XGzs is greater than 10% and less than 25%, it is judged as intermediate, and medium yellow is used for grade color identification; when the relevant disease index XGzs is greater than 25%, it is judged as advanced, and dark yellow is used for grade color identification.

[0026] The method for visual enhancement of joint ultrasound images based on neural network includes the following steps:

[0027] Step 1: Collect the patient's basic information Hx, joint lesion-related disease information Bx, and historical joint ultrasound image information Cx, and form a patient basic information data set, a joint lesion-related disease information data set, and a historical joint ultrasound image information data set respectively;

[0028] Step 2: Evaluate the HZqt of the patient group data set using the patient basic information data set and the historical joint ultrasound image information data set;

[0029] Step 3: Evaluate the relevant disease index XGzs based on the joint lesion related disease information dataset and the historical joint ultrasound image information dataset;

[0030] Step 4: the evaluation output module marks the current joint ultrasound image with a light red color according to the HZqt evaluation result of the patient group data group;

[0031] Step 5. The evaluation results of the relevant disease index XGzs are divided into primary, intermediate and advanced levels, and the level colors are marked as light yellow, medium yellow and dark yellow respectively.

[0032] Compared with the prior art, the present invention provides a joint ultrasound image visual enhancement system and method based on a neural network, which has the following beneficial effects:

[0033] 1. The present invention forms a patient basic information data set by combining the patient's basic information age, gender, living habits, and family medical history, comprehensively considers the patient's basic information and the analysis results of the ultrasound image, reduces the possibility of misdiagnosis and missed diagnosis, comprehensively analyzes the patient's own condition, and comprehensively interprets the joint ultrasound image to improve the effect of joint ultrasound image interpretation. The patient's basic information data set and the historical joint ultrasound image information data set are used to perform HZqt evaluation on the patient group data group of age groups of patients of different genders, and the differences in joint growth changes of different genders and age groups are fully combined to perform joint ultrasound image analysis, which is helpful for individualized disease assessment and diagnosis, and understanding the impact of gender and age on joint health. It can help patients better understand their own disease conditions and improve their recognition and compliance with treatment plans.

[0034] 2. The present invention combines joint lesion-related disease information data, fully considers the impact of joint lesion-related diseases on joints, and combines joint ultrasound images to more accurately identify and diagnose different types of joint lesions, more comprehensively evaluate joint structure changes, and help judge the severity of the disease. At the same time, with the support of lesion information, ultrasound examination can help identify early lesions and promote early treatment. The relevant disease index XGzs is evaluated based on the joint lesion-related disease information data set and the historical joint ultrasound image information data set, and then the relevant disease index XGzs evaluation results are divided into primary, intermediate, and advanced levels, and the level colors are marked in turn according to light yellow, medium yellow, and dark yellow. This helps to improve the visual intensity of joint ultrasound images in combination with joint lesion-related disease information data, and facilitates doctors to analyze joint ultrasound images. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the system flow of the present invention;

[0036] Figure 2 It is a step diagram of the method of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] See also Figure 1-2 , a visual enhancement system for joint ultrasound images based on neural network, including an information acquisition module, an analysis and processing module, and an evaluation and output module;

[0039] The information collection module includes a patient information collection unit, a related disease information collection unit and a historical joint ultrasound image information collection unit. The patient information collection unit is connected to the hospital central system through the network to collect the patient's basic information Hx and form a patient basic information data set. The related disease information collection unit is connected to the hospital central system through the network to collect joint lesion related disease information Bx and form a joint lesion related disease information data set. The historical joint ultrasound image information collection unit is connected to the hospital central system through the network to collect historical joint ultrasound image information Cx and form a historical joint ultrasound image information data set. The information collection module sends the collected data set to the analysis and processing module;

[0040] The expression of the patient basic information data set is: {Hx1 a 、Hx2 a 、Hx3 a、…、Hxn a}, where Hx1 a Represents the first patient basic information in the patient basic information dataset, Hxn a represents the last patient's basic information in the patient basic information dataset, n represents the number of patients whose basic information is collected, and a represents the patient's basic information, including the patient's age, gender, lifestyle, and family medical history;

