Development of a joint space width measurement method based on radiographic hand images

Area of Science:

• Medical Imaging
• Rheumatology
• Computer Vision

Background:

• Rheumatoid arthritis (RA) diagnosis and monitoring rely on assessing joint damage.
• Quantifying joint space width (JSW) from radiographic images is crucial for evaluating RA.
• Existing methods may lack precision or automation in analyzing hand radiographs.

Purpose of the Study:

• To develop and validate a novel algorithm for automated measurement of joint space widths (JSWs) in hand radiographs.
• To assess the algorithm's performance in detecting joints and measuring JSWs in patients with rheumatoid arthritis (RA).
• To compare JSWs between healthy individuals and RA patients using the developed algorithm.

Main Methods:

• Radiographic hand images were preprocessed to extract phalangeal regions and branch paths.
• An algorithm automatically detected 14 joints by identifying grayscale intensity changes along phalangeal paths.
• Joint location accuracy and mean JSWs were evaluated using 35 radiographic images from control and RA groups.

Main Results:

• The algorithm achieved a high joint detection rate of 94.69%, with limitations in cases of severe RA deformities.
• Mean JSW was significantly greater in the control group compared to the RA group (p<0.05).
• The standard deviation of JSW was significantly lower in the control group than in RA groups (p<0.005), indicating less joint space variability.

Conclusions:

• The developed algorithm offers an automated and accurate method for measuring JSWs in hand radiographs.
• The findings highlight significant differences in JSW between healthy individuals and RA patients, supporting the algorithm's clinical relevance.
• This tool has the potential to improve the quantitative assessment and monitoring of rheumatoid arthritis progression.