LRU-Net: lightweight and multiscale feature extraction for localization of ACL tears region in MRI images

TL;DR

LRU-Net is a lightweight deep learning model that accurately detects ACL tears in MRI images with high precision and efficiency.

Contribution

LRU-Net introduces a lightweight residual U-Net with attention mechanisms and dynamic feature extraction for improved ACL tear localization.

Findings

LRU-Net achieved a Dice Coefficient Score of 97.93% and an IoU of 96.40% in experiments.

The model outperformed several benchmark models like Attention-Unet and Swin-UNet.

It maintains high accuracy while reducing computational requirements.

Abstract

Anterior cruciate ligament (ACL) injuries hold significant clinical importance, making the development of accurate and efficient diagnostic tools essential. Deep learning has emerged as an effective method for detecting ACL tears. However, current models often struggle with multiscale and boundary-sensitive tear patterns and tend to be computationally intensive.

We present LRU-Net, a lightweight residual U-Net designed for ACL tear segmentation. LRU-Net integrates an advanced attention mechanism that emphasizes gradients and leverages the anatomical position of the ACL, thereby improving boundary sensitivity. Furthermore, it employs a dynamic feature extraction module for adaptive multiscale feature extraction. A dense decoder featuring dense connections enhances feature reuse.

In experimental evaluations, LRU-Net achieves a Dice Coefficient Score of 97.93% and an Intersection over Union (IoU) of 96.40%.

It surpasses benchmark models such as Attention-Unet, Attention-ResUnet, InceptionV3-Unet, Swin-UNet, Trans-UNet and Rethinking ResNets. With a reduced computational footprint, LRU-Net provides a practical and highly accurate solution for the clinical analysis of ACL tears.

Figures