Auxiliary diagnosis system for child developmental dysplasia of hip joint

By analyzing demographic characteristics and hip ultrasound images using convolutional neural networks, the selection of ultrasound or X-ray detection methods is assisted, solving the problem of ineffective detection in the diagnosis of developmental dysplasia of the hip in children aged 4-6 months, optimizing the detection process, and reducing the radiation risk to children.

CN120853843APending Publication Date: 2025-10-28THE FIRST AFFILIATED HOSPITAL OF ARMY MEDICAL UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202510867480.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-26
Publication Date
2025-10-28

AI Technical Summary

Technical Problem

In the diagnosis of developmental dysplasia of the hip in children aged 4-6 months, the combination of ultrasound and X-ray examinations increases the workload of medical staff and may cause radiation exposure to children, and cannot effectively reduce the number of invalid examinations.

Method used

Convolutional neural networks are used to analyze demographic characteristics and hip ultrasound images to determine the diagnostic needs of ultrasound examination, assist in the selection of examination methods, and reduce invalid examinations.

Benefits of technology

By reducing unnecessary ultrasound examinations, we can optimize the use of medical resources, reduce radiation risks to children, and improve diagnostic efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120853843A_ABST
    Figure CN120853843A_ABST
Patent Text Reader

Abstract

The invention relates to the field of medical systems, in particular to an auxiliary diagnosis system for child developmental dysplasia of hip joints, which comprises an information input module used for inputting multi-aspect information of children; the analysis module comprises a trained convolutional neural network, the convolutional neural network is used for training hip joint ultrasonic images of children of 4-6 months old under multi-aspect information, the convolutional neural network is used for judging whether diagnosis features in the hip joint ultrasonic images meet diagnosis requirements or not, and the convolutional neural network is used for collecting input information of the information input module; and outputting the reference probability that the ultrasonic detection meets the diagnosis requirement. By the adoption of the technical scheme, the method is used for predicting the developmental hip joint dysplasia probability of 4-6-month-old children through multi-aspect information, medical workers are assisted to make examination mode decisions, and the frequency of invalid ultrasonic detection conducted by the medical workers is reduced.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of medical systems, specifically to an auxiliary diagnostic system for developmental dysplasia of the hip in children. Background Technology

[0002] Developmental dysplasia of the hip (DDH) in children is an abnormality that occurs during hip joint development. It may manifest as a shallow acetabulum, femoral head subluxation or complete dislocation, and is commonly seen in infants born breech, those with a family history of the condition, or those who have been improperly swaddled. Early symptoms include asymmetrical skin creases on both legs, limited abduction, and limping during the toddler stage. Early treatment yields better results; Pavlik slings can be used for correction in children under 6 months, while older children may require plaster casts or surgery. If left untreated, it can lead to arthritis or limb deformities; therefore, parents should prioritize screening, avoid improper swaddling, and seek medical attention promptly if any abnormalities are observed.

[0003] In existing technologies, for example, patent publication number CN119444726A discloses an auxiliary diagnostic system for developmental dysplasia of the hip (DHD) in children. This system detects feature points and performs IHDI grading by sending images into a feature point localization model to achieve accurate diagnosis. However, pediatric imaging involves various methods, including ultrasound and X-ray. In children aged 0-4 months, secondary ossification centers of the femoral head have not yet appeared, so ultrasound is used for detection. In children over 6 months, femoral head ossification begins, and X-ray is used for detection. However, in children aged 4-6 months, the femoral head development is in a state between ossification and non-ossification. Therefore, a detection method of ultrasound followed by X-ray is typically used. This is because X-rays have radiation that may have adverse effects on children; therefore, radiation-free ultrasound is preferred. X-ray, which has some radiation, is used only when ultrasound cannot provide sufficient diagnostic features. While this process reduces the impact on children, it increases the workload of medical staff. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides an auxiliary diagnostic system for developmental dysplasia of the hip in children. This system uses multiple pieces of information to predict the probability of developmental dysplasia of the hip in children aged 4-6 months when assessed by ultrasound. It assists medical staff in making decisions regarding the selection of testing methods and reduces the number of ineffective ultrasound examinations performed by medical staff.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A diagnostic auxiliary system for developmental dysplasia of the hip in children, comprising:

[0006] Information input module: used to input the child's demographic characteristics, whether there are other malformations, delivery position, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling and binding their legs, and feeding method;

[0007] Analysis module: Includes a trained convolutional neural network. The convolutional neural network is trained based on the demographic characteristics of the sample population, delivery position, whether the infant is premature, pregnancy nutrition and complications, whether the infant has a habit of swaddling and binding legs, and the hip ultrasound images of children aged 4-6 months under different feeding methods. The hip ultrasound images are labeled with data to indicate diagnostic features. The convolutional neural network is used to determine whether the diagnostic features in the hip ultrasound images meet the diagnostic requirements. The convolutional neural network collects the input information of the information input module and outputs the reference probability that the ultrasound detection meets the diagnostic requirements.

