Evaluating Artificial Intelligence for Identifying Key Anatomical Landmarks on Arthroscopic Images During Adult Knee ACL Reconstruction

What this Trial Is About

Researchers are evaluating an artificial-intelligence software designed to recognize key anatomical landmarks used during anterior cruciate ligament ACL reconstruction surgery in adults. The study aims to determine if the software can automatically identify when the arthroscopic image is clear enough to see these landmarks and accurately outline important bony and cartilaginous features on the femur that guide proper tunnel placement. This observational study focuses on improving surgical accuracy without altering the clinical procedure. Participants undergoing standard primary arthroscopic ACL reconstruction at a single center will have short video segments recorded from the existing arthroscopic camera used in their surgery. These videos capture progressive cleaning steps of the lateral wall of the intercondylar notch and the positioning of surgical instruments relative to key anatomical zones. The investigational software analyzes these recordings offline to train and validate algorithms for landmark recognition without affecting surgical decisions or procedure time. During the study, participants follow their usual care while videos are pseudonymized and securely stored for analysis. Researchers will assess the softwares performance using measures like the area under the receiver operating characteristic curve for cleanliness classification and Dice similarity coefficient for anatomical landmark segmentation. No additional devices or surgical maneuvers are involved, and no risks beyond standard surgery are expected. The study will continue through the completion of data collection and analysis over approximately 12 months.

Augmented Reality in Surgery