CN113610809B - Fracture detection method, fracture detection device, electronic equipment and storage medium - Google Patents
Fracture detection method, fracture detection device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN113610809B CN113610809B CN202110907714.0A CN202110907714A CN113610809B CN 113610809 B CN113610809 B CN 113610809B CN 202110907714 A CN202110907714 A CN 202110907714A CN 113610809 B CN113610809 B CN 113610809B
- Authority
- CN
- China
- Prior art keywords
- fracture
- name information
- bone image
- bone
- detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 170
- 210000000988 bone and bone Anatomy 0.000 claims abstract description 182
- 238000000034 method Methods 0.000 claims description 25
- 238000013145 classification model Methods 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 12
- 238000002372 labelling Methods 0.000 claims description 6
- 238000003745 diagnosis Methods 0.000 abstract description 9
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 238000013135 deep learning Methods 0.000 abstract description 2
- 208000010392 Bone Fractures Diseases 0.000 description 191
- 238000010586 diagram Methods 0.000 description 18
- 238000004891 communication Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 6
- 210000002411 hand bone Anatomy 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000012706 support-vector machine Methods 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000011176 pooling Methods 0.000 description 4
- 210000003813 thumb Anatomy 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000034656 Contusions Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000009519 contusion Effects 0.000 description 1
- 210000003811 finger Anatomy 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 210000000707 wrist Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30008—Bone
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Image Analysis (AREA)
Abstract
The disclosure provides a fracture detection method, a fracture detection device, electronic equipment and a storage medium, relates to the technical field of artificial intelligence, in particular to the technical field of computer vision and deep learning, and can be applied to fracture detection and other scenes. The specific implementation scheme is as follows: acquiring a bone image to be detected; outputting a preliminary detection result indicating whether or not the bone image has a fracture site by inputting the bone image to the fracture detection model; marking the outline of the fracture part in the bone image by inputting the bone image into the outline recognition model when the primary detection result indicates that the fracture part exists; and determining the name information of the fracture part based on the outline of the fracture part in the bone image, and generating a detection report containing the name information of the fracture part. Specific information of the fracture part can be intuitively obtained based on the detection report, and the diagnosis efficiency and accuracy of doctors can be greatly improved.
Description
Technical Field
The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of computer vision and deep learning, and can be applied to scenes such as fracture detection.
Background
The existing method for detecting the fracture condition through the bone image can only judge whether the bone structure displayed by the bone image is fractured or not, or can only locate the bone structure to the approximate area of the fracture part, and can not further output more detailed information of the fracture part, so that the diagnosis efficiency can not be effectively improved.
Disclosure of Invention
The present disclosure provides a fracture detection method, a fracture detection device, an electronic device, and a storage medium.
According to a first aspect of the present disclosure, there is provided a fracture detection method comprising:
acquiring a bone image to be detected;
outputting a preliminary detection result indicating whether or not the bone image has a fracture site by inputting the bone image to the fracture detection model;
marking the outline of the fracture part in the bone image by inputting the bone image into the outline recognition model when the primary detection result indicates that the fracture part exists;
and determining the name information of the fracture part based on the outline of the fracture part in the bone image, and generating a detection report containing the name information of the fracture part.
According to a second aspect of the present disclosure, there is provided a fracture detection device comprising:
the image acquisition module is used for acquiring a bone image to be detected;
the preliminary detection module is used for outputting a preliminary detection result indicating whether a bone image has a fracture part or not by inputting the bone image into the fracture detection model;
the contour labeling module is used for labeling the contour of the fracture part in the bone image by inputting the bone image into the contour recognition model when the primary detection result indicates that the fracture part exists;
and the result output module is used for determining the name information of the fracture part based on the outline of the fracture part in the skeleton image and generating a detection report containing the name information of the fracture part.
According to a third aspect of the present disclosure, there is provided an electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the fracture detection method described above.
According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the fracture detection method described above.
According to a fifth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the fracture detection method described above.
According to a sixth aspect of the present disclosure, there is provided a fracture detection apparatus comprising the electronic apparatus provided by the embodiments of the present disclosure.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.
