Asymmetric Cascade Networks for Focal Bone Lesion Prediction in Multiple Myeloma

TL;DR

This paper introduces an asymmetric cascade neural network architecture that predicts future bone lesions in multiple myeloma patients using longitudinal MRI data, aiding early detection and risk assessment.

Contribution

It proposes a novel asymmetric cascade U-Net architecture for longitudinal bone lesion prediction in multiple myeloma, combining detection and prediction stages.

Findings

Achieves accurate lesion prediction on a dataset of 63 patients.

Provides detailed volumetric risk maps for early lesion detection.

Demonstrates potential for improved patient outcome prediction.

Abstract

The reliable and timely stratification of bone lesion evolution risk in smoldering Multiple Myeloma plays an important role in identifying prime markers of the disease's advance and in improving the patients' outcome. In this work we provide an asymmetric cascade network for the longitudinal prediction of future bone lesions for T1 weighted whole body MR images. The proposed cascaded architecture, consisting of two distinct configured U-Nets, first detects the bone regions and subsequently predicts lesions within bones in a patch based way. The algorithm provides a full volumetric risk score map for the identification of early signatures of emerging lesions and for visualising high risk locations. The prediction accuracy is evaluated on a longitudinal dataset of 63 multiple myeloma patients.