Accurate Intervertebral Disc Segmentation Approach Based on Deep Learning.

Automatically segmenting specific tissues or structures from medical images is a straightforward task for deep learning models. However, identifying a few specific objects from a group of similar targets can be a challenging task. This study focuses on the segmentation of certain specific intervertebral discs from lateral spine images acquired from an MRI scanner. In this research, an approach is proposed that utilizes MultiResUNet models and employs saliency maps for target intervertebral disc segmentation. First, a sub-image cropping method is used to separate the target discs. This method uses MultiResUNet to predict the saliency maps of target discs and crop sub-images for easier segmentation. Then, MultiResUNet is used to segment the target discs in these sub-images. The distance maps of the segmented discs are then calculated and combined with their original image for data augmentation to predict the remaining target discs. The training set and test set use 2674 and 308 MRI images, respectively. Experimental results demonstrate that the proposed method significantly enhances segmentation accuracy to about 98%. The performance of this approach highlights its effectiveness in segmenting specific intervertebral discs from closely similar discs.

Content-aware specular reflection suppression based on adaptive image inpainting and neural network for endoscopic images.

In this article, a content-aware specular reflection suppression scheme was developed based on adaptive image inpainting and neural network for endoscopic images. To decrease the impact of specular reflection on visual quality, the proposed scheme consists of three parts: reflection detection, reflection region classification, and reflection concealment. To automatically locate specular reflection regions, a thresholding algorithm with a morphological dilation operation is employed. To reduce the effect of specular reflection, an adaptive image inpainting algorithm is devised to deal with different reflection regions. To achieve content-aware image inpainting, a reflection region classification algorithm is designed by analyzing the local image content to adjust the parameters in the proposed image inpainting algorithm. The experimental results demonstrate that the proposed scheme can automatically and correctly not only locate but also conceal specular reflection regions in endoscopic images. Furthermore, since the average SSIM (structural similarity index) value of the proposed scheme is higher than those of the existing methods, our specular reflection suppression scheme is superior to the existing methods.