Method and device for evaluating bone age of puberty elbow by using elbow side X-ray image and artificial intelligence model

Abstract

The invention relates to a bone age evaluation method by a bone age evaluation device. The bone age evaluation method comprises the following steps: (a) acquiring an X-ray image of the side surface of an elbow; (b) segmenting an olecranon region from the X-ray image; (c) analyzing the morphological characteristics of the olecranon ossification center in the segmented olecranon ossification area; and (d) determining the bone age based on the analyzed morphological characteristics.

Description

Technical Field

[0001] The present invention relates to bone age assessment technology, and more specifically, to a bone age assessment method and device using lateral X-ray images of the elbows of adolescent children. Background Art

[0002] Bone age assessment in children and adolescents is crucial for accurately assessing their physical maturity, and this may differ from their chronological age. Currently, the most widely used method for bone age assessment is hand bone age determination. However, this method suffers from inaccurate assessments during puberty, large age discrepancies, and limitations in clinical application. Therefore, elbow X-rays are widely used for bone age assessment in adolescents.

[0003] The first two years of puberty are characterized by the acceleration phase, characterized by rapid growth (growth spurts). This phase signals the peak height velocity (PHV), the period of fastest height growth during puberty, and the onset of sexual maturation at Tanner stage 2. During this period, the ossification of the olecranon osteophyte exhibits highly characteristic morphological changes. Completion of ossification marks the end of the acceleration phase (early puberty) and the beginning of the deceleration phase (late puberty). Therefore, using the rapidly changing ossification of the olecranon osteophyte to determine bone age may allow for accurate assessment of the rapid growth of early puberty.

[0004] Existing elbow bone age assessment methods that utilize the ossification process of the olecranon osteophytes include the method proposed by Sauvegrain et al. and the method proposed by Dimeglio et al. The method proposed by Sauvegrain et al. is the most traditional and widely used, but because it requires two X-ray images of the elbow, both anterior and posterior, and lateral, it results in a high dose of radiation and requires evaluation of the morphology of the various parts that make up the elbow and determination of bone age based on the total score, making interpretation more complex. The method proposed by Dimeglio et al. requires only one X-ray image of the elbow from the lateral side, but it cannot reflect the diversity of the ossification process of the olecranon osteophytes. Therefore, it has many limitations and is inaccurate in actual clinical application.

[0005] Prior art literature

[0006] Patent Literature

[0007] (Patent Document 1) Korean Patent Publication No. 10-2019-0142234. Summary of the Invention

[0008] Problems to be solved by the invention

[0009] The object of the present invention is to provide a bone age assessment method and device that can morphologically subdivide the ossification center formation and fusion stages presented in the maturation process of the ulnar olecranon osteophyte that can be confirmed from the lateral elbow X-ray in adolescence, and reflect the diverse and characteristic change process.

[0010] The problems to be solved by the present invention are not limited to the above problems, and those skilled in the art will clearly understand other problems not mentioned from the following description.

[0011] Solutions for solving problems

[0012] The first aspect of the present invention for achieving the above-mentioned purpose provides a bone age assessment method performed by a bone age assessment device, comprising: (a) acquiring an X-ray image of the side of the elbow; (b) segmenting the olecranon region from the X-ray image; (c) analyzing the morphological characteristics of the ossification center of the olecranon in the segmented olecranon region; and (d) determining the bone age based on the analyzed morphological characteristics.

[0013] Preferably, the step (b) may include: predicting a bone region in the X-ray image in pixel units; and segmenting a region corresponding to the olecranon from the X-ray image based on a result of the pixel-based prediction.

[0014] Preferably, the step (c) may include: determining a region corresponding to the ossification center of the olecranon in the olecranon region; and determining the morphological characteristics of the ossification center of the olecranon.

[0015] Preferably, the step (d) may include: classifying the olecranon bone age based on the morphological characteristics of the ossification center of the olecranon and according to pre-set olecranon characteristic standards for different ages.

[0016] Preferably, in the pre-set olecranon feature standards for each age group, the standards can be set in units of 6 months for girls aged 9.5 to 13 years old, and in units of 6 months for boys aged 11.5 to 15 years old.

[0017] Preferably, in the pre-set olecranon feature standards for different ages, olecranon feature standards for girls aged 11.25 and boys aged 13.25 can also be set.

