Artificial intelligence for automated detection and measurements of carpal instability signs on conventional radiographs
View abstract on PubMed
Summary
This summary is machine-generated.An AI system accurately measures carpal instability on radiographs, detecting joint distances, angles, and arc interruptions. This tool shows potential for improved clinical support in diagnosing carpal instability.
Area Of Science
- Radiology
- Artificial Intelligence
- Orthopedics
Background
- Carpal instability can be challenging to diagnose using conventional radiographs.
- Objective measurement of carpal alignment is crucial for accurate diagnosis.
Purpose Of The Study
- To develop and validate an artificial intelligence (AI) system for measuring and detecting signs of carpal instability on conventional radiographs.
- To assess the AI system's performance against clinical measurements.
Main Methods
- A retrospective case-control study using two datasets of hand and wrist radiographs.
- AI system developed for measuring scapholunate (SL) joint distances, SL and capitolunate (CL) angles, and carpal arc interruptions.
- Observer study comparing AI performance to five clinicians on a subset of radiographs.
Main Results
- The AI system demonstrated low mean absolute error (MAE) for measuring SL distances (0.65 mm), SL angles (7.9 degrees), and CL angles (5.9 degrees).
- Sensitivity and specificity for detecting carpal arc interruptions were 83% and 64%, respectively.
- AI measurements were comparable to clinicians; arc interruption detection was more accurate than most clinicians.
Conclusions
- A newly developed AI system accurately measures and detects signs of carpal instability on conventional radiographs.
- The AI system has the potential to enhance the detection of carpal arc interruptions.
- This AI tool shows promise in supporting clinicians for diagnosing carpal instability.

