CN110555825A - Intelligent diagnostic system and diagnostic method for chest X-ray image - Google Patents
Intelligent diagnostic system and diagnostic method for chest X-ray image Download PDFInfo
- Publication number
- CN110555825A CN110555825A CN201910667141.1A CN201910667141A CN110555825A CN 110555825 A CN110555825 A CN 110555825A CN 201910667141 A CN201910667141 A CN 201910667141A CN 110555825 A CN110555825 A CN 110555825A
- Authority
- CN
- China
- Prior art keywords
- chest
- module
- image
- abnormal
- qualified
- 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
- 238000011976 chest X-ray Methods 0.000 title claims abstract description 47
- 238000002405 diagnostic procedure Methods 0.000 title claims abstract description 11
- 238000003745 diagnosis Methods 0.000 claims abstract description 135
- 230000002159 abnormal effect Effects 0.000 claims abstract description 104
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000002601 radiography Methods 0.000 claims abstract description 27
- 238000010586 diagram Methods 0.000 claims description 73
- 238000004364 calculation method Methods 0.000 claims description 32
- 238000000605 extraction Methods 0.000 claims description 17
- 230000003902 lesion Effects 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 6
- 238000003384 imaging method Methods 0.000 claims description 6
- 230000002746 orthostatic effect Effects 0.000 claims description 6
- 230000003213 activating effect Effects 0.000 claims description 4
- 210000000038 chest Anatomy 0.000 description 170
- 230000005856 abnormality Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 210000004072 lung Anatomy 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000013526 transfer learning Methods 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000000415 inactivating effect Effects 0.000 description 1
- 210000001370 mediastinum Anatomy 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 210000002345 respiratory system Anatomy 0.000 description 1
- 208000023504 respiratory system disease Diseases 0.000 description 1
- 210000004872 soft tissue Anatomy 0.000 description 1
- 210000000779 thoracic wall Anatomy 0.000 description 1
- 201000008827 tuberculosis Diseases 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
-
- 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
- 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
- G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
-
- 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
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30061—Lung
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Biophysics (AREA)
- Surgery (AREA)
- Pathology (AREA)
- High Energy & Nuclear Physics (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Optics & Photonics (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
the invention provides a chest X-ray image intelligent diagnosis system, which comprises an image information management module, an image judgment module and a chest X-ray image diagnosis module, wherein the image information management module is used for transmitting a chest examination image of a patient to the image judgment module; the image judgment module is used for identifying the right chest film image in the image, analyzing the quality of the right chest film image, sending the right chest film image meeting the preset conditions to the image processing module and defining the right chest film image as a qualified chest film; the image processing module performs unified processing on the qualified chest radiographs; the two-classification intelligent diagnosis module is used for carrying out normal or abnormal diagnosis on the processed qualified chest radiograph; the structured report module activates the non-discovery related control in the structured report interface when normal data is received; and when the abnormal data is received, deactivating the non-finding related control. The invention also discloses an intelligent diagnostic method for the chest X-ray image. The invention intelligently diagnoses the chest radiography into normal and abnormal by a two-classification method, and sends the result to the structured report, thereby improving the diagnosis efficiency of doctors.
Description
Technical Field
The invention relates to the field of medical information, in particular to a chest X-ray image intelligent diagnosis system and a chest X-ray image intelligent diagnosis method.
Background
The chest X-ray image clinical examination is called chest film for short, and is widely applied clinically. The positive chest film can display the shapes, positions and contours of the lung, the mediastinum, the diaphragm, the chest wall bone, soft tissues and the like, can observe lung pathological changes and contours of heart great vessels, and can be used for disease diagnosis mainly based on a respiratory system. CNN can be used for the study of Chest X-Ray (Chest X Ray, CXR) intelligent diagnosis, but there is no mature product. The most important work in the existing research is chest X-ray multi-classification diagnosis, and some companies are used for automatic detection of tuberculosis and other diseases.
The prior art does not pay attention to the optimized work flow, and chest X-ray images are a first-line image examination method for diagnosing respiratory system diseases and account for a large proportion of daily work tasks in medical imaging departments. The condition of no discovery (normal) is most common, the condition occupies more than half of the total workload of chest X-ray images, the existing process is that a doctor diagnoses each image one by one, a large amount of time is wasted for diagnosis on the images without discovery (normal), and the diagnosis result does not automatically access a structured report system, so that the work efficiency of the doctor is reduced.
disclosure of Invention
in view of the above, the main objective of the present invention is to provide an intelligent diagnostic system for chest X-ray images and a diagnostic method thereof, which can solve the problem in the prior art that the efficiency of doctor's diagnosis and report writing is not high because a complete workflow is not formed and the doctor needs to diagnose each image one by one.
in order to achieve the purpose, the technical scheme of the invention is realized as follows:
On one hand, the invention provides a chest X-ray image intelligent diagnosis system, which comprises an image information management module, an image judgment module, an image processing module, a two-classification intelligent diagnosis module and a structural report module, wherein the image information management module is connected with the image judgment module and is used for transmitting medical digital imaging and communication DICOM images of a patient to the image judgment module through a DICOM protocol when the patient finishes the examination that an examination item is a chest film; the image judgment module is respectively connected with the image information management module and the image processing module and is used for identifying the chest radiograph in the DICOM image, analyzing the quality of the chest radiograph based on a preset condition and sending the chest radiograph meeting the preset condition to the image processing module; wherein, the righting chest picture image which meets the preset condition is defined as a qualified chest picture; the image processing module is respectively connected with the image judging module and the two-classification intelligent diagnosis module and is used for processing the position, the size, the brightness and the contrast of the qualified chest radiography to generate a picture format with uniform specification and sending the processed qualified chest radiography to the two-classification intelligent diagnosis module; the two-classification intelligent diagnosis module is respectively connected with the image processing module and the structured report module and is used for carrying out normal or abnormal diagnosis on the processed qualified chest radiograph and sending normal data or abnormal data to the structured report module; the structured report module is connected with the two-classification intelligent diagnosis module and is used for activating the non-finding related control in the structured report interface and automatically generating a diagnosis report of a qualified chest radiograph when normal data are received; when the abnormal data is received, the non-finding related control in the structured report interface is inactivated.
