CABLD: Contrast-Agnostic Brain Landmark Detection with Consistency-Based Regularization

TL;DR

CABLD is a self-supervised deep learning framework that accurately detects 3D brain landmarks in unlabeled MRI scans across various contrasts, reducing annotation needs and enhancing generalization.

Contribution

It introduces a contrast-agnostic, self-supervised method using a single reference example and consistency regularization for robust brain landmark detection.

Findings

Outperforms state-of-the-art methods in MRE and SDR metrics.

Effective across diverse MRI contrasts and field strengths.

Reduces reliance on extensive annotated datasets.

Abstract

Anatomical landmark detection in medical images is essential for various clinical and research applications, including disease diagnosis and surgical planning. However, manual landmark annotation is time-consuming and requires significant expertise. Existing deep learning (DL) methods often require large amounts of well-annotated data, which are costly to acquire. In this paper, we introduce CABLD, a novel self-supervised DL framework for 3D brain landmark detection in unlabeled scans with varying contrasts by using only a single reference example. To achieve this, we employed an inter-subject landmark consistency loss with an image registration loss while introducing a 3D convolution-based contrast augmentation strategy to promote model generalization to new contrasts. Additionally, we utilize an adaptive mixed loss function to schedule the contributions of different sub-tasks for optimal outcomes. We demonstrate the proposed method with the intricate task of MRI-based 3D brain landmark detection. With comprehensive experiments on four diverse clinical and public datasets, including both T1w and T2w MRI scans at different MRI field strengths, we demonstrate that CABLD outperforms the state-of-the-art methods in terms of mean radial errors (MREs) and success detection rates (SDRs). Our framework provides a robust and accurate solution for anatomical landmark detection, reducing the need for extensively annotated datasets and generalizing well across different imaging contrasts. Our code is publicly available at https://github.com/HealthX-Lab/CABLD.