EDITOR’S CHOICE Convolutional Neural Network for Second Metacarpal Radiographic Osteoporosis Screening Nahom Tecle, MD, * Jack Teitel, MS,† Michael R. Morris, MD, * Numair Sani, BS,† David Mitten, MD, *† Warren C. Hammert, MD* Purpose Osteoporosis and osteopenia are extremely common and can lead to fragility frac- tures. The purpose of this study was to determine whether a computer learning system could classify whether a hand radiograph demonstrated osteoporosis based on the second meta- carpal cortical percentage. Methods We used the second metacarpal cortical percentage as the osteoporosis predictor. A total of 4,000 posteroanterior (PA) radiographs of the hand were standardized through lat- erality correction, vertical alignment correction, segmentation, proxy osteoporosis predictor, and full pipeline. Laterality was classified using a LeNet convolutional neural network (CNN). Vertical alignment classification used 2,000 PA x-rays to determine vertical align- ment of the second metacarpal. We employed segmentation to determine which pixels belong to the second metacarpal from 1,000 PA x-rays using the FSN-8 CNN. The full pipeline was tested on 265 previously unseen PA x-rays. Results Laterality classification accuracy was 99.62%, with a specificity of 100% and sensi- tivity of 99.3%. Rotation of the hand within 10  of vertical was accurate in 93.2% of films. Segmentation was 94.8% accurate. Proxy osteoporosis predictor was 88.4% accurate. Full pipeline accuracy was 93.9%. In the testing data set, the CNN had a sensitivity of 82.4% and specificity of 95.7%. In the balanced data set, 6 of 39 osteoporotic films were classified as nonosteoporotic; sensitivity was 82.4% and specificity, 94.3%. Conclusions We have created a series of CNN that can accurately identify osteoporosis from non-osteoporosis. Furthermore, our CNN is able to make adjustments to images based on laterality and vertical alignment. Clinical relevance Convolutional neural network and computer learning can be used as an adjunct to dual-energy x-ray absorptiometry scans or to screen and make appropriate referrals for further workup in patients with suspected osteoporosis. (J Hand Surg Am. 2020;45(3):175e181. Copyright Ó 2020 by the American Society for Surgery of the Hand. All rights reserved.) Key words Computer neural network, osteoporosis, screening. O STEOPOROSIS AND OSTEOPENIA ARE a public health concern and may result in fragility fractures. As the population ages, the num- ber of individuals at risk for fragility fractures also increases, which presents a considerable potential cost to the health care system. In 2005, more than 2 million osteoporosis-related fractures were reported in the United States at a cost of $17 billion. 1 From From the *Department of Orthopaedics and Rehabilitation and †Health Lab, University of Rochester, Rochester, NY. Received for publication December 27, 2018; accepted in revised form November 22, 2019. No benefits in any form have been received or will be received related directly or indirectly to the subject of this article. Corresponding author: Warren C. Hammert, MD, Department of Orthopaedics and Rehabilitation, University of Rochester, 601 Elmwood Avenue, Box 665, Rochester, NY 14618; e-mail: Warren_Hammert@URMC.Rochester.edu. 0363-5023/20/4503-0001$36.00/0 https://doi.org/10.1016/j.jhsa.2019.11.019 Ó 2020 ASSH r Published by Elsevier, Inc. All rights reserved. r 175