Unsupervised Task Design to Meta-Train Medical Image Classifiers
This addresses the challenge of limited task availability for meta-training in few-shot medical image classification, offering a more scalable solution for practitioners.
The paper tackles the problem of costly hand-designed classification tasks for meta-training medical image classifiers by proposing an unsupervised method to design tasks, resulting in a pre-trained model that, after fine-tuning, achieves better classification results than other pre-training methods and is competitive with meta-training using hand-designed tasks on a breast DCE-MRI dataset.
Meta-training has been empirically demonstrated to be the most effective pre-training method for few-shot learning of medical image classifiers (i.e., classifiers modeled with small training sets). However, the effectiveness of meta-training relies on the availability of a reasonable number of hand-designed classification tasks, which are costly to obtain, and consequently rarely available. In this paper, we propose a new method to unsupervisedly design a large number of classification tasks to meta-train medical image classifiers. We evaluate our method on a breast dynamically contrast enhanced magnetic resonance imaging (DCE-MRI) data set that has been used to benchmark few-shot training methods of medical image classifiers. Our results show that the proposed unsupervised task design to meta-train medical image classifiers builds a pre-trained model that, after fine-tuning, produces better classification results than other unsupervised and supervised pre-training methods, and competitive results with respect to meta-training that relies on hand-designed classification tasks.