[0041] By combining the patient's basic information, age, gender, living habits, and family medical history, a patient basic information data set is formed. The patient's basic information and ultrasound image analysis results are comprehensively considered to reduce the possibility of misdiagnosis and missed diagnosis, comprehensively analyze the patient's own situation, and comprehensively interpret joint ultrasound images to improve the effect of joint ultrasound image interpretation;

[0042] The expression of the joint disease related disease information dataset is:

[0043] {Bx1, Bx2, Bx3, …, Bxn}, where Bx1 represents the first joint lesion-related disease information in the joint lesion-related disease information data set, Bxn represents the last joint lesion-related disease information in the joint lesion-related disease information data set, and n represents that there are n pieces of joint lesion-related disease information collected;

[0044] By combining the information data of joint lesions, taking into full account the impact of joint lesions on joints, and combining with joint ultrasound images, different types of joint lesions can be more accurately identified and diagnosed, and joint structure changes can be more comprehensively evaluated, which helps to determine the severity of the disease. At the same time, with the support of lesion information, ultrasound examination can help identify early lesions and promote early treatment.

[0045] The expression of the historical joint ultrasound image information dataset is:

[0046] {Cx1, Cx2, Cx3, …, Cxn}, where Cx1 represents the first historical joint ultrasound image information in the historical joint ultrasound image information data set, Cxn represents the last historical joint ultrasound image information in the historical joint ultrasound image information data set, and n represents that there are n pieces of historical joint ultrasound image information collected;

[0047] The analysis and processing module includes a patient information analysis unit and a related disease information analysis unit. The patient information analysis unit evaluates the patient group data group HZqt through the patient basic information data set and the historical joint ultrasound image information data set. The patient group data group HZqt is extracted according to the patient basic information data set, including age groups of patients of different genders. The related disease information analysis unit evaluates the related disease index XGzs according to the joint lesion related disease information data set and the historical joint ultrasound image information data set. The analysis and processing module sends the analyzed patient group data group HZqt and the related disease index XGzs to the evaluation output module;

[0048] The patient group data group HZqt is divided into age groups of patients of different genders according to the patient basic information data set, including female adolescent group, female middle-aged group, female elderly group, male adolescent group, male middle-aged group, and male elderly group. Among them, the female adolescent group includes adolescents with female age range of 0 to 16 years old, the female middle-aged group includes middle-aged people with female age range of 17 to 55 years old, the female elderly group includes elderly people over 55 years old, the male adolescent group includes adolescents with male age range of 0 to 16 years old, the male middle-aged group includes middle-aged people with male age range of 17 to 60 years old, and the male elderly group includes elderly people over 60 years old. The labels of the female adolescent group, female middle-aged group, female elderly group, male adolescent group, male middle-aged group, and male elderly group are: Fs, Fz, Fl, Ms, Mz, Ml;

[0049] The calculation formula for HZqt of the patient population data set is as follows:

[0050]

[0051] In the above formula, Represents the sum of the patient basic information data set, represents the sum of historical joint ultrasound image information datasets, Represents the basic information data of x female teenagers in the patient basic information data set, Represents the basic information data of y middle-aged women in the patient basic information data set, Represents the basic information data of o elderly women in the patient basic information data set, Represents the basic information data of p male teenagers in the patient basic information data set, Represents the basic information data of q middle-aged males in the patient basic information dataset, Represents the basic information data of r elderly males in the patient basic information dataset;

[0052] The calculation method of the relevant disease index XGzs is as follows:

[0053]

[0054] In the above formula, w represents the number of disease information data related to specific joint lesions;

[0055] The evaluation output module compares the patient group data set HZqt with the current joint ultrasound image, matches the groups according to the age groups of patients of different genders, and generates a light red color mark for the corresponding groups;

[0056] The evaluation output module divides the evaluation results of the relevant disease index XGzs into primary, intermediate and advanced levels, and marks the levels in light yellow, medium yellow and dark yellow. When the relevant disease index XGzs is less than 10%, it is judged as primary and marked with light yellow. When the relevant disease index XGzs is greater than 10% and less than 25%, it is judged as intermediate and marked with medium yellow. When the relevant disease index XGzs is greater than 25%, it is judged as advanced and marked with dark yellow.