[0008] Strategy Assist Module: When the reference probability is less than the preset value, the output diagnostic strategy is X-ray detection only; when the reference probability is greater than the preset value, the output diagnostic strategy is ultrasound detection first, and X-ray detection is performed only when ultrasound detection does not meet the diagnostic requirements.

[0009] The above approach has the following beneficial effects:

[0010] 1. In this plan, before conducting diagnostic examinations on children, medical staff first collect and input information such as the child's demographic characteristics, whether there are other malformations, fetal position at delivery, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling their legs, and feeding methods. The analysis module outputs a reference probability that ultrasound examination will meet diagnostic requirements based on the child's various characteristics. Since the femoral head of children aged 4-6 months is in a state between ossification and non-ossification, when the analysis module indicates that ultrasound examination is insufficient for diagnosis, medical staff can skip the ultrasound examination stage based on the reference probability and directly perform X-ray examination. The X-ray results will then determine whether the child has developmental dysplasia of the hip (DDH) or a specific condition, reducing unnecessary medical examinations. When the reference probability is greater than a preset value, ultrasound examination is necessary to reduce the radiation impact of X-ray examination on the child. If the ultrasound examination meets the diagnostic requirements, diagnosis based on the hip joint ultrasound image is sufficient, without the need for additional X-ray examination. If the ultrasound examination still does not meet the diagnostic requirements, then X-ray examination will be performed.

[0011] 2. In this study, demographic characteristics, presence of other malformations, fetal position at delivery, preterm birth, pregnancy nutrition and complications, swaddling habits, and feeding methods can all influence the probability, severity, and rate of bone development in children with developmental dysplasia of the hip (DDH). Therefore, a convolutional neural network is used to establish the relationship between demographic characteristics, presence of other malformations, fetal position at delivery, preterm birth, pregnancy nutrition and complications, swaddling habits, feeding methods, and whether hip ultrasound images can provide diagnostic information. This will assist in the selection of subsequent detection methods and optimize the utilization of medical resources.

[0012] Furthermore, demographic characteristics include sex, age, and family genetic history.

[0013] Beneficial effects: Child's sex, age, and family history are all important factors affecting the ultrasound appearance of the hip joint in developmental dysplasia of the hip.

[0014] Furthermore, in the context of pregnancy nutrition and complications, pregnancy nutrition includes calcium and vitamin D intake, smoking, and alcohol consumption, while complications include gestational hypertension, gestational diabetes, and amniotic fluid abnormalities.

[0015] Beneficial effects: Calcium and vitamin D intake can reflect the nutritional level of early childhood bone development, while smoking and alcohol consumption can reflect whether the child's early development was affected by the mother's unhealthy habits. Furthermore, maternal complications during pregnancy can also affect the level of early childhood bone development.

[0016] Furthermore, the convolutional neural network is used to determine whether the hip joint ultrasound image meets the diagnostic requirements based on whether the number of diagnostic features labeled in the data is greater than a preset value.

[0017] Beneficial effects: Hip ultrasound images can be used to determine whether they meet diagnostic requirements by counting the number of features that match the diagnostic criteria. When there are more features that match the diagnostic criteria, it indicates that the imaging level is better and that sufficient effective information can be obtained.

[0018] Furthermore, convolutional neural networks are also used to establish a mapping relationship between demographic characteristics, whether other malformations are present, delivery position, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling their legs, feeding methods, and the probability of a child having developmental dysplasia of the hip. Convolutional neural networks are used to obtain the input information from the information input module and output the probability of a child having developmental dysplasia of the hip.

[0019] Beneficial effects: In addition to assisting in the detection strategies used in diagnosis, it can also predict whether a child is likely to have developmental dysplasia of the hip, thereby assisting doctors in giving medical advice and precautions, and preventing potential developmental dysplasia of the hip in advance.

[0020] Furthermore, the convolutional neural network is trained based on hip ultrasound images and X-ray images of children at various ages. The convolutional neural network is used to input hip ultrasound images and X-ray images, extract diagnostic features from the children's hip ultrasound images and X-ray images, and determine whether the children have developmental dysplasia of the hip.

