Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Texture feature analysis for classification of early-stage prostate cancer in mpMRI

Muftah, Asmail, Shermer, S.M. and Langbein, Frank ORCID: https://orcid.org/0000-0002-3379-0323 2024. Texture feature analysis for classification of early-stage prostate cancer in mpMRI. Presented at: First International Conference on Artificial Intelligence in Healthcare, Swansea, UK, 4-6 September 2024. Published in: Xie, Xianghua, Styles, Iain, Powathil, Gibin and Ceccarelli, Marco eds. Artificial Intelligence in Healthcare. Lecture Notes in Computer Science , vol.14976 (14976) Springer, 10.1007/978-3-031-67285-9_9

[thumbnail of aiih2024.pdf]
Preview
PDF - Accepted Post-Print Version
Download (3MB) | Preview

Abstract

Magnetic resonance imaging (MRI) has become a crucial tool in the diagnosis and staging of prostate cancer, owing to its superior tis- sue 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 multi- ple 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 Ma- chines (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 tex- ture features, and local binary patterns to the classification. Using cor- relation 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.

Item Type: Conference or Workshop Item (Poster)
Date Type: Publication
Status: Published
Schools: Advanced Research Computing @ Cardiff (ARCCA)
Computer Science & Informatics
Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
Publisher: Springer
ISBN: 9783031672842
Related URLs:
Date of First Compliant Deposit: 1 July 2024
Date of Acceptance: 26 May 2024
Last Modified: 21 Aug 2024 15:02
URI: https://orca.cardiff.ac.uk/id/eprint/170191

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics