Apparatus and method for precise analysis of severity of arthritis

An apparatus for a precise analysis of a severity of arthritis, comprising:

• an image collection unit configured to collect a medical image having captured a joint of a user;
• a region detection unit configured to detect one or more regions of interest for analyzing arthritis in the medical image through a learned automatic region detection model;
• an individual analysis unit configured to extract quantitative feature values from the detected regions of interest, and derive one or more individual analysis data from among a severity of arthritis, a severity of osteoproliferation, and a severity of subchondral bone hardness based on the feature values;
• an integrated analysis unit configured to finely classify a severity of degenerative arthritis through an integrated analysis model learned based on the individual analysis data;
• and
• a learning unit configured to:
• detect regions of interest from a training medical image based on training data in which training medical images having captured a joint are matched with result data of reading the training medical images;
• and
• learn an individual analysis model configured to analyze a correlation between feature values extracted from the detected regions of interest and the result data,
• wherein the individual analysis model comprises:
• an arthritis severity analysis model configured to extract quantitative feature values for estimating reduction of a joint cavity based on a region of interest in which a joint cavity region is detected;
• an osteoproliferation severity analysis model configured to extract quantitative feature values for presence or absence of osteoproliferation or a progressing degree of osteoproliferation based on a region of interest in which an osteoproliferation probable region is detected;
• and
• a subchondral bone hardness severity analysis model configured to extract quantitative feature values for a progressing degree of subchondral bone hardness based on a region of interest in which a subchondral bone region is detected.

A program configured to be stored in a non-transitory computer-readable storage medium and to be executed by a processor, the program comprising instructions to execute:

• detecting, based on a collected medical image, one or more regions of interest for analyzing arthritis in the medical image through an automatic region detection model;
• extracting quantitative feature values from the detected regions of interest, and deriving one or more individual analysis data from among a severity of arthritis, a severity of osteoproliferation, and a severity of subchondral bone hardness based on the feature values, through an individual analysis model;
• finely classifying a severity of degenerative arthritis analyzed based on the individual analysis data, through an integrated analysis model;
• and
• outputting the classified severity of degenerative arthritis and the individual analysis data to a connected terminal,
• wherein the automatic region detection model is a model that has been learned to detect a region of interest for estimating result data from a training medical image based on training data in which training medical images are matched with result data of reading the training medical images;
• wherein the individual analysis model is a model that has been learned configured to extract feature values from the detected region of interest and analyze a correlation between individual analysis data of the feature values and the result data;
• and
• wherein the integrated analysis model is a model that has been learned to analyze a correlation between data in which the feature values acquired from the individual analysis model are integratively analyzed and the result data.