[0021] Beneficial effects: In addition to the detection strategies used in auxiliary diagnosis, convolutional neural networks can also assist physicians in making diagnoses based on ultrasound and X-ray images of children's hip joints. By identifying specific features in the ultrasound and X-ray images of the hip joints, it can determine whether a child has developmental dysplasia of the hip.

[0022] Furthermore, diagnostic features in hip ultrasound images include the hip joint α-angle, β-angle, femoral head coverage, and acetabular depth-width ratio.

[0023] Beneficial effects: The α angle represents the inclination of the bony portion of the acetabular roof, reflecting the development of the bony acetabulum; the β angle represents the inclination of the cartilaginous roof of the acetabulum, reflecting the cartilage coverage. Femoral head coverage and the acetabular depth-width ratio can reflect the development of a child's hip joint, thus determining whether the child has developmental dysplasia of the hip.

[0024] Furthermore, diagnostic features on X-ray images include pelvic landmarks, acetabular index, central rim of the acetabulum, and Sharp angle.

[0025] Beneficial effects: Pelvic landmarks include the bony portion of the superior acetabulum, the growth cartilage at the base of the acetabulum, the horizontal line connecting the tops of the bilateral Y-shaped cartilages, the vertical line perpendicular to the Hilgenreiner line from the outer edge of the acetabulum, and the continuous arc between the medial border of the femoral neck and the superior border of the obturator foramen. The acetabular index reflects the steepness of acetabular development, the central acetabular rim reflects the angle between the line connecting the center of the femoral head to the outer edge of the acetabulum and the vertical line, and the Sharp angle is used to reflect the acetabular tilt. By combining these parameters, it is possible to determine whether a child has developmental dysplasia of the hip.

[0026] Furthermore, both the ultrasound and X-ray images of the hip joint underwent noise reduction processing.

[0027] Beneficial effects: Noise reduction can reduce artifacts in an image, thereby improving image processing efficiency and reducing the difficulty of image processing.

[0028] Furthermore, the hip joint ultrasound image is either a two-dimensional ultrasound image or a three-dimensional ultrasound image.

[0029] Beneficial effects: Two-dimensional ultrasound images are low in cost but provide less information, while three-dimensional ultrasound images are more expensive but provide more information and are more conducive to various judgments.

[0030] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of an embodiment of the auxiliary diagnostic system for developmental dysplasia of the hip in children according to the present invention. Detailed Implementation

[0032] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0034] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0035] The following detailed description illustrates the specific implementation method:

[0036] Example 1:

[0037] As attached Figure 1 As shown: A diagnostic aid system for developmental dysplasia of the hip in children, comprising:

[0038] Information Input Module: Used to input the child's demographic characteristics, whether there are other malformations, fetal position at delivery, whether the infant is premature, pregnancy nutrition and complications, whether the infant has a habit of swaddling, and feeding method. Demographic characteristics include sex, age, and family history. Pregnancy nutrition and complications include calcium and vitamin D intake, smoking, and alcohol consumption. Complications include gestational hypertension, gestational diabetes, and amniotic fluid abnormalities.

[0039] The analysis module includes a trained convolutional neural network (CNN). This CNN is trained on hip ultrasound images of 4-6 month old infants, considering demographic characteristics of the sample population, presence of other malformations, fetal position at delivery, preterm birth, pregnancy nutrition and complications, presence of leg binding, and feeding methods. All hip ultrasound images are labeled with diagnostic features. The CNN determines whether these diagnostic features within the hip ultrasound images meet diagnostic requirements. Specifically, it checks if the number of labeled diagnostic features exceeds a preset value to determine if the hip ultrasound image meets diagnostic requirements. The CNN also collects input information from the input module and outputs a reference probability that the ultrasound detection meets diagnostic requirements.

[0040] The number of training samples for a convolutional neural network should be at least 800, and the learning rate should be set between 0.2 and 0.3.

[0041] The activation function of a convolutional neural network is:

[0042] ReLU(x) = max(0,x),

[0043] Where ReLU(x) represents the output of the activation function, and x is the input value.

[0044] Max pooling:

[0045] Output Input (i·s+m,j·s+n),

[0046] Where i and j are the position coordinates of the output feature map, m and n are the local coordinates within the pooling window, and s is the stride.

[0047] Fully connected layer:

[0048] y = Wx + b,

[0049] Where x is the flattened input vector, W is the weight matrix, and b is the bias.

[0050] Loss function:

[0051]

[0052] Where L is the scalar loss value, N is the number of categories, and y i For the i-th component of the real label, This represents the probability of the i-th category predicted by the model.

