Automatic Segmentation of Ulna and Radius in Forearm Radiographs

Abstract

Automatic segmentation of ulna and radius (UR) in forearm radiographs is a necessary step for single X-ray absorptiometry bone mineral density measurement and diagnosis of osteoporosis. Accurate and robust segmentation of UR is difficult, given the variation in forearms between patients and the nonuniformity intensity in forearm radiographs. In this work, we proposed a practical automatic UR segmentation method through the dynamic programming (DP) algorithm to trace UR contours. Four seed points along four UR diaphysis edges are automatically located in the preprocessed radiographs. Then, the minimum cost paths in a cost map are traced from the seed points through the DP algorithm as UR edges and are merged as the UR contours. The proposed method is quantitatively evaluated using 37 forearm radiographs with manual segmentation results, including 22 normal-exposure and 15 low-exposure radiographs. The average Dice similarity coefficient of our method reached 0.945. The average mean absolute distance between the contours extracted by our method and a radiologist is only 5.04 pixels. The segmentation performance of our method between the normal- and low-exposure radiographs was insignificantly different. Our method was also validated on 105 forearm radiographs acquired under various imaging conditions from several hospitals. The results demonstrated that our method was fairly robust for forearm radiographs of various qualities.

Automatic Segmentation for Favourable Delineation of Ten Wrist Bones on Wrist Radiographs Using Convolutional Neural Network

“…Faisal et al found that the range of Dice coefficients for the segmentation of eight carpal bones was 0.83~0.94 when the locally weighted K-means variational level set was applied, whereas the range was 0.91~0.96 when Fine Mask R-CNN was employed in our study [ 22 ]. Goo et al showed that the mean Dice coefficient of the automatic segmentation of the distal ulna and radius with dynamic programing was about 0.90, when using forearm radiographs [ 17 ], while we achieved a mean [SD] Dice of 0.96 (0.01) with Fine Mask R-CNN. The use of a fracture detection CNN without segmentation, based on a Dense-161, for distal radio-ulnar fractures on plain radiographs showed a sensitivity of 90.3%, with a specificity of 90.3% [ 38 ].…”

“…However, most wrist bone segmentation methods use conventional mathematical methods. Gou et al conducted automatic segmentation through a dynamic programming algorithm [ 17 ], and Manos et al employed the region growing [ 18 ] and region merging algorithms sequentially after pre-processing, using a Canny edge detector [ 19 ]. In addition, some advanced algorithms have been applied to overcome the disadvantages related to each medical image domain by combining these conventional methods [ 20 , 21 ].…”

Deep learning approach for automatic segmentation of ulna and radius in dual-energy X-ray imaging

“…The local entropy method was developed for the detection and segmentation of the radius and ulna bones [ 20 ]. Furthermore, the dynamic programming algorithm was applied to segment the ulna and radius for single-energy X-ray absorptiometry BMD measurement [ 21 ]. However, those methods are easily affected by noise.…”