CN113688942A - Method and device for automatically evaluating cephalic and lateral adenoid body images based on deep learning - Google Patents
Method and device for automatically evaluating cephalic and lateral adenoid body images based on deep learning Download PDFInfo
- Publication number
- CN113688942A CN113688942A CN202111088436.7A CN202111088436A CN113688942A CN 113688942 A CN113688942 A CN 113688942A CN 202111088436 A CN202111088436 A CN 202111088436A CN 113688942 A CN113688942 A CN 113688942A
- Authority
- CN
- China
- Prior art keywords
- image
- adenoid
- deep learning
- adenoid body
- body region
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a device for automatically evaluating a cephalic and cephalic adenoid body image based on deep learning, wherein an original cephalic and cephalic plate is input into a preprocessing module to extract a adenoid body region image and perform gray level normalization processing by constructing a deep learning model; defining the adenoid body region image as two categories of a adenoid body pathologic hypertrophy image and a adenoid body normal image, inputting the image of the adenoid body region into a deep learning model, and carrying out forward reasoning; and obtaining the prediction result of the image belonging category output by the deep learning model. According to the method provided by the invention, the deep learning model is constructed, the adenoid body region sub-image is input to the deep learning model for recognition, and the adenoid body image recognition result corresponding to the original head side sheet is obtained.
Description
Technical Field
The invention relates to the technical field of computer image processing, in particular to a method and a device for automatically evaluating a head lateral adenoid body image based on deep learning.
Background
The oral cephalic X-ray film is a conventional X-ray film used for diagnosis and analysis by doctors and is widely applied to the field of orthodontic science. The adenoid is a kind of soft tissue located in the throat, and its hypertrophy may cause the decrease of the nasal cavity ventilation function, so that the patient is forced to breath, and the pathological manifestations such as tooth irregularity, bucktooth, mandible retraction, etc. are caused.
Adenoids are divided into two different types, physiological hypertrophy and pathological hypertrophy. Wherein physiological hypertrophy can improve itself with the increase of age, and special treatment is not generally needed; pathological hypertrophy is caused by the fact that adenoids and palatine tonsils are located at the first defense part of respiratory tract, and are subjected to various inflammatory stimuli for a long time, so that attention needs to be paid to the adenoids and palatine tonsils, and the operation is generally considered clinically for patients with poor curative effect through medicines at present. Because the enlarged adenoid blocks an airway to a certain extent, the respiration is influenced, and the whole body health of the child is greatly influenced, the diagnosis can be timely and accurately made, and further, a doctor is assisted to make a reasonable treatment plan.
At present, the clinical practice is manually completed by oral specialists or otorhinolaryngological doctors, different doctors have different sensitivities to gray level differences in head and side films, and the judgment result of a measurement boundary lacks objectivity, so that subsequent calculation deviation is caused, and the accuracy of a diagnosis result is influenced. At present, medical resources in China are scarce, the number of medical experts is insufficient, and if the workers completely depend on manual operation, the workers can feel eyestrain, so that the accuracy and credibility of the judgment result are further reduced. Under such circumstances, there is an urgent need to develop a method and apparatus for automatically processing a cephalic plate to assist a physician in discriminating an adenoid image.
Disclosure of Invention
In view of the above, the present invention provides a method and an apparatus for automatically recognizing an adenoid region image of a cranial flap based on deep learning.
In order to achieve the purpose, the invention provides the following technical scheme:
the invention provides a method for automatically evaluating a cephalic and lateral adenoid body image based on deep learning, which comprises the following steps of:
acquiring a head side image;
inputting the head lateral image into a preprocessing module to extract a glandular body region image and perform gray level normalization processing;
the adenoid body region map is defined as two categories of a adenoid body pathologic hypertrophy image and a adenoid body normal image, and the image of the adenoid body region is input into a deep learning model for forward reasoning;
and obtaining the prediction result of the image belonging category output by the deep learning model.
Further, the head side plate is an X-ray image formed by X-ray examination.
Further, the pretreatment comprises the following steps:
a map of the adenoid region is taken from the cephalad panel,
and carrying out gray level normalization on the intercepted adenoid area image to obtain an adenoid area matrix.
Further, the deep learning model comprises an adenoid body region automatic intercepting unit and a neural network classifier;
the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate;
the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
Further, the automatic intercepting unit of the adenoid body area is used for detecting the resolution of the input head side film, carrying out down-sampling on the head side film according to the preset resolution to obtain a down-sampling image, and intercepting a preset rectangular area in the down-sampling image to obtain the adenoid body area image.
