Breast Cancer Tissue Region Classification from Second Harmonic Generation Images via Machine Learning

Abstract

In this paper, we investigate various machine learning methods for the identification of regions of interest related to tissue sub-types relevant to the prediction of breast cancer recurrence. We formulate the task as a multi-class classification problem and use support vector machine as the inference engine and focus on exploring the feature extraction stage. To that end, we evaluate methods leveraging hand-crafted and pre-trained deep features, as well as features extracted via transfer learning. The results show a steady trend of improvement on the classification accuracy as the features become more data-driven and customized to the task at hand.