Learning co-plane attention across MRI sequences for diagnosing twelve types of knee abnormalities

• Published in: Nature communications Vol. 15; no. 1; pp. 7637 - 11
• Main Authors: Qiu, Zelin, Xie, Zhuoyao, Lin, Huangjing, Li, Yanwen, Ye, Qiang, Wang, Menghong, Li, Shisi, Zhao, Yinghua, Chen, Hao
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 02.09.2024; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/1305; 631/114/1314; 692/700/1421/1628; Abnormalities; Accuracy; Adult; Aged; Classification; Decision making; Deep Learning; Female; Humanities and Social Sciences; Humans; Image enhancement; Image sequencing; Knee; Knee - diagnostic imaging; Knee Joint - diagnostic imaging; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medical imaging; Middle Aged; multidisciplinary; Observational learning; Performance evaluation; Radiologists; Reproducibility of Results; ROC Curve; Science; Science (multidisciplinary); Young Adult
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-024-51888-4

Multi-sequence magnetic resonance imaging is crucial in accurately identifying knee abnormalities but requires substantial expertise from radiologists to interpret. Here, we introduce a deep learning model incorporating co-plane attention across image sequences to classify knee abnormalities. To assess the effectiveness of our model, we collected the largest multi-sequence knee magnetic resonance imaging dataset involving the most comprehensive range of abnormalities, comprising 1748 subjects and 12 types of abnormalities. Our model achieved an overall area under the receiver operating characteristic curve score of 0.812. It achieved an average accuracy of 0.78, outperforming junior radiologists (accuracy 0.65) and remains competitive with senior radiologists (accuracy 0.80). Notably, with the assistance of model output, the diagnosis accuracy of all radiologists was improved significantly ( p  < 0.001), elevating from 0.73 to 0.79 on average. The interpretability analysis demonstrated that the model decision-making process is consistent with the clinical knowledge, enhancing its credibility and reliability in clinical practice. The authors present a deep learning model that incorporates co-plane attention across image sequences with a performance comparable to senior radiologists in classifying 12 knee abnormalities from MRI. The model significantly improves diagnostic performance and aligns with clinical observations.