[0018] Preferably, according to the fusion order of the ossification center and the accessory ossification center of the olecranon, the preset characteristic standards of the olecranon at different ages can be classified into the first ossification process and the second ossification process.

[0019] Preferably, the first ossification process may include: a stage of a small ulnar ossification center of the ulnar olecranon that is more than 20% and less than 50% of the height of the main body of the ulnar olecranon; a stage of an enlarged ulnar ossification center of the ulnar olecranon that is more than 50% of the height of the main body of the ulnar olecranon; a stage of the appearance of an accessory ossification center; a stage of fusion of the ulnar olecranon ossification center and the accessory ossification center; a stage of enlargement of the fused ossification center; a stage of partial fusion of less than 50% of the ulnar olecranon osteophytes; a stage of partial fusion of more than 50% of the ulnar olecranon osteophytes and a stage of complete fusion of the ulnar olecranon osteophytes.

[0020] Preferably, the stage between the fusion stage of the ossification center of the olecranon and the accessory ossification center and the enlargement stage of the fused ossification center may further include: a stage in which other accessory ossification centers appear after the fusion of the ossification center of the olecranon and the accessory ossification center.

[0021] Preferably, the second ossification process may include a stage of a small ulnar ossification center that is more than 20% and less than 50% of the height of the main body of the ulnar olecranon; a stage of an enlarged ulnar ossification center that is more than 50% of the height of the main body of the ulnar olecranon; a stage of the appearance of an accessory ossification center; a stage of the fusion of the accessory ossification center with the main body of the ulnar olecranon; a stage of the enlargement of the fused accessory ossification center-the main body of the ulnar olecranon and the unfused ulnar olecranon ossification center; a stage of partial fusion of the ulnar olecranon ossification center and the fused accessory ossification center-less than 50% of the main body of the ulnar olecranon; a stage of partial fusion of more than 50% of the ulnar olecranon osteophytes and a stage of complete fusion of the ulnar olecranon osteophytes.

[0022] Preferably, the stage between the fusion stage of the accessory ossification center and the main body of the ulnar olecranon and the enlargement stage of the fused accessory ossification center - the main body of the ulnar olecranon and the unfused ulnar olecranon ossification center may also include: a stage in which other accessory ossification centers appear after the fusion of the ulnar olecranon ossification center and the accessory ossification center.

[0023] A second aspect of the present invention for achieving the aforementioned purpose provides a bone age assessment device, comprising: an image acquisition unit for acquiring an X-ray image of the side of the elbow; a region segmentation unit for segmenting the olecranon region from the X-ray image; a morphological analysis unit for analyzing the morphological features of the ossification center of the olecranon in the segmented olecranon region; and a bone age determination unit for determining bone age based on the analyzed morphological features.

[0024] Preferably, the region segmentation unit may be used to predict a bone region in the X-ray image in pixel units, and segment the region corresponding to the olecranon from the X-ray image based on the result of the pixel-based prediction.

[0025] Preferably, the morphological analysis unit can be used to determine a region corresponding to the ossification center of the olecranon in the ulna region, and determine the morphological characteristics of the ossification center of the olecranon.

[0026] Preferably, the bone age determination unit can be used to classify the bone age of the olecranon based on the morphological characteristics of the ossification center of the olecranon and according to pre-set olecranon characteristic standards for different ages.

[0027] A third aspect of the present invention for achieving the aforementioned object provides a computer program stored on a computer-readable medium, which performs a bone age assessment method when instructions of the computer program are executed.

[0028] Effects of the Invention

[0029] According to the present invention as described above, by using a single X-ray image of the side of the elbow, the radiation exposure during the examination can be reduced, and the age range that can be determined can be expanded and refined.

[0030] In addition, by reflecting the characteristic morphological changes of the olecranon osteophytes that can be observed in lateral elbow X-ray images, the classification can more accurately assess the bone age in various clinical cases when determining the bone age of adolescents.

[0031] In addition, accurate adolescent bone age assessment can play a key role in evaluating the musculoskeletal maturity of children, diagnosing various endocrine diseases, short stature, fractures, scoliosis or lower limb length differences, and determining the timing of treatment.

[0032] In addition, by improving the workflow of radiologists, simpler and more intuitive judgments can be achieved, and using AI programs to assist radiologists, pediatric and adolescent surgeons, or plastic surgeons can achieve rapid interpretation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a block diagram of the bone age assessment method of a preferred embodiment of the present invention.

