Margin-Aware Intra-Class Novelty Identification for Medical Images
This work addresses the challenge of identifying rare diseases in medical images, which is crucial for healthcare diagnostics, though it appears incremental as it builds on existing autoencoder and classifier-based methods.
The paper tackles the problem of intra-class novelty detection in medical images, where existing methods struggle with subtle variations and lack of out-of-distribution training data, and proposes a hybrid model called TEND that outperforms state-of-the-art approaches on both natural and medical image datasets.
Traditional anomaly detection methods focus on detecting inter-class variations while medical image novelty identification is inherently an intra-class detection problem. For example, a machine learning model trained with normal chest X-ray and common lung abnormalities, is expected to discover and flag idiopathic pulmonary fibrosis which a rare lung disease and unseen by the model during training. The nuances from intra-class variations and lack of relevant training data in medical image analysis pose great challenges for existing anomaly detection methods. To tackle the challenges, we propose a hybrid model - Transformation-based Embedding learning for Novelty Detection (TEND) which without any out-of-distribution training data, performs novelty identification by combining both autoencoder-based and classifier-based method. With a pre-trained autoencoder as image feature extractor, TEND learns to discriminate the feature embeddings of in-distribution data from the transformed counterparts as fake out-of-distribution inputs. To enhance the separation, a distance objective is optimized to enforce a margin between the two classes. Extensive experimental results on both natural image datasets and medical image datasets are presented and our method out-performs state-of-the-art approaches.