[0053] Gradient descent parameter update:

[0054]

[0055] Where W is the convolution kernel parameter and η is the learning rate. The gradient of the loss with respect to the weight W.

[0056] Strategy Assist Module: When the reference probability is less than the preset value, the output diagnostic strategy is X-ray detection only; when the reference probability is greater than the preset value, the output diagnostic strategy is ultrasound detection first, and X-ray detection is performed only when ultrasound detection does not meet the diagnostic requirements.

[0057] Before conducting diagnostic examinations on children, medical staff collect and input information such as the child's demographic characteristics, whether there are other malformations, fetal position at delivery, whether the child was premature, pregnancy nutrition and complications, whether the child had a habit of swaddling their legs, and feeding methods. Based on these characteristics, the analysis module outputs a reference probability that the ultrasound examination will meet the diagnostic requirements.

[0058] Because estrogen receptor expression leads to ligament laxity, increasing the risk of hip dislocation, female sex influences the incidence of developmental dysplasia of the hip (DDH). Furthermore, children with a family history of DDH have a significantly increased risk, making family history a contributing factor. Breech presentation, premature birth, and swaddling all increase the risk of DDH. Therefore, these parameters can effectively determine the probability and severity of DDH in children. Early or severe DDH will have more pronounced characteristics, making diagnosis easier with ultrasound.

[0059] Nutrition or complications during pregnancy, as well as feeding methods, can affect the level of bone development in early childhood, that is, change the degree and timing of ossification. The lower the degree of ossification, the better the ultrasound detection will show.

[0060] Because the femoral head of children aged 4-6 months is in a state between ossification and non-ossification, when the analysis module indicates that ultrasound examination is insufficient for diagnosis, medical staff can skip the ultrasound examination stage based on the reference probability and directly perform X-ray examination on the child. The X-ray results can then be used to determine whether the child has developmental dysplasia of the hip (DDH) or the specific condition, thus reducing unnecessary medical examinations. When the reference probability is greater than a preset value, ultrasound examination is necessary first to reduce the radiation impact of X-ray examination on the child. If the ultrasound examination meets the diagnostic requirements, diagnosis based on the hip joint ultrasound image is sufficient, without the need for additional X-ray examination. If the ultrasound examination still does not meet the diagnostic requirements, then X-ray examination is performed.

[0061] Demographic characteristics, presence of other malformations, fetal position at delivery, preterm birth, pregnancy nutrition and complications, swaddling habits, and feeding methods can all influence the probability, severity, and rate of bone development in children with developmental dysplasia of the hip (DDH). Therefore, a convolutional neural network can be used to establish the relationship between demographic characteristics, presence of other malformations, fetal position at delivery, preterm birth, pregnancy nutrition and complications, swaddling habits, feeding methods, and whether hip ultrasound images can provide diagnostic information. This will assist in the selection of subsequent detection methods and optimize the utilization of medical resources.

[0062] Example 2:

[0063] The difference from the above embodiments is that the convolutional neural network is also used to establish a mapping relationship between demographic characteristics, whether other malformations are present, delivery position, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling and binding their legs, feeding methods and the probability of the child having developmental dysplasia of the hip. The convolutional neural network is used to obtain the input information of the information input module and output the probability of the child having developmental dysplasia of the hip.

[0064] In addition to the testing strategies used in auxiliary diagnosis, it can also predict whether a child is likely to have developmental dysplasia of the hip, thereby assisting doctors in giving medical advice and precautions, and preventing potential developmental dysplasia of the hip in advance.

[0065] Example 3:

[0066] The difference from the above embodiments is that the convolutional neural network is also trained based on hip ultrasound images and X-ray images of children at different ages. The convolutional neural network is used to input hip ultrasound images and X-ray images, extract diagnostic features from the hip ultrasound images and X-ray images of children, and determine whether the child has developmental dysplasia of the hip.

[0067] Diagnostic features in hip ultrasound images include the hip joint α angle, β angle, femoral head coverage, and acetabular depth-width ratio. Diagnostic features in X-ray images include pelvic landmarks, acetabular index, acetabular central rim, and Sharp angle.

[0068] The α angle represents the inclination of the bony portion of the acetabulum, reflecting the development of the bony acetabulum, while the β angle represents the inclination of the cartilaginous roof of the acetabulum, reflecting the cartilage coverage. Femoral head coverage and the acetabular depth-to-width ratio can reflect the development of a child's hip joint, thus helping to determine if the child has developmental dysplasia of the hip.