[0057] The HZqt of the patient group data group of different genders and age groups is evaluated through the patient basic information data set and the historical joint ultrasound image information data set. The joint ultrasound image analysis is fully combined with the differences in joint growth changes of different genders and age groups, which is helpful for individualized disease assessment and diagnosis, and understanding the impact of gender and age on joint health. It can help patients better understand their own disease conditions and improve their recognition and compliance with treatment plans;

[0058] By evaluating the relevant disease index XGzs based on the joint lesion-related disease information data set and the historical joint ultrasound image information data set, and then dividing the relevant disease index XGzs evaluation results into primary, intermediate and advanced levels, and marking the levels in light yellow, medium yellow and dark yellow, it helps to improve the visual intensity of joint ultrasound images by combining the joint lesion-related disease information data, and facilitates doctors to analyze joint ultrasound images;

[0059] The specific method of the above-mentioned neural network-based joint ultrasound image visual enhancement system is as follows:

[0060] Step 1: Collect the patient's basic information Hx, joint lesion-related disease information Bx, and historical joint ultrasound image information Cx, and form a patient basic information data set, a joint lesion-related disease information data set, and a historical joint ultrasound image information data set respectively;

[0061] Step 2: Evaluate the HZqt of the patient group data set using the patient basic information data set and the historical joint ultrasound image information data set;

[0062] Step 3: Evaluate the relevant disease index XGzs based on the joint lesion related disease information dataset and the historical joint ultrasound image information dataset;

[0063] Step 4: the evaluation output module marks the current joint ultrasound image with a light red color according to the HZqt evaluation result of the patient group data group;

[0064] Step 5. The evaluation results of the relevant disease index XGzs are divided into primary, intermediate and advanced levels, and the level colors are marked as light yellow, medium yellow and dark yellow respectively.

[0065] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A joint ultrasound image visual enhancement system based on neural network, characterized by: It includes information collection module, analysis and processing module and evaluation output module; The information acquisition module includes a patient information acquisition unit, a related disease information acquisition unit and a historical joint ultrasound image information acquisition unit. The patient information acquisition unit is connected to the hospital central system through a network to collect the patient's basic information Hx and form a patient basic information data set. The related disease information acquisition unit is connected to the hospital central system through a network to collect joint lesion related disease information Bx and form a joint lesion related disease information data set. The historical joint ultrasound image information acquisition unit is connected to the hospital central system through a network to collect historical joint ultrasound image information Cx and form a historical joint ultrasound image information data set. The information acquisition module sends the collected data set to the analysis and processing module. The analysis and processing module includes a patient information analysis unit and a related disease information analysis unit. The patient information analysis unit evaluates the patient group data group HZqt through the patient basic information data set and the historical joint ultrasound image information data set. The patient group data group HZqt is extracted according to the patient basic information data set, including age groups of patients of different genders. The related disease information analysis unit evaluates the related disease index XGzs according to the joint lesion related disease information data set and the historical joint ultrasound image information data set. The analysis and processing module sends the analyzed patient group data group HZqt and the related disease index XGzs to the evaluation output module; The evaluation output module performs a light red color mark on the current joint ultrasound image according to the evaluation result of the patient group data group GZqt; The evaluation output module divides the evaluation results of the relevant disease index XGzs into primary, intermediate and advanced levels, and marks the levels in color according to light yellow, medium yellow and dark yellow.

2. The neural network-based joint ultrasound image visual enhancement system according to claim 1, characterized in that: The expression of the patient basic information data set is: {Hx1 a 、Hx2 a 、Hx3 a 、···、Hxn a }, where Hx1 a Represents the first patient basic information in the patient basic information dataset, Hxn a represents the last patient's basic information in the patient basic information dataset, n represents the number of patients for whom basic information is collected, and a represents the patient's basic information, including the patient's age, gender, lifestyle, and family medical history.

