Faculty Bibliography

BACKGROUND:Detection of rotator cuff tears, a common cause of shoulder disability, can be time-consuming and subject to reader variability. Deep learning (DL) has the potential to increase radiologist accuracy and consistency. PURPOSE:The aim of this study was to develop a prototype DL model for detection and classification of rotator cuff tears on shoulder magnetic resonance imaging into no tear, partial-thickness tear, or full-thickness tear. MATERIALS AND METHODS:This Health Insurance Portability and Accountability Act-compliant, institutional review board-approved study included a total of 11,925 noncontrast shoulder magnetic resonance imaging scans from 2 institutions, with 11,405 for development and 520 dedicated for final testing. A DL ensemble algorithm was developed that used 4 series as input from each examination: fluid-sensitive sequences in 3 planes and a sagittal oblique T1-weighted sequence. Radiology reports served as ground truth for training with categories of no tear, partial tear, or full-thickness tear. A multireader study was conducted for the test set ground truth, which was determined by the majority vote of 3 readers per case. The ensemble comprised 4 parallel 3D ResNet50 convolutional neural network architectures trained via transfer learning and then adapted to the targeted domain. The final tear-type prediction was determined as the class with the highest probability, after averaging the class probabilities of the 4 individual models. RESULTS:The AUC overall for supraspinatus, infraspinatus, and subscapularis tendon tears was 0.93, 0.89, and 0.90, respectively. The model performed best for full-thickness supraspinatus, infraspinatus, and subscapularis tears with AUCs of 0.98, 0.99, and 0.95, respectively. Multisequence input demonstrated higher AUCs than single-sequence input for infraspinatus and subscapularis tendon tears, whereas coronal oblique fluid-sensitive and multisequence input showed similar AUCs for supraspinatus tendon tears. Model accuracy for tear types and overall accuracy were similar to that of the clinical readers. CONCLUSIONS:Deep learning diagnosis of rotator cuff tears is feasible with excellent diagnostic performance, particularly for full-thickness tears, with model accuracy similar to subspecialty-trained musculoskeletal radiologists.

Imaging evaluation of postoperative spinal infection is challenging. A systematic approach and keen understanding of multimodality imaging techniques, as well as knowledge of the patient's surgical procedure and clinical presentation, are critical for the radiologist to render an accurate diagnosis. Because of the overlap between diagnostic imaging findings in the postoperative spine and the infected spine, in those situations in which the index of clinical suspicion for spine infection is high, then immediate consideration ought to be given to performing a spine biopsy.

Magnetic resonance imaging is the current imaging modality of choice in the evaluation of patients presenting with myelopathic symptoms in the search for spinal cord lesions. It is important for the radiologist to recognize and differentiate nonneoplastic from the neoplastic process of the spinal cord as the differentiation of the 2 entities is extremely crucial to the neurosurgeon. This article presents a broad spectrum of benign intramedullary spinal abnormalities including syrinx, contusion, abscess, infarction, myelitis, multiple sclerosis, sarcoid, cavernoma, and arteriovenous malformation. Rare intramedullary neoplasms including dermoid tumor, astrocytoma, ependymoma, hemangioblastoma, lymphoma, ganglioneuroblastoma, and metastases are also illustrated. The clinical presentation and magnetic resonance signal characteristics as well as the differential diagnosis of the intramedullary lesions are discussed. The potential pitfalls in the differentiation of tumors from nonneoplastic disease of the spinal cord are also elucidated.