Learning from Anatomy: Supervised Anatomical Pretraining (SAP) for Improved Metastatic Bone Disease Segmentation in Whole-Body MRI

TL;DR

This paper introduces a supervised anatomical pretraining method that leverages skeletal segmentation from healthy MRI scans to significantly improve metastatic bone disease segmentation in whole-body MRI, outperforming existing methods.

Contribution

The novel SAP approach uses anatomical labels for pretraining, enhancing lesion segmentation accuracy and detection sensitivity in metastatic bone disease.

Findings

SAP outperforms baseline and SSL models in segmentation metrics.

Achieves 100% detection sensitivity for lesions larger than 1 ml in most patients.

Provides publicly available code and models for reproducibility.

Abstract

The segmentation of metastatic bone disease (MBD) in whole-body MRI (WB-MRI) is a challenging problem. Due to varying appearances and anatomical locations of lesions, ambiguous boundaries, and severe class imbalance, obtaining reliable segmentations requires large, well-annotated datasets capturing lesion variability. Generating such datasets requires substantial time and expertise, and is prone to error. While self-supervised learning (SSL) can leverage large unlabeled datasets, learned generic representations often fail to capture the nuanced features needed for accurate lesion detection. In this work, we propose a Supervised Anatomical Pretraining (SAP) method that learns from a limited dataset of anatomical labels. First, an MRI-based skeletal segmentation model is developed and trained on WB-MRI scans from healthy individuals for high-quality skeletal delineation. Then, we compare its downstream efficacy in segmenting MBD on a cohort of 44 patients with metastatic prostate cancer, against both a baseline random initialization and a state-of-the-art SSL method. SAP significantly outperforms both the baseline and SSL-pretrained models, achieving a normalized surface Dice of 0.76 and a Dice coefficient of 0.64. The method achieved a lesion detection F2 score of 0.44, improving on 0.24 (baseline) and 0.31 (SSL). When considering only clinically relevant lesions larger than 1~ml, SAP achieves a detection sensitivity of 100% in 28 out of 32 patients. Learning bone morphology from anatomy yields an effective and domain-relevant inductive bias that can be leveraged for the downstream segmentation task of bone lesions. All code and models are made publicly available.