Further, the output image of the deep learning model is processed according to the following steps:
intercepting the adenoid body region of the cephalic and lateral plates, carrying out gray normalization on pixel points of the intercepted region, calculating a loss value by using a cross entropy loss function, calling an Adam gradient descent algorithm to update a model weight, and determining the optimal parameters of the model according to the F1-score value on the verification set.
The invention provides a device for automatically evaluating hypertrophy of a cephalic and collateral adenoids based on deep learning, which comprises a data acquisition module, a preprocessing module and a deep learning model;
the data acquisition module is used for acquiring a head side film image;
the preprocessing module is used for inputting the head lateral image into the preprocessing module to extract the adenoid body region image and carry out gray level normalization processing;
the deep learning model is used for inputting a adenoid body region map into the deep learning model, wherein the adenoid body region is defined as two classifications of a adenoid body pathologic hypertrophy image and a adenoid body normal image; inputting the adenoid body region into a deep learning model for forward reasoning;
and obtaining the prediction result of the image output by the model belonging to the category.
Further, the deep learning model comprises an adenoid body region automatic intercepting unit and a neural network classifier;
the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate;
the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
Further, the automatic intercepting unit of the adenoid body area is used for detecting the resolution of the input head side film, carrying out down-sampling on the head side film according to the preset resolution to obtain a down-sampling image, and intercepting a preset rectangular area in the down-sampling image to obtain the adenoid body area image.
Further, the output image of the deep learning model is processed according to the following steps:
extracting training data from the adenoid body region of the cephalic and lateral plates, carrying out gray level normalization on pixel points, carrying out loss value calculation by using a cross entropy loss function, calling an Adam gradient descent algorithm to update model weight, and determining the optimal parameters of the model according to the F1-score value on the verification set.
The invention has the beneficial effects that:
according to the method and the device for automatically judging the hypertrophy of the side-of-head adenoids based on deep learning, disclosed by the invention, the original side-of-head adenoids are input into a preprocessing module by constructing a deep learning model, then a rectangular frame is automatically generated in an adenoid area, and sub-images are extracted; and inputting the subgraph into a deep learning model for recognition to obtain a glandular body region image recognition result corresponding to the original head lateral sheet. The method and the system can automatically complete the previous manual operation, and the identification method can automatically complete the previous manual operation and has the advantages of objectivity, rapidness, good repeatability and the like.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
In order to make the object, technical scheme and beneficial effect of the invention more clear, the invention provides the following drawings for explanation:
fig. 1 is a flowchart of a method for identifying a lateral cranial adenoid based on deep learning.
FIG. 2 is a schematic block diagram of a deep learning-based lateral cranial adenoid recognition system.
Fig. 3 is a schematic diagram of a cephalic adenoid image.
FIG. 4 is a schematic diagram of the extraction of the lateral cranial adenoids.
Detailed Description
The present invention is further described with reference to the following drawings and specific examples so that those skilled in the art can better understand the present invention and can practice the present invention, but the examples are not intended to limit the present invention.
Example 1
As shown in fig. 1, the method for automatically evaluating a lateral cranial adenoid body image based on deep learning provided by this embodiment includes the following steps:
acquiring a head side image;
inputting the head lateral image into a preprocessing module to extract a glandular body region image and perform gray level normalization processing;
extracting an adenoid body region image from the preprocessed head side image, wherein the adenoid body region image is a sub-image extracted according to an adenoid body region rectangular frame;
the adenoid body region map is defined as two categories of a adenoid body pathologic hypertrophy image and a adenoid body normal image, and the image of the adenoid body region is input into a deep learning model for forward reasoning;
and obtaining the prediction result of the image belonging category output by the deep learning model.
The deep learning model provided by the embodiment comprises an adenoid body region automatic intercepting unit and a neural network classifier;
the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate;
the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
The adenoid reference image information provided by this embodiment is used for decision-making judgment by related medical personnel in combination with other information, and finally determines whether the adenoid is hypertrophic, obtains a judgment result of the hypertrophy of the adenoid corresponding to the original lateral head slice, and provides reference auxiliary information for medical decision-making.
The original head side plate provided by the embodiment is an original X-ray image formed by X-ray examination.
The pretreatment provided by the embodiment comprises the following steps:
a map of the adenoid region is first taken from the cephalad panel,
and then carrying out gray level normalization on the intercepted adenoid area image to obtain an adenoid area matrix.
The preprocessing is to cut the center of the image to obtain 500 × 500 sub-images, and then to normalize the gray scale of the sub-images. The gray scale range is 0-255, that is, the gray scale value of all the pixels is divided by 255 to obtain a floating point number.
The subgraph provided by the embodiment is automatically generated by generating a rectangular frame according to the adenoid region and extracting; in the middle of the obtained image of the X-ray film, the height is divided by 2 width and rounded up to obtain y divided by 2 and rounded up to obtain X, and the (X, y) is the coordinate of the central point of the image. Then, x and y of the coordinates are respectively +/-250 to obtain a 500 × 500 region, namely a subgraph. And then inputting the subgraph into a deep learning model for image classification to obtain an adenoid evaluation result corresponding to the original head lateral slide.
Wherein the deep learning-based adenoid hypertrophy automatic identification model is obtained by the following steps:
(1) a model design stage: the network structure comprises two stages, namely, adenoid body region automatic interception and a neural network feature extractor, wherein the adenoid body automatic interception stage needs to detect the resolution of an input cephalic side slice, down-samples the cephalic side slice to the resolution of 1537 × 1752 and then intercepts a rectangular region of 500 × 500 in the center of an image; the neural network feature extractor is a simplified model based on VGG16, and a rectangular region is input to VGG16 for secondary classification to obtain final probability prediction of adenoid hypertrophy;
(2) a model training stage: intercepting the adenoid body region of the cephalic and lateral plates, carrying out gray normalization on pixel points of the intercepted region, calculating a loss value by using a cross entropy loss function, calling an Adam gradient descent algorithm to update a model weight, and determining the optimal parameters of the model according to the F1-score value on the verification set.
The gray normalization is a conventional operation in image processing, and a pixel point is an integral number of 8-bit. The value of each pixel is divided by 255 to obtain a 32-bit single-precision floating point number of 0-1. The model training is an optimization process, and parameters stored after the training are finished are technically called as current optimal parameters.
The method for randomly initializing the model weight comprises the following steps: he normal distribution initialization method He _ normal, uniform distribution initialization method mount _ uniform, and random value random _ normal output from the normal distribution.
The He normal distribution initialization method He _ normal, (which extracts numbers from a truncated normal distribution centered at 0 with standard deviation stddev = sqrt (2/fan _ in), where fan _ in is the number of input units in the weight tensor);
the uniform distribution initialization method lecun _ uniform, (which extracts a number from the uniform distribution in [ -limit, limit ], where limit is sqrt (3/fan _ in), and fan _ in is the number of input units in the weight tensor);
the random initialization method random _ normal (which generates random numbers centered at 0 and having a standard deviation of 0.1).
As shown in fig. 2, the apparatus for automatically evaluating a lateral cranial adenoid body image based on deep learning of the present embodiment includes a data acquisition module, a preprocessing module and a deep learning model;
the data acquisition module is used for acquiring a head side film image;
the preprocessing module is used for inputting the head lateral image into the preprocessing module to extract the adenoid body region image and carry out gray level normalization processing;
the deep learning model defines the adenoid body area as two classifications of a adenoid body pathologic hypertrophy image and a adenoid body normal image; inputting the adenoid body region into a deep learning model for forward reasoning; and obtaining the prediction result of the image output by the model belonging to the category.
The deep learning model comprises an adenoid body region automatic intercepting unit and a neural network classifier; the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate; the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
As shown in fig. 3, the head side plate provided by the present embodiment is photographed at a preset angle and orientation, and finally the adenoid region of the photographed head side plate is exactly in the center. Therefore, the determination of the area of the adenoid is realized by acquiring the coordinates of the central point of the image, the x axis of the image is divided by 2, and the y axis of the image is divided by 2 to obtain the coordinates of the central point of the image; then acquiring a 500 multiplied by 500 area of the adenoid body area, namely (x +/-250, y +/-250) according to the coordinate deviation of the central point of the image, wherein the coordinate deviation is set to +/-250; where xy is the length and width of the image; and finally outputting 1, namely, representing the positive evaluation result of the adenoid body recognized by the picture through the deep learning model, and under the guidance of the evaluation result, requiring a doctor to further diagnose the evaluation result.
