An Interpretable and Accurate Deep-Learning Diagnosis Framework Modeled With Fully and Semi-Supervised Reciprocal Learning
- 21 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (1) , 392-404
Abstract
The deployment of automated deep-learning classifiers in clinical practice has the potential to streamline the diagnosis process and improve the diagnosis accuracy, but the acceptance of those classifiers relies on both their accuracy and interpretability. In general, accurate deep-learning classifiers provide little model interpretability, while interpretable models do not have competitive classification accuracy. In this paper, we introduce a new deep-learning diagnosis framework, called InterNRL, that is designed to be highly accurate and interpretable. InterNRL consists of a student-teacher framework, where the student model is an interpretable prototype-based classifier (ProtoPNet) and the teacher is an accurate global image classifier (GlobalNet). The two classifiers are mutually optimised with a novel reciprocal learning paradigm in which the student ProtoPNet learns from optimal pseudo labels produced by the teacher GlobalNet, while GlobalNet learns from ProtoPNet’s classification performance and pseudo labels. This reciprocal learning paradigm enables InterNRL to be flexibly optimised under both fully- and semi-supervised learning scenarios, reaching state-of-the-art classification performance in both scenarios for the tasks of breast cancer and retinal disease diagnosis. Moreover, relying on weakly-labelled training images, InterNRL also achieves superior breast cancer localisation and brain tumour segmentation results than other competing methods.Keywords
Funding Information
- Australian Government under the Medical Research Future Fund for the Transforming Breast Cancer Screening with Artificial Intelligence (BRAIx) Project (MRFAI000090)
- Australian Research Council (FT190100525)
This publication has 47 references indexed in Scilit:
- Automated Analysis of Unregistered Multi-View Mammograms With Deep LearningIEEE Transactions on Medical Imaging, 2017
- Deep Multi-instance Networks with Sparse Label Assignment for Whole Mammogram Classification2017
- DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph searchBiomedical Optics Express, 2017
- Deep Residual Learning for Image Recognition2016
- Learning Deep Features for Discriminative Localization2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)IEEE Transactions on Medical Imaging, 2015
- Deep learningNature, 2015
- Fully automated detection of diabetic macular edema and dry age-related macular degeneration from optical coherence tomography imagesBiomedical Optics Express, 2014