Quick and accurate selection of hand images among radiographs from various body parts using deep learning

Article type: Research Article

Authors: Fujiwara, Kohei^{a; 1} | Fang, Wanxuan^{b; 1} | Okino, Taichi^{c; 1} | Sutherland, Kenneth^d | Furusaki, Akira^e | Sagawa, Akira^e | Kamishima, Tamotsu^{b; *}

Affiliations: [a] Department of Health Sciences, Hokkaido University, Kita-ku, Sapporo, Japan | [b] Faculty of Health Sciences, Hokkaido University, Kita-ku, Sapporo, Japan | [c] Graduate School of Health Sciences, Hokkaido University, Kita-ku, Sapporo, Japan | [d] Global Center for Biomedical Science and Engineering, Hokkaido University, Kita-ku, Sapporo, Japan | [e] Sagawa Akira Rheumatology Clinic, Chuo-ku, Sapporo, Japan

Correspondence: [*] Corresponding author: Tamotsu Kamishima, MD, PhD, Faculty of Health Sciences, Hokkaido University, North-12 West-5, Kita-ku, Sapporo, 060-0812, Japan. Tel./Fax: +81 11 706 2824; E-mail: ktamotamo2@yahoo.co.jp.

Note: [1] These authors are equally contributed to this study.

Abstract: BACKGROUND:Although rheumatoid arthritis (RA) causes destruction of articular cartilage, early treatment significantly improves symptoms and delays progression. It is important to detect subtle damage for an early diagnosis. Recent software programs are comparable with the conventional human scoring method regarding detectability of the radiographic progression of RA. Thus, automatic and accurate selection of relevant images (e.g. hand images) among radiographic images of various body parts is necessary for serial analysis on a large scale. OBJECTIVE:In this study we examined whether deep learning can select target images from a large number of stored images retrieved from a picture archiving and communication system (PACS) including miscellaneous body parts of patients. METHODS:We selected 1,047 X-ray images including various body parts and divided them into two groups: 841 images for training and 206 images for testing. The training images were augmented and used to train a convolutional neural network (CNN) consisting of 4 convolution layers, 2 pooling layers and 2 fully connected layers. After training, we created software to classify the test images and examined the accuracy. RESULTS:The image extraction accuracy was 0.952 and 0.979 for unilateral hand and both hands, respectively. In addition, all 206 test images were perfectly classified into unilateral hand, both hands, and the others. CONCLUSIONS:Deep learning showed promise to enable efficiently automatic selection of target X-ray images of RA patients.

Keywords: Rheumatoid arthritis, deep learning, image classification, convolutional neural network

DOI: 10.3233/XST-200694

Journal: Journal of X-Ray Science and Technology, vol. 28, no. 6, pp. 1199-1206, 2020