Cancer Detection with Multiple Radiologists via Soft Multiple Instance Logistic Regression and $L_1$ Regularization

TL;DR

This paper introduces a soft multiple instance logistic regression model with L1 regularization for cancer detection, effectively utilizing multiple expert annotations to improve diagnostic probability estimation in digital images.

Contribution

It presents a novel approach that combines soft multiple instance learning with L1 regularization, reducing model complexity while leveraging multiple annotations for better cancer detection.

Findings

Achieves similar accuracy to existing models like RVMMIL.

Offers a more compact and practical model for clinical use.

Demonstrates effectiveness on real lung cancer detection data.

Abstract

This paper deals with the multiple annotation problem in medical application of cancer detection in digital images. The main assumption is that though images are labeled by many experts, the number of images read by the same expert is not large. Thus differing with the existing work on modeling each expert and ground truth simultaneously, the multi annotation information is used in a soft manner. The multiple labels from different experts are used to estimate the probability of the findings to be marked as malignant. The learning algorithm minimizes the Kullback Leibler (KL) divergence between the modeled probabilities and desired ones constraining the model to be compact. The probabilities are modeled by logit regression and multiple instance learning concept is used by us. Experiments on a real-life computer aided diagnosis (CAD) problem for CXR CAD lung cancer detection demonstrate that the proposed algorithm leads to similar results as learning with a binary RVMMIL classifier or a mixture of binary RVMMIL models per annotator. However, this model achieves a smaller complexity and is more preferable in practice.