Preferably, the image judgment module further comprises a feedback unit connected to the structured report module, and configured to send the unique ID of the patient corresponding to the non-orthostatic chest image and the orthostatic chest image that does not meet the preset condition in the DICOM image to the structured report module for further diagnosis by the doctor.
Preferably, the two-classification intelligent diagnosis module comprises a feature extraction unit and a calculation unit, wherein the feature extraction unit is connected with the calculation unit and is used for extracting feature data of the processed qualified chest radiograph and sending the feature data to the calculation unit; the calculation unit is connected with the feature extraction unit and is used for performing combined calculation of the and or the non-relation of features on the feature data and the full-connection layer trained by the plurality of normal chest pictures and the abnormal chest pictures, calculating the normal probability and the abnormal probability of the qualified chest pictures after the processing, and outputting the normal data when the normal probability is greater than a first preset threshold value; and outputting abnormal data when the abnormal probability is greater than a second preset threshold value.
preferably, when the two-classification intelligent diagnosis module diagnoses the processed qualified chest radiograph as abnormal, the two-classification intelligent diagnosis module further comprises: and the thermodynamic diagram generation unit is connected with the calculation unit and is used for marking suspicious lesion areas on the abnormal qualified chest films based on the models and the weights trained by the plurality of normal chest films and abnormal chest films, and generating a thermodynamic diagram to be sent to the structured report module.
on the other hand, the invention also provides an intelligent diagnostic method for chest X-ray images, which comprises the following steps: when the patient finishes the examination of the chest radiography, the image information management module transmits the medical digital imaging and communication DICOM image of the patient to the image judgment module through the DICOM protocol; the image judgment module identifies the chest radiograph in the DICOM image, analyzes the quality of the chest radiograph based on a preset condition and sends the chest radiograph meeting the preset condition to the image processing module; wherein, the righting chest picture image which meets the preset condition is defined as a qualified chest picture; the image processing module processes the position, size, brightness and contrast of the qualified chest radiograph to generate a uniform picture format, and sends the processed qualified chest radiograph to the two-classification intelligent diagnosis module; the two-classification intelligent diagnosis module carries out normal or abnormal diagnosis on the processed qualified chest radiograph and sends normal data or abnormal data to the structured report module; when normal data are received, the structured report module activates related control without finding in the structured report interface, and automatically generates a diagnosis report of a qualified chest radiograph; when the abnormal data is received, the non-finding related control in the structured report interface is inactivated.
Preferably, the method further comprises: and a feedback unit in the image judgment module sends the unique ID of the patient corresponding to the non-righting chest image in the DICOM image and the righting chest image which does not meet the preset condition to the structured report module for further diagnosis by a doctor.
Preferably, the diagnosing of normality or abnormality of the processed qualified chest radiograph comprises: a feature extraction unit in the two-classification intelligent diagnosis module extracts the processed feature data of the qualified chest radiograph and sends the feature data to a calculation unit; the calculation unit performs combined calculation of the and or not relation of the features on the feature data and the full-connection layer trained by the plurality of normal chest pictures and abnormal chest pictures, calculates the normal probability and the abnormal probability of the qualified chest pictures after the processing, and outputs normal data when the normal probability is greater than a first preset threshold value; and outputting abnormal data when the abnormal probability is greater than a second preset threshold value.
Preferably, when the processed qualified chest radiograph is diagnosed as abnormal by the binary intelligent diagnosis module, the method further comprises: a thermodynamic diagram generation unit in the two-classification intelligent diagnosis module marks suspicious focus areas on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, generates a thermodynamic diagram and sends the thermodynamic diagram to the structured report module.
the invention has the technical effects that:
1. Because the invention is provided with the image judgment module, the image processing module, the two-classification intelligent diagnosis module and the structured report module, the right chest picture image can be identified, the right chest picture image which meets the preset condition is processed, the two-classification intelligent diagnosis module carries out normal and abnormal diagnosis on the processed qualified chest picture, and the structured report module activates or deactivates the non-finding related control in the structured report interface according to the normal data or the abnormal data; the system can intelligently divide the chest radiography positive position images into two types: normal and abnormal, the normal chest radiography image can automatically form a structured diagnosis report, thereby reducing the work load of doctor reading, improving the diagnosis efficiency and the speed of writing the structured report;
2. Because the feedback unit is arranged, the unique ID of the patient corresponding to the non-righting chest image in the DICOM image and the righting chest image which does not meet the preset condition is sent to the structured report module, so that a doctor can know which chest images do not enter the diagnosis process of the two-classification intelligent diagnosis module in time, the doctor can make further diagnosis, and the process of the system is more humanized;
3. Because the thermodynamic diagram generating unit is arranged, suspicious focus regions can be marked on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, the thermodynamic diagram is generated and sent to the structural report module, and the thermodynamic diagram is presented to a doctor in a key image form on a structural report interface, so that the doctor can be assisted in further diagnosis of the chest films, and the workload of the doctor is reduced.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
Fig. 1 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a first embodiment of the invention;
fig. 2 is a schematic diagram illustrating that when a structured report module in a chest X-ray image intelligent diagnosis system receives normal data, no finding related control in a structured report interface is activated;
Fig. 3 is a schematic diagram illustrating a diagnosis report automatically generating a qualified chest radiograph when a structured report module in an intelligent chest X-ray image diagnosis system receives normal data according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a second embodiment of the invention;
fig. 5 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a third embodiment of the invention;
Fig. 6 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a fourth embodiment of the present invention;
Fig. 7 is a schematic diagram of a thermodynamic diagram generation unit in the intelligent diagnostic system for chest X-ray images generating thermodynamic diagrams and sending the thermodynamic diagrams to a structured report interface according to a fourth embodiment of the invention;
FIG. 8 is a flowchart of a chest X-ray image intelligent diagnosis method according to a fifth embodiment of the present invention;
Fig. 9 is a schematic diagram illustrating that when the structured report module receives normal data in the intelligent diagnostic method for chest X-ray images according to the fifth embodiment of the present invention, no finding related control in the structured report interface is activated;
Fig. 10 is a schematic diagram illustrating that when the structured report module receives normal data, a qualified chest radiograph is automatically generated according to the fifth embodiment of the present invention;
Fig. 11 is a schematic diagram of a thermodynamic diagram generation unit generating a thermodynamic diagram and sending the thermodynamic diagram to a structured report interface in the chest X-ray image intelligent diagnosis method according to the fifth embodiment of the present invention;
Fig. 12 is a flowchart illustrating a detailed process of a chest X-ray image intelligent diagnosis method according to a sixth embodiment of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Example one
Fig. 1 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a first embodiment of the invention; as shown in fig. 1, the system includes: the image information management module 10, the image judgment module 20, the image processing module 30, the two-classification intelligent diagnosis module 40 and the structured report module 50, wherein,
The image information management module 10 is connected with the image judgment module and is used for transmitting the medical digital imaging and communication DICOM image of the patient to the image judgment module 20 through a DICOM protocol when the patient finishes the examination of the chest radiography which is an examination item;
Wherein, the image Information management module is an RIS (radio Information System) system; the types of patients are outpatient patients, physical examination patients and routine examinations before hospitalization.
