Deep Learning-Based End-to-End Diagnosis System for Avascular Necrosis of Femoral Head

As the first diagnostic imaging modality of avascular necrosis femoral head (AVNFH), accurately staging AVNFH from a plain radiograph is critical yet challenging for orthopedists. Thus, we propose deep learning-based diagnosis system (AVN-net). The proposed AVN-net reads radiographs pelvis, conducts diagnosis, and visualizes results automatically. Deep convolutional neural networks are trained to provide an end-to-end solution, covering tasks detection, exam-view identification, side classification, key clinical notes generation. able obtain state-of-the-art testing AUC 0.97 ($95\%$ CI: $0.97-0.98$) in detection significantly greater F1 scores than less-to-moderately experienced orthopedists all tests (p <; 0.01). Furthermore, two real-world pilot studies were conducted support education assistance, respectively, assess utility AVN-net. experimental promising. With as reference, accuracy consistency considerably improved while requiring only 1/4 time. Students self-studying using can learn better faster control group. To best our knowledge, this study research on prospective use by conducting representing application scenarios. We have demonstrated that achieves expert-level performance, provides efficient decision-making, effectively passes experience students.