Identification of Vertebral Fractures by Convolutional Neural Networks to Predict Nonvertebral and Hip Fractures: A Registry-based Cohort Study of Dual X-ray Absorptiometry

Abstract

Background: Detection of vertebral fractures (VFs) aids in management of osteoporosis and targeting of fracture prevention therapies.Purpose: To determine whether convolutional neural networks (CNNs) can be trained to identify VFs at VF assessment (VFA) performed with dual-energy x-ray absorptiometry and if VFs identified by CNNs confer a similar prognosis compared with the expert reader reference standard. Materials and Methods: In this retrospective study, 12 742 routine clinical VFA images obtained from February 2010 to December 2017 and reported as VF present or absent were used for CNN training and testing. All reporting physicians were diagnostic imaging specialists with at least 10 years of experience. Randomly selected training and validation sets were used to produce a CNN ensemble that calculates VF probability. A test set (30%; 3822 images) was used to assess CNN agreement with the human expert reader reference standard and CNN prediction of incident non-VFs. Statistical analyses included area under the receiver operating characteristic curve, two-tailed Student t tests, prevalence-and bias-adjusted k value, Kaplan-Meier curves, and Cox proportional hazard models.Results: This study included 12 742 patients (mean age, 76 years 6 7; 12 013 women). The CNN ensemble demonstrated an area under the receiver operating characteristic curve of 0.94 (95% confidence interval [CI]: 0.93, 0.95) for VF detection that corresponded to sensitivity of 87.4% (534 of 611), specificity of 88.4% (2838 of 3211), and prevalence-and bias-adjusted k value of 0.77. On the basis of incident fracture data available for 2813 patients (mean follow up, 3.7 years), hazard ratios adjusted for baseline fracture probability were 1.7 (95% CI: 1.3, 2.2) for CNN versus 1.8 (95% CI: 1.3, 2.3) for expert reader-detected VFs for incident non-VF and 2.3 (95% CI: 1.5, 3.5) versus 2.4 (95% CI: 1.5, 3.7) for incident hip fracture. Conclusion:Convolutional neural networks can identify vertebral fractures on vertebral fracture assessment images with high accuracy, and these convolutional neural network-identified vertebral fractures predict clinical fracture outcomes.

Machine Learning Solutions for Osteoporosis—A Review

“…Among the 32 studies that investigated fracture detection (Table 3), (66–97 ) 11 were on vertebral fractures, ( 66–76 ) 17 hip fractures, ( 74–90 ) and 10 other fracture sites such as humerus or wrist. ( 75,76,90–97 ) Nineteen studies developed CNN models for image analysis.…”

“…( 69,70,73,75,86–89,92,96 ) Three studies shared their code for model development replication. ( 67,75,77 )…”

Joint Associations of Prevalent Radiographic Vertebral Fracture and Abdominal Aortic Calcification With Incident Hip, Major Osteoporotic, and Clinical Vertebral Fractures

“…Substantial progress has already been made developing automated algorithms to accurately recognize vertebral fracture on lateral spine bone density images. ( 35 )…”

Vertebral fracture: epidemiology, impact and use of DXA vertebral fracture assessment in fracture liaison services

“…The use of Hologic Discovery, a later model, improved the sensitivity and specificity per vertebra to 83.6% and 99.1% respectively [71]. Computer-based deep learning methods have been developed for VFA, and a recent study including over 12,000 VFAs has reported an area under the receiver operating characteristic curve of 0.94 (95% confidence interval [CI]: 0.93, 0.95) for vertebral fracture detection [corresponding to a sensitivity of 87.4% (534 of 611) and specificity of 88.4% (2838 of 3211)] and performed comparably to the study radiologists (all who had more than 10 years of experience) [72].…”