# Monte-Carlo Sampling applied to Multiple Instance Learning for Histological Image Classification

## TL;DR

This paper introduces a Monte-Carlo sampling strategy for high-resolution histological image classification within Multiple Instance Learning, demonstrating improved generalization over traditional sampling methods across multiple datasets.

## Contribution

It presents a novel sequential Monte-Carlo patch sampling method that enhances classification performance in high-resolution histological image analysis.

## Key findings

- Outperforms grid and uniform sampling in generalization accuracy
- Validated on artificial and real histological datasets
- Improves classification results for breast cancer and sun exposure

## Abstract

We propose a patch sampling strategy based on a sequential Monte-Carlo method for high resolution image classification in the context of Multiple Instance Learning. When compared with grid sampling and uniform sampling techniques, it achieves higher generalization performance. We validate the strategy on two artificial datasets and two histological datasets for breast cancer and sun exposure classification.