The image judgment module 20 is respectively connected with the image information management module 10 and the image processing module 30, and is used for identifying the chest radiograph in DICOM images, analyzing the quality of the chest radiograph based on preset conditions, and sending the chest radiograph meeting the preset conditions to the image processing module 30; wherein, the righting chest picture image which meets the preset condition is defined as a qualified chest picture;
the image judging module firstly identifies the positive image and then analyzes whether the image quality of the positive image meets the preset condition.
And the image processing module 30 is respectively connected with the image judging module 20 and the two-classification intelligent diagnosis module 40, and is used for processing the position, the size, the brightness and the contrast of the qualified chest radiograph, generating a picture format with unified specification, and sending the processed qualified chest radiograph to the two-classification intelligent diagnosis module 40.
the generated picture format with unified specification is a format suitable for a two-classification intelligent diagnosis module, the two-classification intelligent diagnosis module is an AI model for judging normality and abnormality of the positive chest image, and a convolutional neural network can be used for training the model. The size of the processed acceptable chest film can be 256 pixels by 256 pixels, and the processing of the window width window level needs to be processed according to the corresponding window width window level given by meta of the DICOM file.
The second classification intelligent diagnosis module 40 is respectively connected with the image processing module 30 and the structured report module 50, and is used for carrying out normal or abnormal diagnosis on the processed qualified chest radiograph and sending normal data or abnormal data to the structured report module 50;
The structured report module 50 is connected with the two-classification intelligent diagnosis module 40 and is used for activating the non-finding related control in the structured report interface when normal data are received, and automatically generating a diagnosis report of a qualified chest radiograph; when the abnormal data is received, the non-finding related control in the structured report interface is inactivated.
fig. 2 is a schematic diagram illustrating that when a structured report module in an intelligent diagnostic system for chest X-ray images receives normal data, a control that is not found in a structured report interface is activated, as shown in fig. 2, an AI result displayed in the structured report interface is normal, and at the same time, a "control that is not found to be abnormal" is automatically selected.
fig. 3 is a schematic diagram of a diagnosis report automatically generating a qualified chest radiograph when a structured report module in an intelligent chest X-ray image diagnosis system receives normal data according to an embodiment of the present invention, and as shown in fig. 3, the diagnosis impression is "no abnormality in diaphragm of both lungs".
The embodiment of the invention is provided with an image judgment module, an image processing module, a two-classification intelligent diagnosis module and a structural report module, which can identify the chest radiography image and process the chest radiography image which meets the preset condition, the two-classification intelligent diagnosis module carries out normal and abnormal diagnosis on the processed qualified chest radiography, and the structural report module activates or deactivates the non-finding related control in the structural report interface according to the normal data or the abnormal data; the system can intelligently divide the chest radiography positive position images into two types: normal and abnormal, the normal chest radiography image can automatically form a structured diagnosis report, the work load of doctor reading is reduced, the diagnosis efficiency is improved, and the speed of writing the structured report is increased.
example two
Fig. 4 shows a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a second embodiment of the present invention, and as shown in fig. 4, the image determination module 20 further includes a feedback unit 202, connected to the structured report module 50, for sending the unique patient ID corresponding to the non-orthostatic chest image and the orthostatic chest image that does not meet the preset condition in the DICOM image to the structured report module for further diagnosis by the doctor.
For example, the chest side image in the received DICOM image is an unsatisfactory chest, or the chest side image in the right position has an unsatisfactory image quality due to some reasons, and is also an unsatisfactory chest. The feedback unit sends the unique ID of the patient corresponding to the chest film which does not meet the requirement to the structured report module, so that the doctor can analyze the reason of the non-compliance and make further diagnosis.
The embodiment of the invention is provided with a feedback unit, and the unique ID of the patient corresponding to the non-righting chest image in the DICOM image and the righting chest image which does not meet the preset condition is sent to the structured report module, so that a doctor can know which chest images do not enter the diagnosis process of the two-classification intelligent diagnosis module in time for further diagnosis of the doctor, and the process of the system is more humanized.
EXAMPLE III
Fig. 5 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a third embodiment of the invention; as shown in fig. 5, the intelligent diagnosis module for binary classification 40 includes a feature extraction unit 402 and a calculation unit 404, wherein,
the feature extraction unit 402 is connected with the calculation unit 404 and is used for extracting feature data of the processed qualified chest radiograph and sending the feature data to the calculation unit 404;
A calculating unit 404, connected to the feature extracting unit 402, for performing a combined calculation of the and or non-relation of features on the feature data and the fully connected layer trained by the plurality of normal chest pictures and abnormal chest pictures, calculating the normal probability and abnormal probability of the qualified chest pictures after the processing, and outputting the normal data when the normal probability is greater than a first preset threshold; and outputting abnormal data when the abnormal probability is greater than a second preset threshold value.