[0034] Figure 2 FIG. 1 is a flowchart illustrating a bone age assessment method according to an embodiment.

[0035] Figure 3 This is an illustrative diagram for illustrating a bone age assessment method according to an embodiment.

[0036] Figure 4 This is an illustrative diagram for explaining the segmentation of the olecranon region according to one embodiment.

[0037] Figure 5 This is an illustrative diagram for explaining analysis of morphological features of ossification centers according to one embodiment.

[0038] Figure 6 and Figure 7This is an example diagram for illustrating a deep learning algorithm according to an embodiment.

[0039] Figure 8 and Figure 9 This is an illustrative diagram for explaining bone age determination according to one embodiment.

[0040] Figure 10 This is an illustrative diagram for illustrating the performance of a bone age assessment method according to one embodiment.

[0041] Description of reference numerals:

[0042] 100: Bone Age Assessment Device

[0043] 110: Image Acquisition Department

[0044] 120: Area Segmentation Department

[0045] 130: Morphological Analysis Department

[0046] 140: Bone Age Determination Department

[0047] 150: Control Department DETAILED DESCRIPTION

[0048] Below, the advantages and features of the present invention, as well as the methods for achieving these advantages and features, will gradually become clear in the process of the embodiments described in detail below in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but can be embodied in various different forms. These embodiments are only used to fully disclose the present invention and are provided to fully inform the scope of the invention to those skilled in the art to which the present invention belongs. The present invention is defined by the claims. The same reference symbols throughout the entire specification refer to the same constituent elements. "And / or" includes each of the items involved and all combinations of more than one thereof.

[0049] Although "first," "second," and the like are used to describe various elements, constituents, and / or parts, these elements, constituents, and / or parts are not limited by these terms. These terms are used only to distinguish one element, constituent, or part from other elements, constituents, or parts. Therefore, it is obvious that the first element, first constituent, or first part mentioned below may also be the second element, second constituent, or second part within the technical concept of the present invention.

[0050] In addition, the identifiers in each step (e.g., a, b, c, etc.) are only used for convenience of description and do not indicate the order of the steps. Unless the context clearly indicates a specific order, the steps may occur in an order different from the order indicated. In other words, the steps may occur in the same order as the order indicated, may be performed substantially simultaneously, or may be performed in the reverse order.

[0051] The terms used in this specification are intended to illustrate the embodiments and are not intended to limit the present invention. In this specification, unless otherwise expressly stated, the singular also includes the plural. The terms "comprises" and / or "comprising" used in this specification to refer to constituent elements, steps, actions, and / or elements do not exclude the presence or addition of one or more other constituent elements, steps, actions, and / or elements mentioned.

[0052] Unless otherwise defined, all terms (including technical and scientific terms) used in this specification should be used with the meanings that are generally understood by those skilled in the art in the technical field of the present invention. In addition, unless clearly defined, terms defined in commonly used dictionaries should not be idealized or over-interpreted.

[0053] Furthermore, when describing the embodiments of the present invention, if it is determined that a detailed description of a known function or configuration would unnecessarily obscure the main purpose of the present invention, its detailed description will be omitted. Furthermore, the terms described below are defined based on the functions in the embodiments of the present invention and may vary depending on the intentions or habits of the user or operator. Therefore, these definitions should be based on the overall content of this specification.

[0054] Figure 1 2 is a block diagram showing a bone age assessment device according to a preferred embodiment of the present invention.

[0055] Bone age assessment device 100 is used to perform the bone age assessment method of the present invention. It assesses bone age by receiving X-ray images of the lateral portion of the elbow and analyzing the bone shape. Specifically, bone age assessment device 100 utilizes the characteristics of the olecranon ossification process, which is characteristic of the acceleration phase two years before puberty, to assess the bone age of adolescents.

[0056] Preferably, the bone age assessment device 100 is a computer that can install and execute an application or program for the bone age assessment method and has a user interface (User Interface) so as to control the input and output of data. Among them, a computer refers to all types of hardware devices including at least one processor, and according to an embodiment, it can be understood to also include a software configuration for actions on the hardware device. For example, a computer can be understood as but not limited to a smartphone, a tablet computer, a desktop computer, a laptop computer, and all user clients and applications driven on each device.

