A deep learning-based method for the diagnosis of vertebral fractures on spine MRI: retrospective training and validation of ResNet

• Published in: European spine journal 2022-08, Vol.31 (8), p.2022-2030
• Main Authors: Yeh, Lee-Ren, Zhang, Yang, Chen, Jeon-Hor, Liu, Yan-Lin, Wang, An-Chi, Yang, Jie-Yu, Yeh, Wei-Cheng, Cheng, Chiu-Shih, Chen, Li-Kuang, Su, Min-Ying
• Format: Article
• Language: English
• Subjects: Deep Learning; Diagnosis; Diagnosis, Differential; Humans; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Medicine; Medicine & Public Health; Neurosurgery; Original Article; Retrospective Studies; Spinal Fractures - diagnosis; Spinal Neoplasms - pathology; Spine; Surgical Orthopedics; Vertebrae
• ISSN: 0940-6719; 1432-0932
• DOI: 10.1007/s00586-022-07121-1

Purpose To improve the performance of less experienced clinicians in the diagnosis of benign and malignant spinal fracture on MRI, we applied the ResNet50 algorithm to develop a decision support system. Methods A total of 190 patients, 50 with malignant and 140 with benign fractures, were studied. The visual diagnosis was made by one senior MSK radiologist, one fourth-year resident, and one first-year resident. The MSK radiologist also gave the binary score for 15 qualitative imaging features. Deep learning was implemented using ResNet50, using one abnormal spinal segment selected from each patient as input. The T1W and T2W images of the lesion slice and its two neighboring slices were considered. The diagnostic performance was evaluated using tenfold cross-validation. Results The overall reading accuracy was 98, 96, and 66% for the senior MSK radiologist, fourth-year resident, and first-year resident, respectively. Of the 15 imaging features, 10 showed a significant difference between benign and malignant groups with p