For example, the feature extraction unit extracts features of the chest radiographs by using a transfer learning method and a convolution layer of two deep learning classification models with structures of VGG16 and ResNet101 which are pre-trained by a large number of pictures, and the reason for using a plurality of models is to improve the classification (abnormal and normal) effect. The extracted features are from macro to micro, and mainly comprise various abstract features such as textures, forms and the like.
and the calculation unit is used for performing characteristic combination on the full-connection layers trained by a large number of normal chest pictures and abnormal chest pictures, summing output results of the VGG16 model and the ResNet101 model, averaging the output results of each category respectively, and finally outputting an array, wherein the calculation result is a number between two 0-1, and can be understood as the probability of being recognized as each category. And judging the output result through a group of preset threshold values specified by a doctor, wherein if [0.5,0.5] represents that more than 0.5 is 1, less than 0.5 is 0, and the value output finally is the value if [0, 1] represents normal, and [1, 0] represents abnormal.
The above description of the feature extraction unit and the calculation unit is only an embodiment, and the number of models used by the feature extraction unit and the calculation principle of the calculation unit are not limited herein.
Example four
Fig. 6 is a schematic structural diagram of a chest X-ray image intelligent diagnosis system according to a fourth embodiment of the present invention; as shown in fig. 6, when the bi-classification intelligent diagnosis module 40 diagnoses the processed qualified chest radiograph as abnormal, the bi-classification intelligent diagnosis module 40 further includes: a thermodynamic diagram generating unit 406, connected to the calculating unit 404, for marking suspicious lesion areas on the qualified chest films of the abnormality based on the models and weights trained by the plurality of normal chest films and abnormal chest films, and generating a thermodynamic diagram to be sent to the structured reporting module 50.
If the qualified chest film is diagnosed as abnormal, the thermodynamic diagram generating unit extracts a main basis identified as abnormal by a machine according to the result output by the calculating unit and the weight of the model, marks suspicious lesion areas on the abnormal qualified chest film, and generates a thermodynamic diagram to be sent to the structured reporting module. The structured report interface is presented to the doctor in the form of a key image, for example, the lesion area is marked by displaying a relatively obvious warm tone, and the marking mode is not limited herein.
the format of the generated thermodynamic diagram may be JPG, which is not limited herein.
fig. 7 is a schematic diagram of a thermodynamic diagram generation unit in the intelligent diagnostic system for chest X-ray images generating thermodynamic diagrams and sending the thermodynamic diagrams to a structured report interface according to a fourth embodiment of the invention; as shown in fig. 7, the left AI result shows abnormal (abnormal), the right side shows the thermodynamic diagram of the chest film, and a warm tone indicating a lesion region is displayed.
the embodiment of the invention is provided with the thermodynamic diagram generation unit, suspicious lesion areas can be marked on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, the thermodynamic diagram is generated and sent to the structured report module, and the thermodynamic diagram is presented to a doctor in a key image form on a structured report interface, so that the doctor can be assisted in further diagnosis of the chest films, and the workload of the doctor is reduced.
EXAMPLE five
FIG. 8 is a flowchart of a chest X-ray image intelligent diagnosis method according to a fifth embodiment of the present invention; as shown in fig. 8, the method comprises the steps of:
Step S501, when the examination item of the patient is the examination of the chest radiography, the image information management module transmits the medical digital imaging and communication DICOM image of the patient to the image judgment module through the DICOM protocol;
Wherein, the image Information management module is an RIS (radio Information System) system; the types of patients are outpatient patients, physical examination patients and routine examinations before hospitalization.
Step S502, the image judgment module identifies the chest radiography image in the DICOM image, analyzes the quality of the chest radiography image based on the preset condition, and sends the chest radiography image meeting the preset condition to the image processing module; wherein, the righting chest picture image which meets the preset condition is defined as a qualified chest picture;
the image judging module firstly identifies the positive image and then analyzes whether the quality of the positive image meets the preset condition.
Step S503, the image processing module processes the position, size, brightness and contrast of the qualified chest radiography to generate a uniform picture format, and sends the processed qualified chest radiography to the two-classification intelligent diagnosis module;
the generated picture format with unified specification is a format suitable for a two-classification intelligent diagnosis module, the two-classification intelligent diagnosis module is an AI model for judging normality and abnormality of the positive chest image, and a convolutional neural network can be used for training the model. The size of the processed acceptable chest film can be 256 pixels by 256 pixels, and the processing of the window width window level needs to be processed according to the corresponding window width window level given by meta of the DICOM file.
Step S504, the two-classification intelligent diagnosis module carries out normal or abnormal diagnosis on the processed qualified chest radiograph and sends normal data or abnormal data to the structured report module;
step S505, when normal data is received, the structured report module activates the non-discovery related control in the structured report interface, and automatically generates a diagnosis report of a qualified chest radiograph; when abnormal data is received, inactivating non-discovery related controls in the structured report interface;
Fig. 9 is a schematic diagram illustrating that when the structured report module receives normal data in the intelligent diagnostic method for chest X-ray images according to the fifth embodiment of the present invention, a control that is not found in the structured report interface is activated, as shown in fig. 9, an AI result displayed in the structured report interface is normal, and at the same time, a "control that is not found to be abnormal" is automatically selected.
fig. 10 is a schematic diagram of a diagnosis report automatically generating a qualified chest radiograph when the structured report module receives normal data in the chest X-ray image intelligent diagnosis method according to the fifth embodiment of the present invention, and as shown in fig. 10, the diagnosis impression is "no abnormality in diaphragm of both lungs".
wherein, the method also comprises: and a feedback unit in the image judgment module sends the unique ID of the patient corresponding to the non-righting chest image in the DICOM image and the righting chest image which does not meet the preset condition to the structured report module for further diagnosis by a doctor.
For example, the chest side image in the received DICOM image is an unsatisfactory chest, or the chest side image in the right position has an unsatisfactory image quality due to some reasons, and is also an unsatisfactory chest. The feedback unit sends the unique ID of the patient corresponding to the chest film which does not meet the requirement to the structured report module, so that the doctor can analyze the reason of the non-compliance and make further diagnosis.
wherein, the diagnosis of the normal or abnormal qualified chest radiography after the treatment comprises the following steps: a feature extraction unit in the two-classification intelligent diagnosis module extracts the processed feature data of the qualified chest radiograph and sends the feature data to a calculation unit; the calculation unit performs combined calculation of the and or not relation of the features on the feature data and the full-connection layer trained by the plurality of normal chest pictures and abnormal chest pictures, calculates the normal probability and the abnormal probability of the qualified chest pictures after the processing, and outputs normal data when the normal probability is greater than a first preset threshold value; and outputting abnormal data when the abnormal probability is greater than a second preset threshold value.