[0057] Reference Figure 1The bone age assessment device 100 includes an image acquisition unit 110, a region segmentation unit 120, a morphology analysis unit 130, a bone age determination unit 140, and a control unit 150. The control unit 150 controls the operations and data flows of the image acquisition unit 110, the region segmentation unit 120, the morphology analysis unit 130, and the bone age determination unit 140.

[0058] The image acquisition unit 110 acquires an X-ray image. Preferably, the X-ray image can be captured by an independent imaging device connected to the bone age assessment device 100 by wire or wirelessly and transmitted to the bone age assessment device 100, or the X-ray image can be captured by an imaging device included in the bone age assessment device 100 and transmitted to the image acquisition unit 110.

[0059] The region segmentation unit 120 segments the olecranon region, which is a region required for determining bone age, from the X-ray image acquired by the image acquisition unit 110. Preferably, various deep learning algorithms can be applied to the method of segmenting the olecranon region.

[0060] The morphological analysis unit 130 analyzes the bone morphological features in the olecranon region. Preferably, the morphological analysis unit 130 can use a deep learning algorithm to determine the region corresponding to the ossification center of the olecranon and analyze the morphological features of each bone constituting the ossification center of the olecranon.

[0061] The bone age determination unit 140 determines the bone age based on the morphological characteristics of the bones. Preferably, the bone age determination unit 140 can determine the bone age according to a pre-set bone age determination standard that matches the morphological characteristics of the bones with the bone age.

[0062] Below, refer to Figure 2 , through Figure 1 The actions performed by each structure of the bone age assessment device 100 shown in FIG. Figure 2 The steps described are performed by different structures, but the present invention is not limited thereto. According to embodiments, at least a portion of the steps may be performed by the same or different structures.

[0063] Figure 2 FIG. 1 is a flowchart illustrating a bone age assessment method according to an embodiment.

[0064] Reference Figure 2 , the image acquisition unit 110 acquires an X-ray image of the side of the elbow (step S210). Figure 3 The image acquisition unit 110 can pre-process the X-ray image. That is, the image acquisition unit 110 can exclude unnecessary areas from the X-ray image, identify and select a lateral view, and perform normalization.

[0065] The region segmentation unit 120 segments the olecranon region from the X-ray image (step S220). Preferably, the region segmentation unit 120 can apply a deep learning algorithm to the X-ray image, predict the bone region in pixel units, and segment the region corresponding to the olecranon from the X-ray image based on the prediction result in pixel units. More specifically, referring to Figure 3 The region segmentation unit 120 can apply YOLOv5, a deep learning algorithm, to the X-ray image as the input image to detect the bone portion corresponding to the olecranon. The region segmentation unit 120 can apply the YOLOv5 network to automatically crop the olecranon region from the X-ray image and determine the cropped olecranon region image as the region of interest (ROI). For example, referring to Figure 4 , Figure 4 (a) corresponds to the original X-ray image, and (b) corresponds to the image of the segmented olecranon region. The region segmentation unit 120 can apply YOLOv5, the first step of the cascade model, to the X-ray image shown in (a) to detect the bone portion corresponding to the olecranon. As shown in (b), only the region corresponding to the olecranon is segmented to obtain an image. That is, the region segmentation unit 120 uses the YOLOv5 network to crop only the olecranon region from the X-ray image and extract it as the region of interest. Based on the extracted region of interest, step S230 described below can be performed.

[0066] In one embodiment, the region segmentation unit 120 can apply YOLOv5 to the input X-ray image to detect the region corresponding to the olecranon, and segment the region corresponding to the olecranon using EfficientDet. EfficientDet is a deep learning algorithm based on the EfficientNet model backbone, with an additional segmentation layer, which can be used to segment the olecranon region from the region segmentation unit 120.

[0067] The morphological analysis unit 130 analyzes the morphological characteristics of the olecranon ossification center in the segmented olecranon region (step S230). Preferably, the morphological analysis unit 130 can apply a deep learning algorithm to the olecranon region to determine the region corresponding to the olecranon ossification center to determine the morphological characteristics of the olecranon ossification center. Figure 3The morphological analysis unit 130 can use the deep learning algorithm of cascade training, namely EfficientDet, to classify the morphological features of the ossification center of the olecranon. That is, the morphological analysis unit 130 applies EfficientDet to the segmented olecranon region to classify the olecranon region and osteophytes. Among them, EfficientDet uses a compound scaling that adjusts depth, width, and resolution at the same time to achieve high efficiency and accuracy. It is conducive to conducting accurate object investigations on the elbow appearance that varies from person to person. For example, referring to Figure 5 , Figure 5 (a) corresponds to the olecranon region obtained by the region segmentation unit 120, Figure 5 (b) corresponds to the ulnar olecranon ossification center region that serves as a judgment standard for the morphological characteristics of the ulnar olecranon ossification center. The morphological analysis unit 130 can apply EfficientDet, a deep learning algorithm, to the ulnar olecranon region shown in (a) to determine the region corresponding to the ulnar olecranon ossification center of the ulnar olecranon region shown in (b), and perform morphological characteristic judgment on the ulnar olecranon ossification center region. That is, EfficientDet, a deep learning algorithm, is based on Figure 5 The highlight area shown in (b) is used to analyze the morphological characteristics.

[0068] More specifically, EfficientDet, a deep learning algorithm applied to the region segmentation unit 120, the morphological analysis unit 130, and the bone age determination unit 140 described later, is introduced into the backbone of the EfficientNet model and is configured to be suitable for the present invention, such as Figure 6As shown, the EfficientNet on the left serves as the backbone. This backbone EfficientNet is applied to b4 (i.e., EfficientNet-b4). Four Bi-directional Feature Pyramid Network (BiFPN) layers are applied at the end as the feature network. The Box Network (BoxNetwork) is removed from the level network that classifies objects and the box network that determines the bounding box position and size of objects, and a segmentation layer is added. The "b4" in EfficientNet-b4 is a number applied based on the layer depth, such as b1, b2, or b4. The layer depth can be applied or changed in various ways. This deep learning algorithm adds a segmentation prediction logit that predicts the segmentation of X-ray images on a pixel-by-pixel basis. This segmentation prediction logit predicts bone regions in X-ray images on a pixel-by-pixel basis. This improved EfficientDet reduces error rates and eliminates the need to examine every bounding box, resulting in faster computation.

[0069] Reference Figure 7, as the backbone structure of EfficientDet, the Efficient-Net used as the backbone network includes multiple convolution (Convolution, Conv) layers, MBconv blocks, conv 1x1 and pooling & fully connected (FC) layers. In Conv3x3, a 3x3 kernel with 32 channels and a convolution layer are stacked, and then an MBConv1 block with a 3x3 kernel and 16 channels, two MBConv6 blocks with a 3x3 kernel and 24 channels, two MBConv6 blocks with a 5x5 kernel and 40 channels, and three MBConv6 blocks with a 3x3 kernel and 80 channels are connected. Among them, MBConv6 is again subjected to deep batch normalization and swish process in MBconv. Finally, a convolution layer (Convlayer) with a 1x1 kernel, a pooling layer, and a fully connected layer as a dense layer are stacked. Since the data input into the deep learning algorithm is a two-dimensional X-ray image rather than a three-dimensional z-axis depth image, the layers are stacked in units of 4 to input->P1 / 2->P2 / 4->P3 / 8->P4 / 16->P5 / 32 in this way, and the features of different scales are integrated by segmenting the BiFPN layer, so as to improve the accuracy of object detection. This is equivalent to a network structure in which the segmentation logit is attached to the layer passing through the BiFPN layer to perform segmentation of the elbow area and classification of the area at the end. At this time, the layers can be stacked deeper as needed. Among them, segmentation is the segmentation of identifying the precise boundary of the elbow area and distinguishing it from other areas by the region segmentation unit 120. The region segmentation unit 120 can also use EfficientDet to segment the ulnar olecranon area, and classification is the bone age determination performed by the bone age determination unit 140 described later.

[0070] The bone age determination unit 140 determines the bone age based on the morphological characteristics of the ossification center of the ulna olecranon analyzed in step S230 (step S240). Preferably, the bone age determination unit 140 can apply EfficientDet, a deep learning algorithm, to classify the ulna olecranon bone age based on the morphological characteristics of the ossification center of the ulna olecranon according to pre-set ulna olecranon feature standards for each age group. The deep learning algorithm is the same as the algorithm applied to step S230.

[0071] Preferably, among the pre-set olecranon feature standards for each age group, the standard can be set in units of 6 months for girls aged 9.5 to 13 years old, and the standard can be set in units of 6 months for boys aged 11.5 to 15 years old. In particular, the olecranon feature standards can also be set for girls aged 11 to 11.5 years old, i.e., 11.25 years old, and for boys aged 13 to 13.5 years old, i.e., 13.25 years old. Figure 8 As can be seen, the pre-set standard table of olecranon characteristics for each age group includes the characteristics and shapes of the olecranon for girls (F) and boys (M). An 11-year-old girl and a 13-year-old boy, an 11.5-year-old girl and a 13.5-year-old boy, and a 12-year-old girl and a 14-year-old boy will exhibit two morphologies corresponding to the D and D' stages, the L and L' stages, and the G and G' stages, respectively. These correspond to the morphologies that may occur during ossification according to the order of fusion.

[0072] More specifically, the pre-set characteristic standards for the olecranon at each age can be categorized into the primary ossification process and the secondary ossification process based on the order of fusion of the ossification center and the accessory ossification center of the olecranon. That is, during the ossification process, if the ossification center of the olecranon fuses with the accessory ossification center first after the accessory ossification center appears, the process progresses to the primary ossification process. If the accessory ossification center fuses with the main body of the olecranon first, the process progresses to the secondary ossification process.

[0073] First, refer to Figure 9 The first ossification process may include:

[0074] - stage O2 (A small ossification center (20-50% height of olecranonbody)) corresponding to 9.5 years of age in girls and 11.5 years of age in boys;

[0075] - corresponds to stage O3 (An enlarged olecranon ossification center (>50% height of olecranon body)) in girls aged 10 years and boys aged 12 years;

[0076] - Stage B (Appearance of an accessory ossification center (two ossification centers)) corresponding to 10.5 years for girls and 12.5 years for boys;

[0077] - Stage D (Fusion of two ossification centers (a half-moon shape)) corresponding to 11 years of age in girls and 13 years of age in boys;

[0078] - Stage L (Enlargement of fused ossification centers (a rectangular shape)) corresponding to 11.5 years in girls and 13.5 years in boys;

[0079] - Stage G (Partial fusion (<50%) of olecranon apophysis) corresponding to 12 years of age in girls and 14 years of age in boys;

[0080] - Stage H (Partial fusion (>50%) of olecranon apophysis) corresponding to 12.5 years of age in girls and 14.5 years of age in boys; and

[0081] - corresponds to stage F (Complete fusion of olecranon apophysis) in girls aged 13 years and boys aged 15 years.

[0082] Next, refer to Figure 9 , the second ossification process may include:

[0083] - stage O2 (A small ossification center (20-50% height of olecranonbody)) corresponding to 9.5 years of age in girls and 11.5 years of age in boys;

[0084] - corresponds to stage O3 (enlarged olecranon ossification center (>50% height of olecranon body)) in girls aged 10 years and boys aged 12 years;

[0085] - Stage B (Appearance of an accessory ossification center (two ossification centers)) corresponding to 10.5 years for girls and 12.5 years for boys;

[0086] - D' stage (Fusion of accessory ossification center and olecranon body) corresponding to 11 years of age in girls and 13 years of age in boys;

[0087] - L' stage (Enlargement of unfused olecranonossification center and fused accessory ossification center-olecranon body) corresponding to 11.5 years of age in girls and 13.5 years of age in boys;

[0088] - G' stage (Partial fusion (<50%) of fused accessory ossification center-olecranon body and olecranon ossification center) corresponding to 12 years of age in girls and 14 years of age in boys);

[0089] *- corresponds to stage H (Partial fusion (>50%) of olecranon apophysis) in girls aged 12.5 years and boys aged 14.5 years; and

[0090] - corresponds to stage F (Complete fusion of olecranon apophysis) in girls aged 13 years and boys aged 15 years.

[0091] Among them, the period between the D stage and the L stage of the first ossification process and between the D' step and the L' step of the second ossification process may also include the BD stage (Appearance of another accessory ossification center (after fusion of previous ossification centers; a half-moonshape+additional accessory ossification center)) corresponding to 11.25 years old for girls and 113.25 years old for boys) in the first ossification process and the D' step in the second ossification process.

[0092] That is, the bone age determination unit 140 determines the bone age as the age corresponding to the stage at which the morphological characteristics of the ossification center of the olecranon obtained by the morphological analysis unit 130 match each stage of the pre-set olecranon characteristic standards for each age.

[0093] Figure 10 Used to illustrate the performance of a bone age assessment method according to one embodiment.

[0094] Reference Figure 10 , showing the consistency between two pediatric imaging physician observers, it can be seen that the bone age assessment methods of the present invention (Novel classification and Olecranon BA (Olecranon Bone Age)) have higher consistency than the existing ulnar olecranon bone age assessment methods, namely Elbow BA_Dimeglio and Elbow BA_Sauvegrain. Among them, the novel classification (Novel classification) shows the results of analysis based on the stage represented by 02 to F as the standard, and Olecranon BA shows the results of analysis based on age as the standard. Stage D and stage D', stage L and stage L', and stage G and stage G' respectively represent the same bone age but different stage classifications, so a differentiated analysis was performed.

[0095] In addition, referring to the following [Table 1], it can be seen that the bone age assessment method of the present invention shows higher reliability when compared with the existing elbow bone age assessment methods, namely Sauvegrain (Elbow BA_Sauvegrain) and Dimeglio (Elbow BA_Dimeglio) and the existing hand bone age assessment methods, namely Greulich-Pyle and Tanner-Whitehouse hybrid (Hand BA_GP / TW3) and the Korean children's standard bone age (Hand BA_KS).

[0096] [Table 1]

[0097]

[0098]

[0099] Moreover, the AI ​​model of the deep learning algorithm applied to the bone age assessment method of the present invention presents high accuracy and low error rate. Specifically, the internal data set of the novel classification (Novel classification) shows an accuracy of 0.96, a sensitivity of 0.80 and a specificity of 0.98, and the external data set shows an accuracy of 0.87, a sensitivity of 0.76 and a specificity of 0.91. In addition, the internal validation of the ulnar olecranon bone age (Olecranon bone age) presents an MSE of 0.088 and an RMSE of 0.296, and the external validation presents an MSE of 0.125 and an RMSE of 0.398. Among them, internal validation is used to evaluate whether the algorithm model or classification method applied for the bone age assessment method meets the purpose of training data, and external validation is used to evaluate the popularity of the algorithm model or classification method for new data.

[0100] On the one hand, the steps of the method or algorithm described in conjunction with the embodiments of the present invention can be implemented directly in hardware, or can be implemented as a software module executed by hardware, or can be implemented as a combination of hardware and software modules. The software module can reside in random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or any other form of computer-readable recording medium known in the art to which the present invention belongs.

[0101] The constituent elements of the present invention can be implemented as programs (or applications) that are combined with computer hardware and stored on a medium for execution. The constituent elements of the present invention can be implemented as software programs or software elements. Similarly, embodiments can be implemented in programming languages ​​or scripting languages ​​such as C, C++, Java, assemblers, etc., including various algorithms implemented in the form of combinations of data structures, processes, routines, or other programming elements. Functional aspects can be implemented as algorithms executed on more than one processor.

[0102] Although the above describes the preferred embodiments of the bone age assessment method and device of the present invention, the present invention is not limited thereto. Various modifications can be implemented within the scope of the claims, specific embodiments and drawings of the present invention, and these modifications also belong to the present invention.

Claims

1. A bone age assessment method using a bone age assessment device, characterized in that: include: (a) Obtain an X-ray image of the lateral aspect of the elbow; (b) segmenting the olecranon region from the X-ray image; (c) analyzing the morphological characteristics of the ossification center of the olecranon in the segmented olecranon region; as well as (d) Determine bone age based on the analyzed morphological features.

2. The bone age assessment method according to claim 1, characterized in that: The step (b) comprises: Predicting a bone region in the X-ray image in pixel units; and According to the pixel-based prediction result, the region corresponding to the olecranon is segmented from the X-ray image.

3. The bone age assessment method according to claim 2, characterized in that: The step (c) comprises: Determining a region corresponding to the ossification center of the olecranon in the olecranon region; and Determine the morphological characteristics of the ossification center of the olecranon.

4. The bone age assessment method according to claim 3, characterized in that: The step (d) comprises: Based on the morphological characteristics of the ossification center of the ulnar olecranon, the bone age of the ulnar olecranon is classified according to the pre-set characteristic standards of the ulnar olecranon at different ages.

5. The bone age assessment method according to claim 4, characterized in that: In the pre-set olecranon feature standards for each age group, the standards are set in units of 6 months for girls aged 9.5 to 13 years, and the standards are set in units of 6 months for boys aged 11.5 to 15 years.

6. The bone age assessment method according to claim 5, characterized in that: Among the pre-set olecranon feature standards for each age group, olecranon feature standards for 11.25-year-old girls and 13.25-year-old boys are also set.

7. The bone age assessment method according to claim 4, characterized in that: According to the fusion order of the ossification center and the accessory ossification center of the olecranon, the preset characteristic standards of the olecranon at different ages are classified into the first ossification process and the second ossification process.

8. The bone age assessment method according to claim 7, characterized in that: The first ossification process includes: The olecranon is a small ossification center stage with a height of more than 20% and less than 50% of the main body of the olecranon; The height of the olecranon is more than 50% of the central stage of ossification of the olecranon; The stage of emergence of paraossification centers; The fusion stage of the ossification center and accessory ossification center of the olecranon; the enlargement phase of the fused ossification center; Partial fusion stage of less than 50% of the olecranon osteophyte; Partial fusion of more than 50% of the olecranon osteophyte; and Complete fusion stage of the olecranon osteophyte.

9. The bone age assessment method according to claim 8, characterized in that: Between the fusion stage of the olecranon ossification center and the accessory ossification center and the enlargement stage of the fused ossification center, the following steps are also included: The stage at which other accessory ossification centers appear follows the fusion of the ossification center and the accessory ossification center of the olecranon.

10. The bone age assessment method according to claim 7, characterized in that: The second ossification process includes: The olecranon is a small ossification center stage with a height of more than 20% and less than 50% of the main body of the olecranon; The height of the olecranon is more than 50% of the central stage of ossification of the olecranon; The stage of emergence of paraossification centers; the stage of fusion of the accessory ossification center with the main body of the olecranon; Fused accessory ossification center - stage of enlargement of the main body of the olecranon and the unfused ossification center of the olecranon; Ossification center of the olecranon and fused accessory ossification center - the stage of partial fusion of less than 50% of the main body of the olecranon; Partial fusion of more than 50% of the olecranon osteophyte; and Complete fusion stage of the olecranon osteophyte.

11. The bone age assessment method according to claim 10, characterized in that: The stage between the fusion of the accessory ossification center and the main body of the olecranon and the enlargement of the fused accessory ossification center-the main body of the olecranon and the unfused olecranon ossification center also includes: The stage at which other accessory ossification centers appear follows the fusion of the ossification center and the accessory ossification center of the olecranon.

12. A bone age assessment device, characterized in that: include: An image acquisition unit, used for acquiring an X-ray image of the side of the elbow; a region segmentation unit, configured to segment the olecranon region from the X-ray image; A morphological analysis unit, configured to analyze the morphological characteristics of the ossification center of the olecranon in the segmented olecranon region; as well as The bone age determination unit is used to determine the bone age based on the analyzed morphological features.

13. The bone age assessment device according to claim 12, characterized in that: The region segmentation unit is used to predict the bone region in the X-ray image in pixel units, and segment the region corresponding to the olecranon from the X-ray image based on the result of the pixel unit prediction.

14. The bone age assessment device according to claim 13, characterized in that: The morphological analysis unit is used to determine a region corresponding to the ossification center of the olecranon in the ulna region and to determine the morphological characteristics of the ossification center of the olecranon.

15. The bone age assessment device according to claim 14, characterized in that: The bone age determination unit is used to classify the bone age of the olecranon based on the morphological characteristics of the ossification center of the olecranon and according to pre-set olecranon characteristic standards for different ages.

16. The bone age assessment device according to claim 15, characterized in that: In the pre-set olecranon feature standards for each age group, the standards are set in units of 6 months for girls aged 9.5 to 13 years, and the standards are set in units of 6 months for boys aged 11.5 to 15 years.

17. The bone age assessment device according to claim 16, characterized in that: In the pre-set olecranon feature standards for each age group, olecranon feature standards for girls aged 11.25 and boys aged 13.25 are also set.

18. The bone age assessment device according to claim 15, characterized in that According to the fusion order of the ossification center and the accessory ossification center of the olecranon, the preset characteristic standards of the olecranon at different ages are classified into the first ossification process and the second ossification process.

19. A computer program stored on a computer-readable medium, characterized in that When the instructions of the computer program are executed, the method according to any one of claims 1 to 11 is performed.

Images