A Self-ensembling Framework for Semi-supervised Knee Cartilage Defects Assessment with Dual-Consistency

TL;DR

This paper introduces a semi-supervised self-ensembling framework with dual-consistency for improved knee cartilage defect assessment, reducing annotation needs and enhancing classification and localization accuracy.

Contribution

It presents a novel dual-consistency self-ensembling approach that improves semi-supervised knee OA diagnosis by leveraging attention-based learning and teacher-student networks.

Findings

Significantly improved classification accuracy.

Enhanced lesion localization performance.

Reduced dependence on labeled data.

Abstract

Knee osteoarthritis (OA) is one of the most common musculoskeletal disorders and requires early-stage diagnosis. Nowadays, the deep convolutional neural networks have achieved greatly in the computer-aided diagnosis field. However, the construction of the deep learning models usually requires great amounts of annotated data, which is generally high-cost. In this paper, we propose a novel approach for knee cartilage defects assessment, including severity classification and lesion localization. This can be treated as a subtask of knee OA diagnosis. Particularly, we design a self-ensembling framework, which is composed of a student network and a teacher network with the same structure. The student network learns from both labeled data and unlabeled data and the teacher network averages the student model weights through the training course. A novel attention loss function is developed to obtain accurate attention masks. With dual-consistency checking of the attention in the lesion classification and localization, the two networks can gradually optimize the attention distribution and improve the performance of each other, whereas the training relies on partially labeled data only and follows the semi-supervised manner. Experiments show that the proposed method can significantly improve the self-ensembling performance in both knee cartilage defects classification and localization, and also greatly reduce the needs of annotated data.