Apparatus for quality management of medical image interpretation using machine learning, and method thereof

Abstract

Provided are a computerized image interpretation method and a device for analyzing a medical image. The image interpretation method may include receiving, at a processor, a medical image, and receiving report information including a healthcare worker's judgement result of the medical image. The method may also include generating, at the processor, result information representing correspondence between first lesion information, which is related to a lesion in the medical image acquired on the basis of the medical image, and second lesion information, which is related to a lesion in the medical image acquired on the basis of the report information, by applying the first lesion information and the second lesion information to a third analysis model. The method may further include outputting, at the processor, the result information.

What is claimed is: 
     
         1 . A computerized image analysis method comprising:
 receiving, at a medical image analysis device comprising at least one machine-trained model, a medical image;   processing the medical image, using a first machine-trained model, to obtain first abnormality information relating to at least one first abnormality in the medical image;   receiving, at the medical image analysis device, report information comprising a healthcare worker's judgement result of the medical image;   processing the report information, using a second machine-trained model, to obtain second abnormality information relating to at least one second abnormality in the medical image;   processing the first abnormality information and the second abnormality information, using a third machine-trained model, to generate result information relating to a correspondence between the first abnormality information and the second abnormality information; and   outputting, at the medical image analysis device, the result information.   
     
     
         2 . The image analysis method of  claim 1 , wherein the first abnormality information includes first abnormality area information, wherein the second abnormality information includes second abnormality area information, and wherein each of the first abnormality area information and the second abnormality area information comprises information relating to at least one of a size, location, type or shape of abnormality. 
     
     
         3 . The image analysis method of  claim 2 , wherein processing the first abnormality information and the second abnormality information comprises at least one of determining a combined area of the first and second abnormality areas and determining an overlapping area of the first and second abnormality areas. 
     
     
         4 . The image analysis method of  claim 3 , wherein processing the first abnormality information and the second abnormality information comprises obtaining the result information based on at least one of the combined area and the overlapping area. 
     
     
         5 . The image analysis method of  claim 1 , further comprising transmitting the result information to a server such that healthcare worker evaluation information is generated based on accumulated result information. 
     
     
         6 . The image analysis method of  claim 1 , further comprising transmitting the result information to a server such that hospital evaluation information is generated based on accumulated result information. 
     
     
         7 . The image analysis method of  claim 1 , further comprising transmitting the result information to a server such that medical payment reference information is generated based on accumulated result information. 
     
     
         8 . The image analysis method of  claim 1 , further comprising transmitting the result information to a server such that a patient worklist is generated based on the result information. 
     
     
         9 . The image analysis method of  claim 1 , wherein the first machine-trained model includes a model which has machine-learned correlations between a plurality of past medical images and a plurality of pieces of first past abnormality information regarding the plurality of past medical images,
 wherein the second machine-trained model includes a model which has machine-learned correlations between a plurality of pieces of past report information and a plurality of pieces of second past abnormality information regarding the plurality of pieces of past report information, and   wherein the third machine-trained model includes a model which has machine-learned the plurality of pieces of first past abnormality information, the plurality of pieces of second past abnormality information, and past result information related a correspondence between the plurality of pieces of first past abnormality information and the plurality of pieces of second past abnormality information.   
     
     
         10 . A device for analyzing a medical image, the device comprising:
 a memory storing computer-executable instructions; and   a processor configured to executed the computer-executable instructions,   wherein the processor is configured, by executing the computer-executable instructions, to:
 receive a medical image; 
 process the medical image, using a first machine-trained model, to obtain first abnormality information relating to at least one first abnormality in the medical image; 
 receive report information comprising a healthcare worker's judgement result of the medical image; 
 process the report information, using a second machine-trained model, to obtain second abnormality information relating to at least one second abnormality in the medical image; 
 process the first abnormality information and the second abnormality information, using a third machine-trained model, to generate result information related a correspondence between the first abnormality information and the second abnormality information; and 
 output the result information. 
   
     
     
         11 . The device of  claim 10 , wherein the first abnormality information includes first abnormality area information, wherein the second abnormality information includes second abnormality area information, and wherein each of the first abnormality area information and the second abnormality area information comprises information relating to at least one of a size, location, type or shape of abnormality. 
     
     
         12 . The device of  claim 11 , wherein in processing the first abnormality information and the second abnormality information, the processor is configured to perform at least one of determining a combined area of the first and second abnormality areas and determining an overlapping area of the first and second abnormality areas. 
     
     
         13 . The device of  claim 12 , wherein the processor is further configured to obtain the result information based on at least one of the combined area and the overlapping area. 
     
     
         14 . The device of  claim 10 , wherein the processor is further configured to transmit the result information to a server such that healthcare worker evaluation information is generated based on accumulated result information. 
     
     
         15 . The device of  claim 10 , wherein the processor is further configured to transmit the result information to a server such that hospital evaluation information is generated based on accumulated result information. 
     
     
         16 . The device of  claim 10 , wherein the processor is further configured to transmit the result information to a server such that medical payment reference information is generated based on accumulated result information. 
     
     
         17 . The device of  claim 10 , wherein the processor is further configured to transmit the result information to a server such that a patient worklist is generated based on the result information. 
     
     
         18 . The device of  claim 10 , wherein the first machine-trained model includes a model which has machine-learned correlations between a plurality of past medical images and a plurality of pieces of first past abnormality information regarding the plurality of past medical images,
 wherein the second first machine-trained model includes a model which has machine-learned correlations between a plurality of pieces of past report information and a plurality of pieces of second past abnormality information regarding the plurality of pieces of past report information, and   wherein the third first machine-trained model includes a model which has machine-learned the plurality of pieces of first past abnormality information, the plurality of pieces of second past abnormality information, and past result information related a correspondence between the plurality of pieces of first past abnormality information and the plurality of pieces of second past abnormality information.