Texture Feature Analysis for Classification of Early-Stage Prostate Cancer in mpMRI

TL;DR

This study evaluates machine learning models like RF and SVM for prostate cancer classification in mpMRI, highlighting the importance of specific texture features and their interpretability for early-stage detection.

Contribution

The paper identifies key texture features influencing prostate cancer classification and emphasizes the need for explainable AI in medical imaging.

Findings

Many features are strongly correlated, reducing the effective feature set.

Most features have negligible impact on classification outcomes.

A small subset of features primarily determines the classification.

Abstract

Magnetic resonance imaging (MRI) has become a crucial tool in the diagnosis and staging of prostate cancer, owing to its superior tissue contrast. However, it also creates large volumes of data that must be assessed by trained experts, a time-consuming and laborious task. This has prompted the development of machine learning tools for the automation of Prostate cancer (PCa) risk classification based on multiple MRI modalities (T2W, ADC, and high-b-value DWI). Understanding and interpreting the predictions made by the models, however, remains a challenge. We analyze Random Forests (RF) and Support Vector Machines (SVM), for two complementary datasets, the public Prostate-X dataset, and an in-house, mostly early-stage PCa dataset to elucidate the contributions made by first-order statistical features, Haralick texture features, and local binary patterns to the classification. Using correlation analysis and Shapley impact scores, we find that many of the features typically used are strongly correlated, and that the majority of features have negligible impact on the classification. We identify a small set of features that determine the classification outcome, which may aid the development of explainable AI approaches.