DRMC: A Generalist Model with Dynamic Routing for Multi-Center PET Image Synthesis

TL;DR

This paper introduces DRMC, a dynamic routing-based generalist model for multi-center PET image synthesis, effectively handling domain shifts and improving cross-center generalizability.

Contribution

The paper proposes a novel dynamic routing strategy with cross-layer connections to mitigate center interference in a shared architecture for multi-center PET image synthesis.

Findings

DRMC outperforms existing methods in cross-center generalization.

Dynamic routing effectively reduces center interference.

The model demonstrates strong robustness across diverse imaging centers.

Abstract

Multi-center positron emission tomography (PET) image synthesis aims at recovering low-dose PET images from multiple different centers. The generalizability of existing methods can still be suboptimal for a multi-center study due to domain shifts, which result from non-identical data distribution among centers with different imaging systems/protocols. While some approaches address domain shifts by training specialized models for each center, they are parameter inefficient and do not well exploit the shared knowledge across centers. To address this, we develop a generalist model that shares architecture and parameters across centers to utilize the shared knowledge. However, the generalist model can suffer from the center interference issue, \textit{i.e.} the gradient directions of different centers can be inconsistent or even opposite owing to the non-identical data distribution. To mitigate such interference, we introduce a novel dynamic routing strategy with cross-layer connections that routes data from different centers to different experts. Experiments show that our generalist model with dynamic routing (DRMC) exhibits excellent generalizability across centers. Code and data are available at: https://github.com/Yaziwel/Multi-Center-PET-Image-Synthesis.