3. The neural network-based joint ultrasound image visual enhancement system according to claim 2, characterized in that: The expression of the joint lesion related disease information dataset is: {Bx1, Bx2, Bx3, …, Bxn}, where Bx1 represents the first joint disease related disease information in the joint disease related disease information data set, Bxn represents the last joint disease related disease information in the joint disease related disease information data set, and n represents that there are n pieces of collected joint disease related disease information.

4. The neural network-based joint ultrasound image visual enhancement system according to claim 3, characterized in that: The expression of the historical joint ultrasound image information data set is: {Cx1, Cx2, Cx3, …, Cxn}, where Cx1 represents the first historical joint ultrasound image information in the historical joint ultrasound image information data set, Cxn represents the last historical joint ultrasound image information in the historical joint ultrasound image information data set, and n represents that there are n pieces of historical joint ultrasound image information collected.

5. The neural network-based joint ultrasound image visual enhancement system according to claim 4, characterized in that: The patient group data group HZqt is divided into age groups of patients of different genders according to the patient basic information data set, including female juvenile group, female middle-aged group, female elderly group, male juvenile group, male middle-aged group, and male elderly group, wherein the female juvenile group includes female juveniles aged 0 to 16 years old, the female middle-aged group includes middle-aged people aged 17 to 55 years old, the female elderly group is elderly people over 55 years old, the male juvenile group includes male juveniles aged 0 to 16 years old, the male middle-aged group includes middle-aged people aged 17 to 60 years old, and the male elderly group is elderly people over 60 years old. The labels of the female juvenile group, female middle-aged group, female elderly group, male juvenile group, male middle-aged group, and male elderly group are: Fs, Fz, Fl, Ms, Mz, Ml.

6. The neural network-based joint ultrasound image visual enhancement system according to claim 5, characterized in that: The calculation formula of HZqt of the patient population data set is as follows: In the above formula, Represents the sum of the patient basic information data set, represents the sum of historical joint ultrasound image information datasets, Represents the basic information data of x female teenagers in the patient basic information data set, Represents the basic information data of y middle-aged women in the patient basic information data set, Represents the basic information data of o elderly women in the patient basic information data set, Represents the basic information data of p male teenagers in the patient basic information data set, Represents the basic information data of q middle-aged males in the patient basic information dataset, Represents the basic information data of r elderly males in the patient basic information dataset.

7. The neural network-based joint ultrasound image visual enhancement system according to claim 6, characterized in that: When the patient group data set HZqt is compared with the current joint ultrasound image, group correspondence is performed according to the age groups of patients of different genders, and a light red color mark is generated for the corresponding group.

8. The neural network-based joint ultrasound image visual enhancement system according to claim 7, characterized in that: The calculation method of the related disease index XGzs is as follows: In the above formula, w represents the number of disease information data related to specific joint diseases.

9. The neural network-based joint ultrasound image visual enhancement system according to claim 8, characterized in that: When the relevant disease index XGzs is less than 10%, it is judged as primary, and light yellow is used for grade color identification. When the relevant disease index XGzs is greater than 10% and less than 25%, it is judged as intermediate, and medium yellow is used for grade color identification. When the relevant disease index XGzs is greater than 25%, it is judged as advanced, and dark yellow is used for grade color identification.

10. A method for visual enhancement of joint ultrasound images based on a neural network, characterized in that: The following steps are involved: Step 1: Collect the patient's basic information Hx, joint lesion-related disease information Bx, and historical joint ultrasound image information Cx, and form a patient basic information data set, a joint lesion-related disease information data set, and a historical joint ultrasound image information data set respectively; Step 2: Evaluate the HZqt of the patient group data set using the patient basic information data set and the historical joint ultrasound image information data set; Step 3: Evaluate the relevant disease index XGzs based on the joint lesion related disease information dataset and the historical joint ultrasound image information dataset; Step 4: the evaluation output module marks the current joint ultrasound image with a light red color according to the HZqt evaluation result of the patient group data group; Step 5. The evaluation results of the relevant disease index XGzs are divided into primary, intermediate and advanced levels, and the level colors are marked as light yellow, medium yellow and dark yellow respectively.

Images