Femur segmentation in X-ray image based on improved U-Net

Abstract

Segmentation of Femur bone from X-ray images is an indispensable step in computer aided analysis of medical images and orthopaedic examinations. It is more complex than segmentation from CT and MR images, due to some associated less dense tissues that are hard to distinguish from the femur bone in X-ray images. This paper presents an improved method based on U-Net to automatically extract the femurs from hip X-ray images. This method changes the structure of the U-Net network, which can effectively map the non-linear relationship between hip image and femur image, and accurately segment femur image. The paper also added the absolute deviation loss function to improve the segmentation effect. Experimental results show that this method is accurate, robust, and achieves an average dice similarity coefficient of 0.966. The segmentation results are satisfactory.

Fully automatic system to detect and segment the proximal femur in pelvic radiographic images for Legg–Calvé–Perthes disease

“…19 In a recent paper, Lianghui et al made further changes to the architecture of the U-Net network. 46 They also removed a a The total number of radiographic images or instances used in the evaluation or analysis.…”

“…We implemented U‐Net with dropout layers to avoid overfitting the training data, which differs from the original network architecture 19 . In a recent paper, Lianghui et al made further changes to the architecture of the U‐Net network 46 . They also removed a pooling layer and added batch normalization layers after each convolution and deconvolution layer to compensate for the model speed.…”

“…They also removed a pooling layer and added batch normalization layers after each convolution and deconvolution layer to compensate for the model speed. To train the network, they use the original cross‐entropy loss function and the absolute loss error function instead of a weighted cross‐entropy loss function 46 …”

Active learning for data efficient semantic segmentation of canine bones in radiographs

“…Precise segmentation of these bone structures is noteworthy as it allows further automated diagnosis of canine hip dysplasia (Moreira da Silva et al, 2021 , 2022 ). However, the joint regions present high noise, low contrast, overlapping tissue, and a narrow gap between the femur and acetabulum (Lianghui et al, 2019 ). As such, the annotation of these regions requires increased attention, a greater level of medical specialization and knowledge of these anatomical structures.…”