Pathological Image Caption Matching for Automatic Subtype Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technologies for diagnosing lesions such as dermatitis and cancer lack accuracy in determining subtypes based on pathological images, relying heavily on a doctor's experience and skills, and do not adequately utilize machine learning for precise subtype identification.
Innovation Solution
A diagnosis assistance apparatus and method utilizing a first learned model to generate captions in a predetermined format for pathological images, followed by evaluating similarities with database findings to determine the most accurate subtype, potentially enhanced by a second learned model for feature information and threshold comparisons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image retrieval technology is used to find similar pathological images, then a similar image can be extracted for reference, but the accuracy in determining lesion subtypes remains insufficient and requires additional atlas reference by doctors
Solution Approach 1:
The patent replaces the manual mechanical process of doctor-based atlas reference with an automated machine learning system. The learned model automatically extracts features from pathological images and determines lesion subtypes by comparing with database findings, substituting the manual mechanical process of visual comparison and expert judgment with automated computational analysis.
Solution Approach 2:
The patent introduces a learned model as an intermediary between the pathological image and the diagnosis outcome. This intermediary automatically extracts features and determines subtypes, mediating the process between image input and diagnostic conclusion, thereby eliminating the need for direct doctor-atlas interaction while maintaining or improving accuracy.
2Measurement precision
If a doctor manually refers to an atlas for final determination of lesion subtype, then diagnostic accuracy can be improved, but the process becomes time-consuming and less efficient
Solution Approach 1:
The patent enables the system to perform subtype determination autonomously without requiring doctor intervention for atlas reference. The learned model self-servingly extracts features, compares with database findings, and determines subtypes automatically, making the system self-sufficient in the diagnostic process while maintaining accuracy.
Solution Approach 2:
The patent replaces the manual mechanical process of atlas reference with automated computational comparison. The system automatically compares extracted features with stored findings in the database to determine subtypes, substituting the time-consuming manual process with rapid computational analysis.
3Reliability
If conventional techniques are used for lesion subtype determination, then the process relies on doctor experience and skills, but this leads to variability and reduced accuracy in automatic determination
Solution Approach 1:
The patent replaces the variable human-based diagnostic process with a consistent machine learning system. The learned model applies the same feature extraction and comparison algorithms to all pathological images, eliminating variability introduced by different doctors' experiences and skills while maintaining or improving accuracy through systematic automated analysis.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
In order to attain an object to improve accuracy in automatic determination of a subtype carried out on the basis of a pathological image, at least one processor included in a diagnosis assistance apparatus carries out: a first acquisition process of acquiring a first caption from a first learned model, the first learned model being constructed by machine learning so as to generate, in a case where a pathological image is inputted, a caption describing content of the pathological image in a predetermined format, the first caption describing, in the predetermined format, content of a first pathological image to be subjected to diagnosis; a first evaluation process of evaluating a first similarity which is a similarity between (i) content of at least a part of a plurality of findings which are accumulated in a database and which represent, in writing in the predetermined format, a respective plurality of pathological subtypes and (ii) content of the first caption; and an output process of outputting information pertaining to a subtype whose first similarity is evaluated to be the highest among those of the plurality of pathological subtypes. The information outputted in the output process is used for decision making in diagnosis by a doctor.