Synthesizing lesions using contextual GANs improves breast cancer classification on mammograms
This addresses the problem of overfitting in medical imaging for healthcare applications, but it is incremental as it applies an existing GAN-based method to a specific domain.
The paper tackled data scarcity and class imbalance in breast cancer classification on mammograms by using a contextual GAN to synthesize and remove lesions for data augmentation, resulting in significant improvement in malignancy classification performance on a test set of real mammogram patches.
Data scarcity and class imbalance are two fundamental challenges in many machine learning applications to healthcare. Breast cancer classification in mammography exemplifies these challenges, with a malignancy rate of around 0.5% in a screening population, which is compounded by the relatively small size of lesions (~1% of the image) in malignant cases. Simultaneously, the prevalence of screening mammography creates a potential abundance of non-cancer exams to use for training. Altogether, these characteristics lead to overfitting on cancer cases, while under-utilizing non-cancer data. Here, we present a novel generative adversarial network (GAN) model for data augmentation that can realistically synthesize and remove lesions on mammograms. With self-attention and semi-supervised learning components, the U-net-based architecture can generate high resolution (256x256px) outputs, as necessary for mammography. When augmenting the original training set with the GAN-generated samples, we find a significant improvement in malignancy classification performance on a test set of real mammogram patches. Overall, the empirical results of our algorithm and the relevance to other medical imaging paradigms point to potentially fruitful further applications.