# Machine Learning on Biomedical Images: Interactive Learning, Transfer Learning, Class Imbalance, and Beyond

## TL;DR

This paper addresses key challenges in machine learning on biomedical images by exploring interactive learning, transfer learning, and data imbalance, demonstrating improved methods and results through three detailed case studies.

## Contribution

It introduces novel approaches and insights for enhancing biomedical image analysis, including a new loss function and effective transfer learning strategies.

## Key findings

- Interactive learning improves exploration efficiency and quality.
- Transfer learning with pre-processing enhances Alzheimer's diagnosis with less data.
- Focal Tversky loss improves segmentation on imbalanced datasets.

## Abstract

In this paper, we highlight three issues that limit performance of machine learning on biomedical images, and tackle them through 3 case studies: 1) Interactive Machine Learning (IML): we show how IML can drastically improve exploration time and quality of direct volume rendering. 2) transfer learning: we show how transfer learning along with intelligent pre-processing can result in better Alzheimer's diagnosis using a much smaller training set 3) data imbalance: we show how our novel focal Tversky loss function can provide better segmentation results taking into account the imbalanced nature of segmentation datasets. The case studies are accompanied by in-depth analytical discussion of results with possible future directions.