Wu, X;
Saeed, SU;
Wang, Y;
Coll, EB;
Hu, Y;
(2026)
Policy to Assist Iteratively Local Segmentation: Optimising Modality and Location Selection for Prostate Cancer Localisation.
In: Cui, Zhiming and Rekik, Islem and Suk, Heung-IL and Ouyang, Xi and Sun, Kaicong and Wang, Sheng, (eds.)
Machine Learning in Medical Imaging.
(pp. pp. 328-337).
Springer Nature: Cham, Switzerland.
Preview |
Text
MICCAI_WorkShop.pdf - Accepted Version Download (7MB) | Preview |
Abstract
Radiologists often mix medical image reading strategies, including inspection of individual modalities and local image regions, using information at different locations from different images independently as well as concurrently. In this paper, we propose a recommend system to assist machine learning-based segmentation models, by suggesting appropriate image portions along with the best modality, such that prostate cancer segmentation performance can be maximised. Our approach trains a policy network that assists tumor localisation, by recommending both the optimal imaging modality and the specific sections of interest for review. During training, a pre-trained segmentation network mimics radiologist inspection on individual or variable combinations of these imaging modalities and their sections - selected by the policy network. Taking the locally segmented regions as an input for the next step, this dynamic decision making process iterates until all cancers are best localised. We validate our method using a data set of 1325 labelled multiparametric MRI images from prostate cancer patients, demonstrating its potential to improve annotation efficiency and segmentation accuracy, especially when challenging pathology is present. Experimental results show that our approach can surpass standard segmentation networks. Perhaps more interestingly, our trained agent independently developed its own optimal strategy, which may or may not be consistent with current radiologist guidelines such as PI-RADS. This observation also suggests a promising interactive application, in which the proposed policy networks assist human radiologists. The code is open-sourced and available at here.
| Type: | Proceedings paper |
|---|---|
| Title: | Policy to Assist Iteratively Local Segmentation: Optimising Modality and Location Selection for Prostate Cancer Localisation |
| Event: | Machine Learning in Medical Imaging (MLMI 2025) |
| ISBN-13: | 9783032095121 |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1007/978-3-032-09513-8_32 |
| Publisher version: | https://doi.org/10.1007/978-3-032-09513-8_32 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. It has been made open access under the Creative Commons (CC BY) licence under the terms of the UCL Intellectual Property (IP) Policy and UCL Publications Policy. |
| Keywords: | Prostate Cancer, Reinforcement Learning, Image Segmentation |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10221328 |
Archive Staff Only
 |
View Item |