For example, the feature extraction unit extracts features of the chest radiographs by using a transfer learning method and a convolution layer of two deep learning classification models with structures of VGG16 and ResNet101 which are pre-trained by a large number of pictures, and the reason for using a plurality of models is to improve the classification (abnormal and normal) effect. The extracted features are from macro to micro, and mainly comprise various abstract features like textures, morphologies and the like.
and the calculation unit is used for performing characteristic combination on the full-connection layers trained by a large number of normal chest pictures and abnormal chest pictures, summing output results of the VGG16 model and the ResNet101 model, averaging the output results of each category respectively, and finally outputting an array, wherein the calculation result is a number between two 0-1, and can be understood as the probability of being recognized as each category. And judging the output result through a group of preset threshold values specified by a doctor, wherein if [0.5,0.5] represents that more than 0.5 is 1, less than 0.5 is 0, and the value output finally is the value if [0, 1] represents normal, and [1, 0] represents abnormal.
the above description of the feature extraction unit and the calculation unit is only an embodiment, and the number of models used by the feature extraction unit and the calculation principle of the calculation unit are not limited herein.
When the two-classification intelligent diagnosis module diagnoses the processed qualified chest radiograph as abnormal, the method further comprises the following steps: a thermodynamic diagram generation unit in the two-classification intelligent diagnosis module marks suspicious focus areas on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, generates a thermodynamic diagram and sends the thermodynamic diagram to the structured report module.
If the qualified chest film is diagnosed as abnormal, the thermodynamic diagram generating unit extracts a main basis identified as abnormal by a machine according to the result output by the calculating unit and the weight of the model, marks suspicious lesion areas on the abnormal qualified chest film, and generates a thermodynamic diagram to be sent to the structured reporting module. The structured report interface is presented to the doctor in the form of a key image, for example, the lesion area is marked by displaying a relatively obvious warm tone, and the marking mode is not limited herein.
the format of the generated thermodynamic diagram may be JPG, which is not limited herein.
fig. 11 is a schematic diagram of a thermodynamic diagram generation unit in the intelligent diagnostic system for chest X-ray images generating thermodynamic diagrams and sending the thermodynamic diagrams to a structured report interface according to a fifth embodiment of the invention; as shown in fig. 11, the left AI result shows abnormal (abnormal), the right side shows a thermodynamic diagram of the chest film, and a warm tone indicating a lesion region is displayed.
The image judgment module, the image processing module, the two-classification intelligent diagnosis module and the structured report module in the embodiment of the invention can identify the chest radiograph in the right position and process the chest radiograph which meets the preset condition, the two-classification intelligent diagnosis module carries out normal and abnormal diagnosis on the processed qualified chest radiograph, and the structured report module activates or deactivates the non-finding related control in the structured report interface according to the normal data or the abnormal data; the method can intelligently divide the chest radiography positive position images into two types: normal and abnormal, the normal chest film normal position image can automatically form a structured diagnosis report, the work load of doctor reading is reduced, the diagnosis efficiency is improved, and the speed of writing the structured report is increased; due to the feedback unit in the embodiment of the invention, the unique ID of the patient corresponding to the non-righting chest images in the DICOM images and the righting chest images which do not meet the preset conditions is sent to the structured report module, so that a doctor can timely know which chest images do not enter the diagnosis process of the two-classification intelligent diagnosis module, the doctor can make further diagnosis, and the process of the system is more humanized; because the thermodynamic diagram generating unit in the embodiment of the invention can mark suspicious lesion areas on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, generate a thermodynamic diagram and send the thermodynamic diagram to the structural report module, and the thermodynamic diagram is presented to a doctor in a key image form on a structural report interface, the thermodynamic diagram generating unit can assist the doctor to make further diagnosis on the chest films, and the workload of the doctor is reduced.
EXAMPLE six
Fig. 12 is a flowchart illustrating a detailed process of a chest X-ray image intelligent diagnosis method according to a sixth embodiment of the invention, as shown in fig. 12, the method has the following steps:
Step S601, if the chest examination item? exists, executing step S602, otherwise, returning to the beginning;
step S602, the DICOM image of the patient is sent to an image judgment module;
step S603, if the DIOCM image is the chest radiograph with correct position?, executing step S604, otherwise, executing step S603-1;
Step S603-1, sending the unique ID of the patient to a structured report module;
Step S604, judging whether the quality of the righting chest picture image meets the preset condition?, if so, executing step S605, otherwise, returning to step S603-1;
Step S605, processing the position, size, brightness and contrast of the qualified chest film;
Step S606, the two-classification intelligent diagnosis module judges whether the qualified chest piece is normal?, if so, the step S608 is executed, if not, the step S607 is executed;
Step S607, the thermodynamic diagram generation unit generates a thermodynamic diagram; step S607-1, the two-classification intelligent diagnosis module sends abnormal data and thermodynamic diagrams to the structural report module; step S607-2, the doctor can further diagnose the qualified chest radiography according to the abnormal data and the thermodynamic diagram;
Step S608, the two-classification intelligent diagnosis module sends normal data to the structured report module;
Step S608-1, after the structured report module receives the normal data, activating the non-finding related control in the structured report interface, and automatically generating the diagnosis report of the qualified chest radiograph.
From the above description, it can be seen that the above-described embodiments of the present invention achieve the following technical effects: because the invention is provided with the image judgment module, the image processing module, the two-classification intelligent diagnosis module and the structured report module, the right chest film image can be identified, the right chest film image which meets the preset condition is processed, the two-classification intelligent diagnosis module carries out normal and abnormal diagnosis on the processed qualified chest film, and the structured report module activates or deactivates the non-finding related control in the structured report interface according to the normal data or the abnormal data; the system can intelligently divide the positive chest radiography images into two types: normal and abnormal, the normal chest film normal position image can automatically form a structured diagnosis report, the work load of doctor reading is reduced, the diagnosis efficiency is improved, and the speed of writing the structured report is increased; because the feedback unit is arranged, the unique ID of the patient corresponding to the non-righting chest image in the DICOM image and the righting chest image which does not meet the preset condition is sent to the structured report module, so that a doctor can know which chest images do not enter the diagnosis process of the two-classification intelligent diagnosis module in time, the doctor can make further diagnosis, and the process of the system is more humanized; because the thermodynamic diagram generating unit is arranged, suspicious focus regions can be marked on abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, the thermodynamic diagram is generated and sent to the structural report module, and the thermodynamic diagram is presented to a doctor in a key image form on a structural report interface, so that the doctor can be assisted in further diagnosis of the chest films, and the workload of the doctor is reduced.
it will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and they may alternatively be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, or fabricated separately as individual integrated circuit modules, or fabricated as a single integrated circuit module from multiple modules or steps. Thus, the present invention is not limited to any specific combination of hardware and software.
the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An intelligent diagnostic system for chest X-ray images, which is characterized by comprising an image information management module, an image judgment module, an image processing module, a two-classification intelligent diagnostic module and a structural report module,
The image information management module is connected with the image judgment module and is used for transmitting the medical digital imaging and communication DICOM image of the patient to the image judgment module through a DICOM protocol when the patient finishes the examination that the examination item is a chest radiography;
The image judging module is respectively connected with the image information management module and the image processing module and is used for identifying the chest radiograph in DICOM images, analyzing the quality of the chest radiograph based on preset conditions and sending the chest radiograph meeting the preset conditions to the image processing module; wherein, the righting chest picture image meeting the preset condition is defined as a qualified chest picture;
the image processing module is respectively connected with the image judging module and the two-classification intelligent diagnosis module and is used for processing the position, the size, the brightness and the contrast of the qualified chest radiography to generate a picture format with uniform specification and sending the processed qualified chest radiography to the two-classification intelligent diagnosis module;
the two-classification intelligent diagnosis module is respectively connected with the image processing module and the structured report module and is used for carrying out normal or abnormal diagnosis on the processed qualified chest radiograph and sending normal data or abnormal data to the structured report module;
The structured report module is connected with the two-classification intelligent diagnosis module and is used for activating the non-finding related control in a structured report interface when the normal data is received, and automatically generating a diagnosis report of the qualified chest radiograph; and when the abnormal data is received, deactivating the non-finding related control in the structured report interface.
2. the intelligent diagnostic system for chest X-ray images according to claim 1, wherein the image determining module further comprises a feedback unit connected to the structural report module, for sending the unique ID of the patient corresponding to the non-orthostatic chest image and the orthostatic chest image not meeting the preset condition in the DICOM image to the structural report module for further diagnosis by the doctor.
3. the intelligent diagnostic system for chest X-ray images as claimed in claim 1, wherein said two-classification intelligent diagnostic module comprises a feature extraction unit and a calculation unit, wherein,
The feature extraction unit is connected with the calculation unit and is used for extracting the feature data of the processed qualified chest radiograph and sending the feature data to the calculation unit;
the computing unit is connected with the feature extraction unit and is used for performing combined computation of the and or the non-relation of features on the feature data and a full connection layer trained by a plurality of normal chest pictures and abnormal chest pictures, computing the normal probability and the abnormal probability of the processed qualified chest pictures, and outputting the normal data when the normal probability is greater than a first preset threshold value; and when the abnormal probability is larger than a second preset threshold value, outputting the abnormal data.
4. the intelligent diagnostic system for chest X-ray images as claimed in claim 3, wherein when the two-classification intelligent diagnostic module diagnoses the processed qualified chest film as abnormal, the two-classification intelligent diagnostic module further comprises: and the thermodynamic diagram generating unit is connected with the calculating unit and is used for marking suspicious lesion areas on the abnormal qualified chest films based on the models and the weights trained by the plurality of normal chest films and abnormal chest films, and generating a thermodynamic diagram to be sent to the structural reporting module.
5. an intelligent diagnostic method for chest X-ray images is characterized by comprising the following steps:
when the patient finishes the examination of the chest radiography, the image information management module transmits the medical digital imaging and communication DICOM image of the patient to the image judgment module through the DICOM protocol;
the image judging module identifies a correct chest image in the DICOM image, analyzes the quality of the correct chest image based on a preset condition, and sends the correct chest image meeting the preset condition to an image processing module; wherein, the righting chest picture image meeting the preset condition is defined as a qualified chest picture;
the image processing module processes the position, size, brightness and contrast of the qualified chest radiograph to generate a uniform picture format, and sends the processed qualified chest radiograph to the two-classification intelligent diagnosis module;
the two-classification intelligent diagnosis module carries out normal or abnormal diagnosis on the processed qualified chest radiograph and sends normal data or abnormal data to the structured report module;
when the normal data is received, the structured report module activates the non-finding related control in the structured report interface to automatically generate the diagnosis report of the qualified chest radiography; and when the abnormal data is received, deactivating the non-finding related control in the structured report interface.
6. the intelligent diagnostic method for chest X-ray images as claimed in claim 5, further comprising: and a feedback unit in the image judgment module sends the non-righting chest images in the DICOM images and the unique ID of the patient corresponding to the righting chest image which does not meet the preset condition to the structured report module for further diagnosis of doctors.
7. the intelligent diagnostic method for chest X-ray images as claimed in claim 5, wherein said diagnosing whether the processed qualified chest film is normal or abnormal comprises:
A feature extraction unit in the two-classification intelligent diagnosis module extracts feature data of the processed qualified chest radiograph and sends the feature data to a calculation unit;
The calculation unit performs characteristic and-or-not relation combined calculation on the characteristic data and the full-connection layer trained by the plurality of normal chest pictures and abnormal chest pictures, calculates the normal probability and the abnormal probability of the processed qualified chest pictures, and outputs the normal data when the normal probability is greater than a first preset threshold value; and when the abnormal probability is larger than a second preset threshold value, outputting the abnormal data.
8. the intelligent diagnostic method for chest X-ray images as claimed in claim 7, wherein when the two-classification intelligent diagnostic module diagnoses the processed qualified chest film as abnormal, the method further comprises: and a thermodynamic diagram generation unit in the two-classification intelligent diagnosis module marks suspicious lesion areas on the abnormal qualified chest films based on models and weights trained by a plurality of normal chest films and abnormal chest films, generates a thermodynamic diagram and sends the thermodynamic diagram to the structural report module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910667141.1A CN110555825A (en) | 2019-07-23 | 2019-07-23 | Intelligent diagnostic system and diagnostic method for chest X-ray image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910667141.1A CN110555825A (en) | 2019-07-23 | 2019-07-23 | Intelligent diagnostic system and diagnostic method for chest X-ray image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110555825A true CN110555825A (en) | 2019-12-10 |
Family
ID=68735843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910667141.1A Pending CN110555825A (en) | 2019-07-23 | 2019-07-23 | Intelligent diagnostic system and diagnostic method for chest X-ray image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110555825A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111415331A (en) * | 2020-03-03 | 2020-07-14 | 北京深睿博联科技有限责任公司 | Abnormality detection method and system based on category relation in positive chest radiograph |
| CN111916185A (en) * | 2020-08-17 | 2020-11-10 | 北京赛迈特锐医疗科技有限公司 | Management system and method for providing image data for AI diagnostic model |
| CN111916186A (en) * | 2020-08-17 | 2020-11-10 | 北京赛迈特锐医疗科技有限公司 | Chest X-ray intelligent diagnosis system and method by sequential AI diagnosis model |
| CN111986182A (en) * | 2020-08-25 | 2020-11-24 | 卫宁健康科技集团股份有限公司 | Auxiliary diagnosis method, system, electronic device and storage medium |
| CN112037212A (en) * | 2020-09-04 | 2020-12-04 | 北京掌引医疗科技有限公司 | Pulmonary tuberculosis DR image identification method based on deep learning |
| CN112168168A (en) * | 2020-10-14 | 2021-01-05 | 北京赛迈特锐医疗科技有限公司 | System and method for automatically and quantitatively evaluating whole body fat by using MR (magnetic resonance) technology |
| CN112241961A (en) * | 2020-09-22 | 2021-01-19 | 深圳视见医疗科技有限公司 | Chest X-ray film auxiliary diagnosis method and system based on deep convolutional neural network |
| CN112255709A (en) * | 2020-10-21 | 2021-01-22 | 中铁第四勘察设计院集团有限公司 | Automatic lifting type flood monitoring device and method based on image recognition |
| CN112259197A (en) * | 2020-10-14 | 2021-01-22 | 北京赛迈特锐医疗科技有限公司 | Intelligent analysis system and method for acute abdomen plain film |
| CN112382360A (en) * | 2020-12-03 | 2021-02-19 | 卫宁健康科技集团股份有限公司 | Automatic generation system of diagnosis report, storage medium and electronic equipment |
| CN113192625A (en) * | 2021-03-08 | 2021-07-30 | 北京航空航天大学 | Lung disease auxiliary diagnosis cloud platform based on deep learning |
| CN113192609A (en) * | 2021-05-20 | 2021-07-30 | 上海商汤智能科技有限公司 | Data display method and device, electronic equipment and storage medium |
| CN113555089A (en) * | 2021-07-14 | 2021-10-26 | 江苏宏创信息科技有限公司 | Artificial intelligence medical image quality control method applied to clinical image |
| WO2021250484A1 (en) * | 2020-06-10 | 2021-12-16 | Parth Chopra | System and method to diagnose respiratory diseases in real-time |
| CN116452579A (en) * | 2023-06-01 | 2023-07-18 | 中国医学科学院阜外医院 | Chest radiography image-based pulmonary artery high pressure intelligent assessment method and system |
| WO2024199554A1 (en) * | 2023-03-30 | 2024-10-03 | 深圳市帝迈生物技术有限公司 | Auxiliary diagnosis information providing apparatus and blood analysis system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| CN107730484A (en) * | 2017-07-26 | 2018-02-23 | 江西中科九峰智慧医疗科技有限公司 | A kind of abnormal rabat intelligent identification Method and system based on deep learning |
| CN107729911A (en) * | 2017-07-26 | 2018-02-23 | 江西中科九峰智慧医疗科技有限公司 | A kind of pulmonary tuberculosis intelligent identification Method and system based on DR |
| CN108324297A (en) * | 2017-01-09 | 2018-07-27 | 通用电气公司 | Image analysis method for assessing image data |
| CN109378043A (en) * | 2017-10-13 | 2019-02-22 | 北京昆仑医云科技有限公司 | Medical image based on patient generates the system and method and medium of diagnosis report |
| CN109411084A (en) * | 2018-11-28 | 2019-03-01 | 武汉大学人民医院（湖北省人民医院） | A kind of intestinal tuberculosis assistant diagnosis system and method based on deep learning |
| CN109512464A (en) * | 2018-11-23 | 2019-03-26 | 清华大学深圳研究生院 | A kind of disorder in screening and diagnostic system |
-
2019
- 2019-07-23 CN CN201910667141.1A patent/CN110555825A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897573A (en) * | 2016-08-01 | 2017-06-27 | 12西格玛控股有限公司 | Use the computer-aided diagnosis system for medical image of depth convolutional neural networks |
| CN108324297A (en) * | 2017-01-09 | 2018-07-27 | 通用电气公司 | Image analysis method for assessing image data |
| CN107730484A (en) * | 2017-07-26 | 2018-02-23 | 江西中科九峰智慧医疗科技有限公司 | A kind of abnormal rabat intelligent identification Method and system based on deep learning |
| CN107729911A (en) * | 2017-07-26 | 2018-02-23 | 江西中科九峰智慧医疗科技有限公司 | A kind of pulmonary tuberculosis intelligent identification Method and system based on DR |
| CN109378043A (en) * | 2017-10-13 | 2019-02-22 | 北京昆仑医云科技有限公司 | Medical image based on patient generates the system and method and medium of diagnosis report |
| CN109512464A (en) * | 2018-11-23 | 2019-03-26 | 清华大学深圳研究生院 | A kind of disorder in screening and diagnostic system |
| CN109411084A (en) * | 2018-11-28 | 2019-03-01 | 武汉大学人民医院（湖北省人民医院） | A kind of intestinal tuberculosis assistant diagnosis system and method based on deep learning |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111415331A (en) * | 2020-03-03 | 2020-07-14 | 北京深睿博联科技有限责任公司 | Abnormality detection method and system based on category relation in positive chest radiograph |
