Skip to main content
Fig. 3 | Arthritis Research & Therapy

Fig. 3

From: Deep learning-based automatic-bone-destruction-evaluation system using contextual information from other joints

Fig. 3

Overview of the binary classification model and the same joint groups. We developed four classification models: A A most-basic classification model that independently classifies each joint as intact (SHS = 0) or non-intact (SHS ≥ 1) (single-input single-output model). B A classification model that inputs the same joints of both hands and outputs intact or non-intact, respectively (multiple-input multiple-output model at a specific joint). C A classification model that receives inputs of the same joint group of one hand and outputs whether they are intact or non-intact, respectively (multiple-input multiple-output model on the one hand). D A classification model that receives inputs of the same joint group of both hands and outputs, whether intact or non-intact, respectively (multiple-input multiple-output model on both hands). E The same joint groups of both hands for erosion and JSN. PIP, proximal interphalangeal; IP, interphalangeal; MCP, metacarpophalangeal; CMC-M, carpometacarpal joint of the thumb and multangular; CMC, carpometacarpal

Back to article page