The beneficial effects that this disclosure provided technical scheme brought are:
in the technical scheme of the disclosure, after the existence of the fracture part in the bone image is determined, the specific position and the name information of the fracture part can be further and accurately identified, a detection report containing the name information of the fracture part is generated, the specific information of the fracture part can be intuitively obtained based on the detection report, and the diagnosis efficiency and accuracy of doctors can be greatly improved.
Drawings
The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
FIG. 1 illustrates a model architecture diagram to which a fracture detection method can be applied, provided in an embodiment of the present disclosure;
FIG. 2 shows a flow diagram of a fracture detection method provided by an embodiment of the present disclosure;
FIG. 3 shows an exemplary architecture diagram of a Faster-RCNN model provided in an embodiment of the disclosure;
FIG. 4 illustrates an exemplary structural schematic diagram of a Mask-RCNN model provided in an embodiment of the disclosure;
FIG. 5 illustrates a flow diagram of another fracture detection method provided by embodiments of the present disclosure;
FIG. 6 illustrates a hand bone image provided by an embodiment of the present disclosure;
FIG. 7 is a schematic diagram of a hand skeleton image after removal of a type identifier provided by an embodiment of the present disclosure;
FIG. 8 illustrates a hand bone image schematic diagram of a bone fracture site outlined according to an embodiment of the present disclosure;
FIG. 9 shows one of the schematic diagrams of a fracture detection device provided by embodiments of the present disclosure;
FIG. 10 illustrates a second schematic view of a fracture detection device provided in an embodiment of the present disclosure;
fig. 11 shows a schematic block diagram of an example electronic device that may be used to implement the fracture detection methods provided by embodiments of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The existing method for detecting the fracture condition through the bone image can only judge whether the bone structure displayed by the bone image is fractured or not, or can only locate the bone structure to the approximate area of the fracture part, and can not further output more detailed information of the fracture part, so that the diagnosis efficiency can not be effectively improved.
The embodiment of the disclosure provides a fracture detection method, a fracture detection device, electronic equipment and a storage medium, which aim to solve at least one of the technical problems in the prior art.
Fig. 1 illustrates a model architecture diagram to which a fracture detection method can be applied, according to an embodiment of the present disclosure, and as illustrated in fig. 1, the architecture includes a fracture detection model and a contour recognition model. When acquiring a bone image to be detected, outputting a preliminary detection result indicating whether the bone image has a fracture part or not by inputting the bone image into a fracture detection model; marking the outline of the fracture part in the bone image by inputting the bone image into the outline recognition model when the primary detection result indicates that the fracture part exists; and determining the name information of the fracture part based on the outline of the fracture part in the bone image, and generating a detection report containing the name information of the fracture part.
Optionally, as shown in fig. 1, the architecture described above may also include a classification model. The embodiment of the disclosure can output name information of the fracture site by inputting the bone image marked with the outline of the fracture site into a classification model.
It can be understood that each model is obtained by training the initial model by using a training set related to fracture images on the basis of the corresponding initial model. The type of the initial model corresponding to each model can be determined according to actual design requirements, for example, a basic model of a fracture detection model can be a fast-RCNN model, a basic model of a contour recognition model can be a Mask-RCNN model, and a basic model of a classification model can be a support vector machine.
Fig. 2 shows a flow chart of a fracture detection method according to an embodiment of the present disclosure, and as shown in fig. 2, the method may mainly include the following steps:
s210: and acquiring a bone image to be detected.
The bone image is an image of a bone structure image containing a body part obtained after scanning the body part by a scanning device. For example, the scanning device may be an X-Ray (X-Ray) scanning device, and after the body part is scanned by the X-Ray scanning device, an X-Ray image including an image of a bone structure of the body part may be obtained, where the X-Ray image is a bone image.
In the embodiment of the disclosure, the bone image to be detected can be directly obtained from the scanning device; alternatively, the bone image scanned by the scanning device may be stored in a designated device (e.g., a server), from which the bone image to be detected is acquired when the detection step needs to be performed. Of course, the bone image to be detected may also be obtained by other means, which are not listed here.
S220: by inputting the bone image to the fracture detection model, a preliminary detection result indicating whether or not the bone image has a fracture site is output.
