Fully RNN for knee ligament tear classification and localization in MRI scans

Diagnosing ligament injuries using MRI scans is a labor-intensive task that requires an expert. In this paper, we propose fully recurrent neural network (RNN) for detecting Anterior Cruciate Ligament (ACL) tears scans. The proposed localizes the ACL and classifies it into several categories: tear, normal healthy. Existing detection methods use deep learning networks based on single sections, in way lose 3D spatial context. To address this, processes sequence of sections so captures YOLOv3 backbone can produce decisions which are then combined by majority voting. Experimental results show improvement over state-of-the-art methods.