Automatic Segmentation of Lumbar Spine MRI Images Based on Improved Attention U-Net

Abstract

Lumbar spine segmentation is important to help doctors diagnose lumbar disc herniation (LDH) and patients’ rehabilitation treatment. In order to accurately segment the lumbar spine, a lumbar spine image segmentation algorithm based on improved Attention U-Net is proposed. The algorithm is based on Attention U-Net, the attention module based on multilevel feature map fusion is adopted, two residual modules are introduced instead of the original convolution blocks. a hybrid loss function is used for prediction during the training process, and finally, the image superposition process is realized. In this experiment, we expanded 420 lumbar MRI images of 180 patients to 1000 images and trained them by different algorithms, respectively, and accuracy, recall, and Dice similarity coefficient metrics were used to analyze these algorithms. The results show that compared with SVM, FCN, R-CNN, U-Net, and Attention U-Net models, the improved model achieved better results in all three evaluations, with 95.50%, 94.53%, and 95.01%, respectively, which proves the better performance of the proposed method for segmentation in lumbar disc and caudal vertebrae.

Deep learning system for automated detection of posterior ligamentous complex injury in patients with thoracolumbar fracture on MRI

“…A previous study reported the attention gates in Attention U-net improve the sensitivity and accuracy for dense label predictions by suppressing feature activations in irrelevant regions 20 . Another study on lumbar spine MRI reported that the use of Attention U-net increased the accuracy of lumbar spine segmentation 27 . Similar to previous studies, our study showed that the fractured vertebral body and the BA segmentation results of the Attention U-net were superior to those of the traditional U-net.…”

A dual channel and spatial attention network for automatic spine segmentation of MRI images

“…In the field of computer vision, the attention mechanism is adopted mainly to achieve better performance through adaptively changing the weights of important features. Attention U‐Net, 31 as a typical network model with an attention mechanism, can improve spine image segmentation performance significantly. In the segmentation of the lumbosacral plexus, the attention mechanism 32 improves the accuracy of the network model.…”

MRI Lumbar Spine Semantic Segmentation Using YOLOv8

“…Wang et al [20] proposed an improved Attention U-Net algorithm for lumbar spine image segmentation. The algorithm uses attention-based fusion and residual modules to improve performance.…”

“…The hyperparameters shown in table I were used for the task of detecting and segmentating vertebrae in lumbar spine MRI images. The use of a large batch size (20) and a large number of workers (8) helps to improve the training efficiency of the model. The use of a constant learning rate (0.01) throughout the training process helps to prevent the model from overfitting the training data.…”