Keywords: Multiparametric prostate MRI, Multiple instance learning, Image classification prostate cancer
Abstract: Prostate lesion localization is a component of the prostate magnetic resonance imaging (MRI) routine evaluation. Localization is essential for targeted biopsy by enabling registration with real-time ultrasound. Most previous work on prostate cancer localization was focused on classification or segmentation assuming availability of radiological annotations. In this work we propose to use unsipervised attention-based multiple instance learning (MIL) method in an application for the prediction and localization of clinically significant prostate cancer. We train our model end-to-end with only image-level labels instead of relying on expensive pixel-level annotations. We extend MIL method by operating both on patches and the whole size image to learn local and global features which further improves classification and localization performance. To better leverage the relationships between multi-modal data we use an architecture with multiple encoding paths, where each path processing one image modality respectively. The model was developed on dataset containing 986 multiparametric prostate MRI and achieved 0.75 ±0.03 AUROC using 3-fold cross-validation in prostate cancer Grade Group classification.
Registration: I acknowledge that publication of this at MIDL and in the proceedings requires at least one of the authors to register and present the work during the conference.
Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.
Paper Type: validation/application paper
Primary Subject Area: Detection and Diagnosis
Secondary Subject Area: Segmentation
Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.