7-Minute 3T Knee MRI with Deep Learning Reconstruction Shows High Accuracy Versus Arthroscopy

In 117 patients, a 7-minute 5-sequence 3T knee MRI with deep learning reconstruction showed high diagnostic accuracy for arthroscopy-verified tears: ACL sensitivity 97%, specificity 98%, AUC 0.97; medial meniscus AUC 0.87. Interreader agreement was good.

• Image quality was rated good (median 4/5) with minimal noise, good edge sharpness, and negligible reconstruction artifacts; interreader agreement for quality metrics was κ≥0.83.
• Seven musculoskeletal radiologists independently reviewed 117 exams; the deep learning reconstruction pipeline (AIR Recon DL, GE HealthCare) enabled a 7-minute five-sequence protocol.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.