[0069] Pelvic landmarks include the bony portion of the superior rim of the acetabulum, the growth cartilage at the base of the acetabulum, the horizontal line connecting the tops of the bilateral Y-shaped cartilages, the vertical line perpendicular to Hilgenreiner's line from the outer rim of the acetabulum, and the continuous arc between the medial border of the femoral neck and the superior border of the obturator foramen. The acetabular index reflects the steepness of acetabular development, the central rim of the acetabulum reflects the angle between the line connecting the center of the femoral head to the outer rim of the acetabulum and the vertical line, and the Sharp angle is used to reflect the acetabular tilt. These parameters are then used to determine whether a child has developmental dysplasia of the hip.

[0070] In addition to the detection strategies used in auxiliary diagnosis, convolutional neural networks can also assist physicians in making diagnoses based on ultrasound and X-ray images of children's hip joints. By identifying specific features in the ultrasound and X-ray images of the hip joints, it can determine whether a child has developmental dysplasia of the hip.

[0071] Example 4:

[0072] The difference from the above embodiments is that both the hip joint ultrasound images and X-ray images are noise-reducing processed.

[0073] Noise reduction processing can reduce artifacts in an image, thereby improving image processing efficiency and reducing the difficulty of image processing.

[0074] Example 5:

[0075] The difference from the above embodiments is that the hip joint ultrasound image is either a two-dimensional ultrasound image or a three-dimensional ultrasound image.

[0076] Two-dimensional ultrasound images are inexpensive but provide less information, while three-dimensional ultrasound images are more expensive but provide more information, making them more suitable for various image analyses and interpretations.

[0077] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. An auxiliary diagnostic system for developmental dysplasia of the hip in children, characterized in that, include: Information input module: used to input the child's demographic characteristics, whether there are other malformations, delivery position, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling and binding their legs, and feeding method; Analysis module: Includes a trained convolutional neural network. The convolutional neural network is trained on hip ultrasound images of 4-6 month old infants based on demographic characteristics of the sample population, whether there are other malformations, fetal position at delivery, whether the infant is premature, nutrition and complications during pregnancy, whether there is a habit of swaddling the legs, and feeding methods. The hip ultrasound images are labeled with data to indicate diagnostic features. The convolutional neural network is used to determine whether the diagnostic features in the hip ultrasound images meet the diagnostic requirements. The convolutional neural network collects the input information of the information input module and outputs a reference probability that the ultrasound detection meets the diagnostic requirements. Strategy Assist Module: When the reference probability is less than the preset value, the output diagnostic strategy is X-ray detection only; when the reference probability is greater than the preset value, the output diagnostic strategy is ultrasound detection first, and X-ray detection is performed only when ultrasound detection does not meet the diagnostic requirements.

2. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 1, characterized in that, Demographic characteristics include sex, age, and family genetic history.

3. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 2, characterized in that, Nutritional factors during pregnancy include calcium and vitamin D intake, smoking, and alcohol consumption. Complications include gestational hypertension, gestational diabetes, and amniotic fluid abnormalities.

4. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 3, characterized in that, Convolutional neural networks are used to determine whether hip ultrasound images meet diagnostic requirements based on whether the number of diagnostic features labeled in the data exceeds a preset value.

5. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 4, characterized in that, Convolutional neural networks are also used to establish a mapping relationship between demographic characteristics, whether other malformations are present, delivery position, whether the child is premature, pregnancy nutrition and complications, whether the child has a habit of swaddling their legs, feeding methods, and the probability of a child having developmental dysplasia of the hip. Convolutional neural networks are used to obtain the input information of the information input module and output the probability of a child having developmental dysplasia of the hip.

6. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 5, characterized in that, The convolutional neural network is also trained on hip ultrasound and X-ray images of children at different ages. The convolutional neural network is used as input to hip ultrasound and X-ray images, extracts diagnostic features from the children's hip ultrasound and X-ray images, and determines whether the children have developmental dysplasia of the hip.

7. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 6, characterized in that, Diagnostic features in hip ultrasound images include the hip joint α-angle, β-angle, femoral head coverage, and acetabular depth-width ratio.

8. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 7, characterized in that, Diagnostic features on X-rays include pelvic landmarks, acetabular index, central rim of the acetabulum, and Sharp angle.

9. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 8, characterized in that, Both ultrasound and X-ray images of the hip joint were noise-reduced.

10. The auxiliary diagnostic system for developmental dysplasia of the hip in children according to claim 9, characterized in that, An ultrasound image of the hip joint can be either a two-dimensional ultrasound image or a three-dimensional ultrasound image.

Citation Information

Patent Citations

  • Child developmental dysplasia hip joint auxiliary diagnosis system based on deep learning

    CN119444726A