In embodiments of the present disclosure, the fracture detection model may be a bi-classification model. The bone image is input to a fracture detection model, which can output two preliminary detection results of the presence of a fracture site and the absence of a fracture site.
In the embodiment of the present disclosure, when the preliminary detection result indicates that a fracture site exists, step S230 may be continuously performed; when the preliminary detection result indicates that no fracture site exists, a detection report indicating that no fracture site exists may be continuously generated, and the fracture detection process may be ended.
S230: when the primary detection result indicates that a fracture site exists, the outline of the fracture site is marked in the bone image by inputting the bone image into the outline recognition model.
In the embodiment of the disclosure, the contour recognition model may be any model capable of recognizing the contour of the fracture site based on a preset characteristic fracture image, the bone image is input into the contour recognition model, and the contour recognition model can determine the fracture site from the bone structure contained in the bone image and mark the contour of the fracture site.
S240: name information of the fracture site is determined based on the contour of the fracture site in the bone image.
In the embodiment of the disclosure, a mapping relationship between the contour of each part in the bone structure and the corresponding name information of each part may be pre-established, the contour of the fracture part is marked in the bone image, and the name information of the fracture part may be determined based on the features of the contour.
Alternatively, embodiments of the present disclosure may input a bone image labeled with the contour of the fracture site into a classification model, and determine name information of the fracture site using the classification model. The classification model may be a support vector machine (Support Vector Machine, SVM) or other types of classification models, which are not limited in this disclosure.
S250: a detection report containing name information of the fracture site is generated.
The embodiment of the disclosure can generate a detection report based on a preset report template, wherein the detection report comprises name information of the fracture part. Here, the detection report may be a visual text report, or may be another report, such as a video report, a chart report, or the like, which is not limited in this disclosure.
According to the fracture detection method provided by the embodiment of the disclosure, after the existence of the fracture part in the bone image is determined, the specific position and the name information of the fracture part can be further accurately identified, a detection report containing the name information of the fracture part is generated, the specific information of the fracture part can be intuitively obtained based on the detection report, and the diagnosis efficiency and accuracy of doctors can be greatly improved.
In the embodiment of the disclosure, the bone image includes a bone type identifier. Prior to step S220, a bone type in the bone image may be determined based on the type identification. When determining the name information of the fracture site based on the contour of the fracture site in the bone image, at least one candidate name information may be determined based on the bone type; the name information of the fracture site is determined from the at least one candidate name information based on the contour of the fracture site in the bone image.
In the embodiment of the present disclosure, after determining the bone type in the bone image based on the type identifier, the type identifier in the bone image may be cleared, and a preliminary detection result indicating whether the bone image has a fracture site may be output by inputting the bone image with the cleared type identifier into the fracture detection model.
In the embodiment of the disclosure, the bone type can also be used as one piece of information in the detection report, so that in the case that the bone type in the bone image is determined based on the type identification, the detection report containing the name information of the bone type and the fracture part can be generated, the information in the detection report is enriched, and the diagnosis efficiency and accuracy are improved.
In the embodiments of the present disclosure, the specific model class of the above-described bi-classification model may be determined according to actual design needs, for example, the fracture detection model may be the Faster-RCNN model. FIG. 3 shows an exemplary architecture diagram of a Faster-RCNN model provided in an embodiment of the disclosure, as shown in FIG. 3, the Faster-RCNN model includes a convolutional layer (Conv layers), a region extraction network (Region Proposal Networks, RPN), a region of interest Pooling layer (ROI Pooling), and a Classifier (Classifier). The convolutional layer may consist of a set of underlying conv+relu+pooling for extracting feature maps (feature maps) of bone images, which may be shared for subsequent region extraction networks and classifiers. The region extraction network may determine a plurality of candidate regions (regions) based on the feature map, determine that the anchors belong to positive or negative through a softmax function, and then correct the anchors by bounding box regression to obtain an accurate target region (regions). The region of interest pooling layer may collect the input feature map and the target region, determine a target region feature map based on the feature map and the target region (proposal feature maps), and input the target region feature map to a classifier, which determines whether the bone image has a fracture site based on the target region feature map, i.e., determining whether the target region has a fracture site.
Optionally, the above-mentioned contour recognition model may be a Mask-RCNN model, fig. 4 shows an exemplary structural schematic diagram of the Mask-RCNN model provided by the embodiment of the present disclosure, and as shown in fig. 4, the Mask-RCNN model includes a region of interest aggregation layer (ROI alignment) and a plurality of convolution layers (Conv), a bone image is input into the region of interest aggregation layer of the Mask-RCNN model, and then the bone structure included in the bone image is processed by the plurality of convolution layers, so that a fracture site may be determined and the contour of the fracture site is marked. It should be noted that, in this step, the contour of the frontal fracture part is marked to be substantially consistent with the contour of the fracture part, and the contour can reach the pixel level precision.
Fig. 5 shows a flow chart of another fracture detection method according to an embodiment of the present disclosure, as shown in fig. 5, the method may mainly include the following steps:
s510: and acquiring a bone image to be detected.
The bone image is an image of a bone structure image containing a body part obtained after scanning the body part by a scanning device. Taking the example that the scanned body part is a hand, fig. 6 shows a hand skeleton image provided by an embodiment of the disclosure, after the hand is scanned by the scanning device, a hand skeleton image including the hand skeleton structure shown in fig. 6 may be obtained.
It should be noted that, in step S510, the specific description of acquiring the bone image to be detected may refer to the description in step S210, which is not repeated herein.
S520: the bone type in the bone image is determined based on the type identifier.
In embodiments of the present disclosure, the type identifier in the bone image may be generated by the scanning device. Specifically, while scanning the body part by the scanning device, the user may input information of the scanned body part at the scanning device, which generates a corresponding type identification based on the information of the scanned body part.
S530: the type identifier in the bone image is cleared.
Embodiments of the present disclosure may clear the type identifier in the bone image by a responsive algorithm. Optionally, the embodiment of the disclosure may utilize a gaussian blur mathematical model to remove the type identifier in the bone image, specifically, according to a difference value between the type identifier and its peripheral color, obtain a pixel color covering the icon, and adjust the pixel color of the type identifier to make the color of the type identifier coincide with the peripheral color thereof, thereby achieving the effect of removing the type identifier.
In the disclosed embodiments, the bone image may also be resized, for example, 224x224. The size of the bone image after adjustment can be determined based on the size of the sample image in the training set used in the training process of the fracture detection model, and the purpose of the bone image after adjustment is to adjust the size of the bone image to be detected to be uniform with the size of the sample image in the training process of the fracture detection model, so that the robustness of the fracture detection model is enhanced.
S540: by inputting the bone image with the cleared type identifier to the fracture detection model, a preliminary detection result indicating whether the bone image has a fracture site is output.
After the type identification in the bone image is removed, the method is equivalent to removing part of noise in the bone image, avoiding adverse effect of the type identification on the judging process of the fracture detection model, and improving accuracy of the preliminary detection result output by the fracture detection model. It should be noted that, in step S540, the specific description of acquiring the bone image to be detected may refer to the description in step S220, which is not repeated herein.
S550: when the primary detection result indicates that a fracture site exists, the outline of the fracture site is marked in the bone image by inputting the bone image into the outline recognition model.
In the embodiment of the present disclosure, the specific description of acquiring the bone image to be detected in step S550 may refer to the description in step S230, which is not repeated herein.
S560: at least one candidate name information is determined based on the bone type.
It can be understood that determining at least one candidate name information based on the bone type can reduce the search range of the name information of the fracture site in the subsequent step and improve the search efficiency. The embodiments of the present disclosure may pre-store name information for each of the various parts in each bone type. Taking a left hand bone as an example, each portion of the left hand bone includes a thumb first joint, a thumb second joint, an index finger first joint, a wrist first joint, and the like.
S570: the name information of the fracture site is determined from the at least one candidate name information based on the contour of the fracture site in the bone image.
Optionally, the embodiment of the disclosure may determine the matching probability of the fracture site and each candidate name information in the at least one candidate name information by inputting the bone image labeled with the contour of the fracture site into the classification model; and determining the candidate name information with the highest corresponding matching probability as the name information of the fracture part. The classification model may be a support vector machine (Support Vector Machine, SVM) or other types of classification models, which are not limited in this disclosure.