As shown in fig. 4, the lateral head slice provided in this embodiment makes the adenoid region exactly in the center of the picture, which improves the speed of determining the adenoid region, so that the deep learning model can process according to the set region, and the difficulty of determining the preset region by the deep learning model is simplified.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (10)
1. The method for automatically evaluating the cephalic and lateral adenoid body images based on deep learning comprises the following steps:
acquiring a head side image;
inputting the head lateral image into a preprocessing module to extract a glandular body region image and perform gray level normalization processing;
the adenoid body region map is defined as two categories of a adenoid body pathologic hypertrophy image and a adenoid body normal image, and the image of the adenoid body region is input into a deep learning model for forward reasoning;
and obtaining the prediction result of the image belonging category output by the deep learning model.
2. The method for automatically evaluating the lateral cranial adenoid body image based on deep learning of claim 1, wherein: the head side plate is an X-ray image formed by X-ray examination.
3. The method for automatically evaluating the lateral cranial adenoid body image based on deep learning of claim 1, wherein: the pretreatment comprises the following steps:
a map of the adenoid region is taken from the cephalad panel,
and carrying out gray level normalization on the intercepted adenoid area image to obtain an adenoid area matrix.
4. The method for automatically evaluating the rostral adenoids based on deep learning of claim 1, wherein: the deep learning model comprises an adenoid body region automatic intercepting unit and a neural network classifier;
the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate;
the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
5. The method for automatically evaluating the lateral cranial adenoid body image based on deep learning of claim 1, wherein: the automatic adenoid body area intercepting unit is used for detecting the resolution of an input head side film, carrying out down-sampling on the head side film according to a preset resolution to obtain a down-sampling image, and intercepting a preset rectangular area in the down-sampling image to obtain an adenoid body area image.
6. The method for automatically evaluating the lateral cranial adenoid body image based on deep learning of claim 1, wherein: the output image of the deep learning model is processed according to the following steps:
intercepting the adenoid body region of the cephalic and lateral plates, carrying out gray normalization on pixel points of the intercepted region, calculating a loss value by using a cross entropy loss function, calling an Adam gradient descent algorithm to update a model weight, and determining the optimal parameters of the model according to the F1-score value on the verification set.
7. Automatic evaluation device of head lateral side adenoid hypertrophy based on degree of deep learning, its characterized in that: the system comprises a data acquisition module, a preprocessing module and a deep learning model;
the data acquisition module is used for acquiring a head side film image;
the preprocessing module is used for inputting the head lateral image into the preprocessing module to extract the adenoid body region image and carry out gray level normalization processing;
the deep learning model is used for inputting a adenoid body region map into the deep learning model, wherein the adenoid body region is defined as two classifications of a adenoid body pathologic hypertrophy image and a adenoid body normal image; inputting the adenoid body region into a deep learning model for forward reasoning;
and obtaining the prediction result of the image output by the model belonging to the category.
8. The apparatus for automatically evaluating a rostral adenoid based on deep learning of claim 7, wherein: the deep learning model comprises an adenoid body region automatic intercepting unit and a neural network classifier;
the automatic adenoid body region intercepting unit is used for acquiring an input adenoid body region image in the head side plate;
the neural network feature extractor is a VGG16 model, and the rectangular area is input to a VGG16 model for image classification.
9. The apparatus for automatically evaluating a cranial adenoid body image based on deep learning of claim 8, wherein: the automatic adenoid body area intercepting unit is used for detecting the resolution of an input head side film, carrying out down-sampling on the head side film according to a preset resolution to obtain a down-sampling image, and intercepting a preset rectangular area in the down-sampling image to obtain an adenoid body area image.
