# Label Stability in Multiple Instance Learning

## TL;DR

This paper investigates the stability of instance labels in multiple instance learning classifiers, especially in medical imaging, proposing an unsupervised measure to evaluate and balance stability with performance.

## Contribution

It introduces an unsupervised measure for assessing instance label stability in MIL classifiers and explores the trade-off between stability and accuracy.

## Key findings

- MIL classifiers show varying stability across datasets.
- A performance-stability trade-off exists in MIL classifiers.
- Medical image datasets reveal instability issues in instance labeling.

## Abstract

We address the problem of \emph{instance label stability} in multiple instance learning (MIL) classifiers. These classifiers are trained only on globally annotated images (bags), but often can provide fine-grained annotations for image pixels or patches (instances). This is interesting for computer aided diagnosis (CAD) and other medical image analysis tasks for which only a coarse labeling is provided. Unfortunately, the instance labels may be unstable. This means that a slight change in training data could potentially lead to abnormalities being detected in different parts of the image, which is undesirable from a CAD point of view. Despite MIL gaining popularity in the CAD literature, this issue has not yet been addressed. We investigate the stability of instance labels provided by several MIL classifiers on 5 different datasets, of which 3 are medical image datasets (breast histopathology, diabetic retinopathy and computed tomography lung images). We propose an unsupervised measure to evaluate instance stability, and demonstrate that a performance-stability trade-off can be made when comparing MIL classifiers.