ProMIL: Probabilistic Multiple Instance Learning for Medical Imaging

TL;DR

ProMIL introduces a probabilistic deep learning approach for medical imaging MIL that automatically determines the optimal positive instance percentage, outperforming standard models in real-world scenarios.

Contribution

ProMIL is a novel instance-based MIL method using Bernstein polynomial estimation that automatically identifies the optimal positive instance threshold.

Findings

ProMIL outperforms standard MIL models in medical imaging tasks.

It automatically detects the optimal positive instance percentage.

Code for ProMIL is publicly available.

Abstract

Multiple Instance Learning (MIL) is a weakly-supervised problem in which one label is assigned to the whole bag of instances. An important class of MIL models is instance-based, where we first classify instances and then aggregate those predictions to obtain a bag label. The most common MIL model is when we consider a bag as positive if at least one of its instances has a positive label. However, this reasoning does not hold in many real-life scenarios, where the positive bag label is often a consequence of a certain percentage of positive instances. To address this issue, we introduce a dedicated instance-based method called ProMIL, based on deep neural networks and Bernstein polynomial estimation. An important advantage of ProMIL is that it can automatically detect the optimal percentage level for decision-making. We show that ProMIL outperforms standard instance-based MIL in real-world medical applications. We make the code available.