| CN111415331B (en) * | 2020-03-03 | 2023-05-23 | 北京深睿博联科技有限责任公司 | Abnormal detection method and system based on category relation in positive chest radiography |
| WO2021250484A1 (en) * | 2020-06-10 | 2021-12-16 | Parth Chopra | System and method to diagnose respiratory diseases in real-time |
| CN111916185A (en) * | 2020-08-17 | 2020-11-10 | 北京赛迈特锐医疗科技有限公司 | Management system and method for providing image data for AI diagnostic model |
| CN111916186A (en) * | 2020-08-17 | 2020-11-10 | 北京赛迈特锐医疗科技有限公司 | Chest X-ray intelligent diagnosis system and method by sequential AI diagnosis model |
| CN111916185B (en) * | 2020-08-17 | 2024-02-20 | 北京赛迈特锐医疗科技有限公司 | Management system and method for providing image data for AI diagnostic model |
| CN111986182A (en) * | 2020-08-25 | 2020-11-24 | 卫宁健康科技集团股份有限公司 | Auxiliary diagnosis method, system, electronic device and storage medium |
| CN112037212A (en) * | 2020-09-04 | 2020-12-04 | 北京掌引医疗科技有限公司 | Pulmonary tuberculosis DR image identification method based on deep learning |
| CN112241961A (en) * | 2020-09-22 | 2021-01-19 | 深圳视见医疗科技有限公司 | Chest X-ray film auxiliary diagnosis method and system based on deep convolutional neural network |
| CN112259197A (en) * | 2020-10-14 | 2021-01-22 | 北京赛迈特锐医疗科技有限公司 | Intelligent analysis system and method for acute abdomen plain film |
| CN112168168A (en) * | 2020-10-14 | 2021-01-05 | 北京赛迈特锐医疗科技有限公司 | System and method for automatically and quantitatively evaluating whole body fat by using MR (magnetic resonance) technology |
| CN112255709A (en) * | 2020-10-21 | 2021-01-22 | 中铁第四勘察设计院集团有限公司 | Automatic lifting type flood monitoring device and method based on image recognition |
| CN112382360A (en) * | 2020-12-03 | 2021-02-19 | 卫宁健康科技集团股份有限公司 | Automatic generation system of diagnosis report, storage medium and electronic equipment |
| CN112382360B (en) * | 2020-12-03 | 2024-05-31 | 卫宁健康科技集团股份有限公司 | Diagnostic report automatic generation system, storage medium and electronic equipment |
| CN113192625A (en) * | 2021-03-08 | 2021-07-30 | 北京航空航天大学 | Lung disease auxiliary diagnosis cloud platform based on deep learning |
| CN113192609A (en) * | 2021-05-20 | 2021-07-30 | 上海商汤智能科技有限公司 | Data display method and device, electronic equipment and storage medium |
| CN113555089A (en) * | 2021-07-14 | 2021-10-26 | 江苏宏创信息科技有限公司 | Artificial intelligence medical image quality control method applied to clinical image |
| WO2024199554A1 (en) * | 2023-03-30 | 2024-10-03 | 深圳市帝迈生物技术有限公司 | Auxiliary diagnosis information providing apparatus and blood analysis system |
| CN116452579A (en) * | 2023-06-01 | 2023-07-18 | 中国医学科学院阜外医院 | Chest radiography image-based pulmonary artery high pressure intelligent assessment method and system |
| CN116452579B (en) * | 2023-06-01 | 2023-12-08 | 中国医学科学院阜外医院 | Chest radiography image-based pulmonary artery high pressure intelligent assessment method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110555825A (en) | Intelligent diagnostic system and diagnostic method for chest X-ray image | |
| US11553874B2 (en) | Dental image feature detection | |
| Ekert et al. | Deep learning for the radiographic detection of apical lesions | |
| JP2019093137A (en) | Systems and methods to deliver point-of-care alerts for radiological findings | |
| US7961921B2 (en) | System and method for medical diagnosis and tracking using three-dimensional subtraction in a picture archiving communication system | |
| JP6768620B2 (en) | Learning support device, operation method of learning support device, learning support program, learning support system, terminal device and program | |
| CN110559007A (en) | Intelligent diagnosis system and method for skull CT flat scan image | |
| US11735314B2 (en) | Methods and apparatus for the application of machine learning to radiographic images of animals | |
| Dyer et al. | Diagnosis of normal chest radiographs using an autonomous deep-learning algorithm | |
| CN110246580B (en) | Cranial image analysis method and system based on neural network and random forest | |
| JP2019061578A (en) | Learning support device, operation method for learning support device, learning support program, learning support system, and terminal device | |
| CN111986182A (en) | Auxiliary diagnosis method, system, electronic device and storage medium | |
| US20240029901A1 (en) | Systems and Methods to generate a personalized medical summary (PMS) from a practitioner-patient conversation. | |
| CN111916186A (en) | Chest X-ray intelligent diagnosis system and method by sequential AI diagnosis model | |
| Almeida et al. | Lung ultrasound for point-of-care COVID-19 pneumonia stratification: computer-aided diagnostics in a smartphone. First experiences classifying semiology from public datasets | |
| CN110473180A (en) | Recognition methods, system and the storage medium of respiratory chest motion | |
| Chen et al. | Detection of various dental conditions on dental panoramic radiography using Faster R-CNN | |
| US10909674B2 (en) | System and method for context-aware imaging | |
| CN115359907A (en) | Intelligent diagnosis system and method for screening heart diseases on chest X-ray film | |
| İçöz et al. | Evaluation of an Artificial Intelligence System for the Diagnosis of Apical Periodontitis on Digital Panoramic Images | |
| CN115359878A (en) | ICU bedside chest X-ray film automatic contrast diagnosis intelligent reporting system and method | |
| CN107256544A (en) | A kind of prostate cancer image diagnosing method and system based on VCG16 | |
| CN101196962A (en) | Image information indication method of medicine imaging system | |
| US11908136B2 (en) | Respiratory status classifying method and system thereof | |
| US20240212153A1 (en) | Method for automatedly displaying and enhancing AI detected dental conditions |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191210 |
|
| RJ01 | Rejection of invention patent application after publication |