Here, the matching probability may represent an accuracy rate of determining the name information of the fracture site, and the greater the matching probability, the higher the accuracy rate of determining the name information of the fracture site.
S580: a detection report is generated which includes the bone type, the name information of the fracture site, and the matching probability corresponding to the name information.
In the embodiment of the disclosure, the matching probability corresponding to the bone type and the name information can also be used as one item of information in the detection report, so that the detection report containing the bone type, the name information of the fracture part and the matching probability corresponding to the name information can be generated under the condition that the bone type in the bone image and the matching probability corresponding to the name information are determined based on the type identification, the information in the detection report is enriched, the diagnosis efficiency and the accuracy are improved, and the reliability of the name information can be intuitively embodied when the matching probability corresponding to the name information is displayed in the detection report.
In an embodiment of the present disclosure, the fracture detection model may be a three-classification model. The bone image is input into a fracture detection model, the fracture detection model can output two primary detection results of the existence of the fracture part and the absence of the fracture part, and when the primary detection result indicates the existence of the fracture part, the primary detection result also comprises the fracture type of the fracture part, wherein the fracture type can comprise implantation, contusion and the like. The fracture type can also be used as one piece of information in the detection report, and under the condition that the fracture type is determined, the detection report containing the fracture type and name information of the fracture part can be generated, so that the information in the detection report is enriched, and the diagnosis efficiency and accuracy are improved.
Alternatively, the type identifier in step S520 may be in the form of text, letters, or patterns. As shown in fig. 6, the letter "L" in fig. 6 is a type identifier, which indicates that the bone structure in the bone image is left-hand bone. The embodiment of the disclosure can input the determined bone image with the type identifier into a fracture detection model, and the fracture detection model can identify the type identifier in the bone image so as to determine the bone type in the bone image.
Fig. 7 shows a schematic diagram of a hand skeleton image after removal of a type identifier provided by an embodiment of the present disclosure. In clearing the type identifier in the bone image, the letter "L" in FIG. 6 may be cleared, as can be seen in FIG. 7, which has cleared the letter "L" as compared to FIG. 6.
Fig. 8 shows a schematic diagram of a hand bone image with contours of a fracture site noted, provided by an embodiment of the present disclosure. When the contour of the fracture site is marked in the bone image, the contour of the fracture site may be marked with a designated color, and as shown in fig. 8, the fracture site is the thumb first joint, and the contour recognition model may mark the contour of the thumb first joint with a dark color.
Based on the same principle as the fracture detection method described above, fig. 9 shows one of schematic diagrams of a fracture detection device provided in an embodiment of the present disclosure. As shown in fig. 9, the fracture detection device 900 includes an image acquisition module 910, a preliminary detection module 920, a contour labeling module 930, and a result output module 940.
The image acquisition module 910 is configured to acquire a bone image to be detected.
The preliminary detection module 920 is configured to output a preliminary detection result indicating whether a bone image exists at a fracture site by inputting the bone image into the fracture detection model.
The contour labeling module 930 is configured to label the contour of the fracture site in the bone image by inputting the bone image into the contour recognition model when the primary detection result indicates the presence of the fracture site.
The result output module 940 is configured to determine name information of the fracture site based on the contour of the fracture site in the bone image, and generate a detection report including the name information of the fracture site.
According to the fracture detection device provided by the embodiment of the disclosure, after the existence of the fracture part in the bone image is determined, the specific position and the name information of the fracture part can be further and accurately identified, a detection report containing the name information of the fracture part is generated, the specific information of the fracture part can be intuitively obtained based on the detection report, and the diagnosis efficiency and accuracy of doctors can be greatly improved.
In the embodiment of the disclosure, the bone image includes a bone type identifier. Fig. 10 illustrates a second schematic diagram of a fracture detection device according to an embodiment of the present disclosure, as shown in fig. 10, the fracture detection device 900 further includes a type identification module 950, where the type identification module 950 is configured to: the bone type in the bone image is determined based on the type identification before a preliminary detection result indicating whether the bone image has a fracture site is output by inputting the bone image into the fracture detection model.
