Deformable MRI Sequence Registration for AI-based Prostate Cancer Diagnosis

TL;DR

This study evaluates how deformable MRI sequence registration impacts AI-based prostate cancer diagnosis, finding improved lesion alignment but only a modest, non-significant boost in diagnostic accuracy.

Contribution

It demonstrates that deformable registration improves lesion overlap but does not significantly enhance diagnostic performance, highlighting the need for integrated registration and AI development.

Findings

Deformable registration increased lesion overlap by 10%.

Rigid registration showed no diagnostic improvement.

Deformable registration led to a non-significant 0.3% AUROC increase.

Abstract

The PI-CAI (Prostate Imaging: Cancer AI) challenge led to expert-level diagnostic algorithms for clinically significant prostate cancer detection. The algorithms receive biparametric MRI scans as input, which consist of T2-weighted and diffusion-weighted scans. These scans can be misaligned due to multiple factors in the scanning process. Image registration can alleviate this issue by predicting the deformation between the sequences. We investigate the effect of image registration on the diagnostic performance of AI-based prostate cancer diagnosis. First, the image registration algorithm, developed in MeVisLab, is analyzed using a dataset with paired lesion annotations. Second, the effect on diagnosis is evaluated by comparing case-level cancer diagnosis performance between using the original dataset, rigidly aligned diffusion-weighted scans, or deformably aligned diffusion-weighted scans. Rigid registration showed no improvement. Deformable registration demonstrated a substantial improvement in lesion overlap (+10% median Dice score) and a positive yet non-significant improvement in diagnostic performance (+0.3% AUROC, p=0.18). Our investigation shows that a substantial improvement in lesion alignment does not directly lead to a significant improvement in diagnostic performance. Qualitative analysis indicated that jointly developing image registration methods and diagnostic AI algorithms could enhance diagnostic accuracy and patient outcomes.