Few-shot Diagnosis of Chest x-rays Using an Ensemble of Random Discriminative Subspaces

Kshitiz, Garvit Garg, Angshuman Paul

Keywords: Chest x-rays, Few-shot learning, Random subspaces

TL;DR: We design a few-shot learning method for the diagnosis of chest x-rays using an ensemble of random subspaces. Our method is computationally efficient and improves classification performance on publicly available chest x-ray dataset.

Abstract: Due to the scarcity of annotated data in the medical domain, few-shot learning may be useful for medical image analysis tasks. We design a few-shot learning method using an ensemble of random subspaces for the diagnosis of chest x-rays (CXRs). Our design is computationally efficient and almost 1.8 times faster than method that uses the popular truncated singular value decomposition (t-SVD) for subspace decomposition. The proposed method is trained by minimizing a novel loss function that helps create well-separated clusters of training data in discriminative subspaces. As a result, minimizing the loss maximizes the distance between the subspaces, making them discriminative and assisting in better classification. Experiments on large-scale publicly available CXR datasets yield promising results. Code for the project will be available at https://github.com/Few-shot-Learning-on-chest-x-ray/fsl_subspace

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2309.00081/code)