10. The apparatus for automatically evaluating a cranial adenoid body image based on deep learning of claim 8, wherein: the output image of the deep learning model is processed according to the following steps:
extracting training data from the adenoid body region of the cephalic and lateral plates, carrying out gray level normalization on pixel points, carrying out loss value calculation by using a cross entropy loss function, calling an Adam gradient descent algorithm to update model weight, and determining the optimal parameters of the model according to the F1-score value on the verification set.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110177733 | 2021-02-09 | ||
| CN2021101777332 | 2021-02-09 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113688942A true CN113688942A (en) | 2021-11-23 |
Family
ID=78586589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111088436.7A Pending CN113688942A (en) | 2021-02-09 | 2021-09-16 | Method and device for automatically evaluating cephalic and lateral adenoid body images based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113688942A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116994070A (en) * | 2023-09-25 | 2023-11-03 | 四川大学 | Tooth image processing method and device based on measurable subspace dynamic classifier |
| CN117726624A (en) * | 2024-02-07 | 2024-03-19 | 北京长木谷医疗科技股份有限公司 | Method and device for intelligently identifying and evaluating adenoid lesions in real time under video stream |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111127390A (en) * | 2019-10-21 | 2020-05-08 | 哈尔滨医科大学 | X-ray image processing method and system based on transfer learning |
| WO2020224123A1 (en) * | 2019-06-24 | 2020-11-12 | 浙江大学 | Deep learning-based seizure focus three-dimensional automatic positioning system |
| CN112184658A (en) * | 2020-09-24 | 2021-01-05 | 上海健康医学院 | Method, medium, and electronic device for predicting non-small cell lung cancer prognostic survival |
-
2021
- 2021-09-16 CN CN202111088436.7A patent/CN113688942A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020224123A1 (en) * | 2019-06-24 | 2020-11-12 | 浙江大学 | Deep learning-based seizure focus three-dimensional automatic positioning system |
| CN111127390A (en) * | 2019-10-21 | 2020-05-08 | 哈尔滨医科大学 | X-ray image processing method and system based on transfer learning |
| CN112184658A (en) * | 2020-09-24 | 2021-01-05 | 上海健康医学院 | Method, medium, and electronic device for predicting non-small cell lung cancer prognostic survival |
Non-Patent Citations (3)
| Title |
|---|
| LIU J L等: "Automated radiographic evaluation of adenoid hypertrophy based on VGG-Lite" * |
| SHEN YI等: "A deep‐learning‐based approach for adenoid hypertrophy diagnosis" * |
| X. HU等: "Photographic Analysis and Machine Learning for Diagnostic Prediction of Adenoid Hypertrophy" * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116994070A (en) * | 2023-09-25 | 2023-11-03 | 四川大学 | Tooth image processing method and device based on measurable subspace dynamic classifier |
| CN116994070B (en) * | 2023-09-25 | 2023-12-01 | 四川大学 | Tooth image processing method and device based on measurable subspace dynamic classifier |
| CN117726624A (en) * | 2024-02-07 | 2024-03-19 | 北京长木谷医疗科技股份有限公司 | Method and device for intelligently identifying and evaluating adenoid lesions in real time under video stream |
| CN117726624B (en) * | 2024-02-07 | 2024-05-28 | 北京长木谷医疗科技股份有限公司 | Method and device for intelligently identifying and evaluating adenoid lesions in real time under video stream |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106909778B (en) | A kind of Multimodal medical image recognition methods and device based on deep learning | |
| EP3806744B1 (en) | Immediate workup | |
| Luo et al. | Micro-vessel image segmentation based on the AD-UNet model | |
| CN112365464B (en) | GAN-based medical image lesion area weak supervision positioning method | |
| CN113688942A (en) | Method and device for automatically evaluating cephalic and lateral adenoid body images based on deep learning | |
| CN111862075A (en) | Lung image analysis system and method based on deep learning | |
| CN103249358A (en) | Medical image processing device | |
| CN113782184A (en) | Cerebral apoplexy auxiliary evaluation system based on facial key point and feature pre-learning | |
| CN113095382A (en) | Interpretable tuberculosis classification network identification method based on CT image | |
| CN113397485A (en) | Scoliosis screening method based on deep learning | |
| CN113469987A (en) | Dental X-ray image lesion area positioning system based on deep learning | |
| Chen et al. | Detection of various dental conditions on dental panoramic radiography using Faster R-CNN | |
| CN113657458A (en) | Airway classification method and device and computer-readable storage medium | |
| CN111986217A (en) | Image processing method, device and equipment | |
| CN113688788B (en) | Automatic recognition method and system for airway image on head side piece based on deep learning | |
| CN115690556A (en) | Image recognition method and system based on multi-modal iconography characteristics | |
| CN113255718B (en) | Cervical cell auxiliary diagnosis method based on deep learning cascade network method | |
| CN115035086A (en) | Intelligent tuberculosis skin test screening and analyzing method and device based on deep learning | |
| CN113592834A (en) | X-ray chest radiography image quality determination method and device | |
| CN116113984A (en) | Skin surface analysis device and skin surface analysis method | |
| CN113450306A (en) | Method of providing a fracture detection tool | |
| CN111462101A (en) | Staging equipment based on novel coronavirus pneumonia CT detection and using method thereof | |
| Velusamy et al. | Faster Region‐based Convolutional Neural Networks with You Only Look Once multi‐stage caries lesion from oral panoramic X‐ray images | |
| CN116612117B (en) | Female genital tract pathogen recognition system based on morphology | |
| CN113822904B (en) | Image labeling device, method and readable storage medium |
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 |