The result output module 940 is specifically configured to, when determining the name information of the fracture site based on the contour of the fracture site in the bone image: determining at least one candidate name information based on the bone type; the name information of the fracture site is determined from the at least one candidate name information based on the contour of the fracture site in the bone image.
In the disclosed embodiment, the preliminary detection module 920 is further configured to: removing the type identifier in the bone image; by inputting the bone image with the cleared type identifier to the fracture detection model, a preliminary detection result indicating whether the bone image has a fracture site is output.
In the embodiment of the present disclosure, the result output module 940 is specifically configured to, when used to generate a detection report including name information of a fracture site: a detection report is generated containing name information of the bone type and the fracture site.
In the embodiment of the present disclosure, the result output module 940 is specifically configured to, when determining the name information of the fracture site from the at least one candidate name information based on the contour of the fracture site in the bone image: the matching probability of the fracture part and each candidate name information in at least one candidate name information is determined by inputting a bone image marked with the outline of the fracture part into a classification model; and determining the candidate name information with the highest corresponding matching probability as the name information of the fracture part.
In the embodiment of the present disclosure, the result output module 940 is specifically configured to, when used to generate a detection report including name information of a fracture site: a detection report is generated which includes the bone type, the name information of the fracture site, and the matching probability corresponding to the name information.
In embodiments of the present disclosure, when the preliminary detection result indicates that a fracture site is present, the preliminary detection result further includes a fracture type of the fracture site; the result output module 940 is specifically configured to, when used to generate a detection report including name information of the fracture site: a detection report is generated containing fracture type and name information for the fracture site.
It will be appreciated that the above-described modules of the fracture detection device of the embodiments of the present disclosure have the function of implementing the corresponding steps of the fracture detection method described above. The functions can be realized by hardware, and can also be realized by executing corresponding software by hardware. The hardware or software includes one or more modules corresponding to the functions described above. The modules may be software and/or hardware, and each module may be implemented separately or may be implemented by integrating multiple modules. For a specific description of the functions of the modules of the fracture detection device, reference may be made to the corresponding description of the fracture detection method, which is not repeated herein.
In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related user personal information all conform to the regulations of related laws and regulations, and the public sequence is not violated.
According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.
Fig. 11 shows a schematic block diagram of an example electronic device that may be used to implement the fracture detection methods provided by embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 11, the electronic device 1100 includes a computing unit 1101 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a Random Access Memory (RAM) 1103. In the RAM1103, various programs and data required for the operation of the electronic device 1100 can also be stored. The computing unit 1101, ROM1102, and RAM1103 are connected to each other by a bus 1104. An input/output (I/O) interface 1105 is also connected to bus 1104.
A number of components in the electronic device 1100 are connected to the I/O interface 1105, including: an input unit 1106 such as a keyboard, a mouse, etc.; an output unit 1107 such as various types of displays, speakers, and the like; a storage unit 1108, such as a magnetic disk, optical disk, etc.; and a communication unit 1109 such as a network card, modem, wireless communication transceiver, or the like. The communication unit 1109 allows the electronic device 1100 to exchange information/data with other devices through a computer network such as the internet and/or various telecommunications networks.
The computing unit 1101 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1101 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 performs the respective methods and processes described above, such as a fracture detection method. For example, in some embodiments, the fracture detection method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 1108. In some embodiments, some or all of the computer programs may be loaded and/or installed onto electronic device 1100 via ROM1102 and/or communication unit 1109. When the computer program is loaded into the RAM1103 and executed by the computing unit 1101, one or more steps of the fracture detection method described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform the fracture detection method by any other suitable means (e.g. by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.
The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.
The present disclosure also provides a fracture detection device according to an embodiment of the present disclosure, including an electronic device 1100 shown in fig. 11.
Optionally, the fracture detection device further comprises a scanning device, which is communicatively connected to the electronic device 1100. After the scanning device scans the body part, the obtained image containing the bone structure image of the body part is input into the electronic device 1100 as a bone image to be detected, and the electronic device 1100 outputs a detection report corresponding to the bone image.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.
The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (16)
1. A method of fracture detection comprising:
acquiring a bone image to be detected, wherein the bone image comprises a bone type identifier;
determining a bone type in the bone image based on the type identifier;
outputting a preliminary detection result indicating whether or not the bone image has a fracture site by inputting the bone image to a fracture detection model;
marking the outline of the fracture part in the bone image by inputting the bone image into an outline recognition model when the preliminary detection result indicates that the fracture part exists;
determining at least one candidate name information based on the bone type;
and determining the name information of the fracture part from the at least one candidate name information based on the outline of the fracture part in the bone image, and generating a detection report containing the name information of the fracture part.
2. The method of claim 1, wherein after determining the bone type in the bone image based on the type identification, further comprising:
clearing the type identifier in the bone image;
outputting a preliminary detection result indicating whether or not the bone image has a fracture site by inputting the bone image from which the type identification has been cleared to a fracture detection model.
3. The method of claim 1, wherein the generating a detection report containing name information of the fracture site comprises: a detection report is generated containing the bone type and name information of the fracture site.
4. The method of claim 1, wherein the determining the name information of the fracture site from the at least one candidate name information based on the contour of the fracture site in the bone image comprises:
determining a matching probability of the fracture site and each candidate name information in the at least one candidate name information by inputting the bone image marked with the contour of the fracture site into a classification model;
and determining the candidate name information with the highest corresponding matching probability as the name information of the fracture part.
5. The method of claim 4, wherein the generating a detection report containing name information of the fracture site comprises: and generating a detection report containing the bone type, the name information of the fracture part and the matching probability corresponding to the name information.
6. The method of claim 1, wherein when the preliminary detection result indicates the presence of a fracture site, the preliminary detection result further comprises a fracture type of the fracture site;
the generating a detection report containing name information of the fracture site includes: a detection report is generated containing fracture type and name information for the fracture site.
7. A fracture detection device, comprising:
the image acquisition module is used for acquiring a bone image to be detected, wherein the bone image comprises a bone type identifier;
the type identification module is used for determining the bone type in the bone image based on the type identification;
the preliminary detection module is used for outputting a preliminary detection result indicating whether a fracture part exists in the bone image by inputting the bone image into a fracture detection model;
the contour labeling module is used for labeling the contour of the fracture part in the bone image by inputting the bone image into a contour recognition model when the preliminary detection result indicates that the fracture part exists;
a result output module for determining at least one candidate name information based on the bone type; and determining the name information of the fracture part from the at least one candidate name information based on the outline of the fracture part in the bone image, and generating a detection report containing the name information of the fracture part.
8. The apparatus of claim 7, wherein the preliminary detection module is further to:
clearing the type identifier in the bone image;
outputting a preliminary detection result indicating whether or not the bone image has a fracture site by inputting the bone image from which the type identification has been cleared to a fracture detection model.
9. The apparatus of claim 7, wherein the result output module, when configured to generate a detection report containing name information of the fracture site, is specifically configured to: a detection report is generated containing the bone type and name information of the fracture site.
10. The apparatus of claim 7, wherein the result output module, when configured to determine the name information of the fracture site from the at least one candidate name information based on the contour of the fracture site in the bone image, is specifically configured to:
determining a matching probability of the fracture site and each candidate name information in the at least one candidate name information by inputting the bone image marked with the contour of the fracture site into a classification model;
and determining the candidate name information with the highest corresponding matching probability as the name information of the fracture part.
11. The apparatus of claim 10, wherein the result output module, when configured to generate a detection report containing name information of the fracture site, is specifically configured to: and generating a detection report containing the bone type, the name information of the fracture part and the matching probability corresponding to the name information.
12. The apparatus of claim 7, wherein when the preliminary detection result indicates the presence of a fracture site, the preliminary detection result further comprises a fracture type of the fracture site;
the result output module is specifically configured to, when being configured to generate a detection report including name information of the fracture site: a detection report is generated containing fracture type and name information for the fracture site.
13. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
14. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-6.
15. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-6.
