Distance Map Supervised Landmark Localization for MR-TRUS Registration

TL;DR

This paper introduces a landmark-guided deep learning approach for MR-TRUS prostate image registration, using distance map regression for improved landmark localization and resulting in more accurate affine transformations than manual registration.

Contribution

It presents a novel distance map regression method for landmark localization that enhances registration accuracy in MR-TRUS prostate imaging.

Findings

Outperforms manual registration in TRE accuracy

Distance map regression improves landmark localization

Automated registration achieves significant accuracy gains

Abstract

In this work, we propose to explicitly use the landmarks of prostate to guide the MR-TRUS image registration. We first train a deep neural network to automatically localize a set of meaningful landmarks, and then directly generate the affine registration matrix from the location of these landmarks. For landmark localization, instead of directly training a network to predict the landmark coordinates, we propose to regress a full-resolution distance map of the landmark, which is demonstrated effective in avoiding statistical bias to unsatisfactory performance and thus improving performance. We then use the predicted landmarks to generate the affine transformation matrix, which outperforms the clinicians' manual rigid registration by a significant margin in terms of TRE.