16. A fracture detection device comprising the electronic device of claim 13.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110907714.0A CN113610809B (en) | 2021-08-09 | 2021-08-09 | Fracture detection method, fracture detection device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110907714.0A CN113610809B (en) | 2021-08-09 | 2021-08-09 | Fracture detection method, fracture detection device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113610809A CN113610809A (en) | 2021-11-05 |
| CN113610809B true CN113610809B (en) | 2024-04-05 |
Family
ID=78307642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110907714.0A Active CN113610809B (en) | 2021-08-09 | 2021-08-09 | Fracture detection method, fracture detection device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113610809B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114974509B (en) * | 2022-05-26 | 2024-05-10 | 哈尔滨工业大学 | Fracture reduction path planning method, fracture reduction method and electronic equipment |
| CN115661138B (en) * | 2022-12-13 | 2023-03-21 | 北京大学第三医院（北京大学第三临床医学院） | Human skeleton contour detection method based on DR image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111325745A (en) * | 2020-03-09 | 2020-06-23 | 北京深睿博联科技有限责任公司 | Fracture region analysis method and device, electronic device and readable storage medium |
| CN111667474A (en) * | 2020-06-08 | 2020-09-15 | 杨天潼 | Fracture identification method, apparatus, device and computer readable storage medium |
| CN112819811A (en) * | 2021-02-24 | 2021-05-18 | 上海商汤智能科技有限公司 | Image analysis method and related device, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6923368B2 (en) * | 2017-06-15 | 2021-08-18 | キヤノンメディカルシステムズ株式会社 | Medical image processing equipment, medical diagnostic imaging equipment and medical image processing programs |
-
2021
- 2021-08-09 CN CN202110907714.0A patent/CN113610809B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111325745A (en) * | 2020-03-09 | 2020-06-23 | 北京深睿博联科技有限责任公司 | Fracture region analysis method and device, electronic device and readable storage medium |
| CN111667474A (en) * | 2020-06-08 | 2020-09-15 | 杨天潼 | Fracture identification method, apparatus, device and computer readable storage medium |
| CN112819811A (en) * | 2021-02-24 | 2021-05-18 | 上海商汤智能科技有限公司 | Image analysis method and related device, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 多层CT及图像后处理技术在诊断隐匿性肋骨骨折中的应用;吴江;;包头医学院学报;20150315(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113610809A (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11854237B2 (en) | Human body identification method, electronic device and storage medium | |
| US20220270382A1 (en) | Method and apparatus of training image recognition model, method and apparatus of recognizing image, and electronic device | |
| CN112597837B (en) | Image detection method, apparatus, device, storage medium, and computer program product | |
| CN113610809B (en) | Fracture detection method, fracture detection device, electronic equipment and storage medium | |
| CN113780098B (en) | Character recognition method, character recognition device, electronic equipment and storage medium | |
| CN113205041B (en) | Structured information extraction method, device, equipment and storage medium | |
| CN113378712B (en) | Training method of object detection model, image detection method and device thereof | |
| EP3846122B1 (en) | Method and apparatus for generating background-free image, device, and medium | |
| CN112561879B (en) | Ambiguity evaluation model training method, image ambiguity evaluation method and image ambiguity evaluation device | |
| US20220027661A1 (en) | Method and apparatus of processing image, electronic device, and storage medium | |
| CN113378836A (en) | Image recognition method, apparatus, device, medium, and program product | |
| CN112115921A (en) | True and false identification method and device and electronic equipment | |
| CN112580620A (en) | Sign picture processing method, device, equipment and medium | |
| CN114677566B (en) | Training method of deep learning model, object recognition method and device | |
| CN113936158A (en) | Label matching method and device | |
| CN113139486A (en) | Method, apparatus, device and storage medium for processing image | |
| CN113361371A (en) | Road extraction method, device, equipment and storage medium | |
| CN114998906B (en) | Text detection method, training method and device of model, electronic equipment and medium | |
| CN112101368B (en) | Character image processing method, device, equipment and medium | |
| CN114792423B (en) | Document image processing method and device and storage medium | |
| CN116611997A (en) | Sample image generation method, device and equipment | |
| CN117975099A (en) | Pixel-level target labeling method and device, electronic equipment and storage medium | |
| CN116682124A (en) | Data entry method, device, equipment and storage medium | |
| CN117094870A (en) | Data processing method, apparatus, device, computer readable storage medium and product | |
| CN116403073A (en) | Training method, device, equipment and storage medium of target detection task model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |