Temporal Neural Cellular Automata: Application to modeling of contrast enhancement in breast MRI

2025-06-23

AI Summary

This study addresses the clinical challenge in breast MRI requiring intravenous gadolinium-based contrast agents and the inability of existing synthesis methods to preserve temporal consistency across dynamic contrast-enhanced (DCE) sequences. We propose a novel, contrast-free paradigm for synthesizing DCE-MRI images. To this end, we introduce TeNCA—a temporally extended neural cellular automaton specifically designed for modeling sparse, irregularly sampled time-series MRI data. TeNCA incorporates a physics-inspired temporal evolution mechanism and an adaptive spatiotemporal loss function to enable end-to-end simulation of dynamic enhancement kinetics. Evaluated on multicenter datasets, TeNCA significantly outperforms state-of-the-art methods, achieving superior structural fidelity, temporal coherence, and physiological plausibility in synthesized DCE sequences. Its robust performance demonstrates strong potential for direct clinical translation.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Synthetic contrast enhancement offers fast image acquisition and eliminates the need for intravenous injection of contrast agent. This is particularly beneficial for breast imaging, where long acquisition times and high cost are significantly limiting the applicability of magnetic resonance imaging (MRI) as a widespread screening modality. Recent studies have demonstrated the feasibility of synthetic contrast generation. However, current state-of-the-art (SOTA) methods lack sufficient measures for consistent temporal evolution. Neural cellular automata (NCA) offer a robust and lightweight architecture to model evolving patterns between neighboring cells or pixels. In this work we introduce TeNCA (Temporal Neural Cellular Automata), which extends and further refines NCAs to effectively model temporally sparse, non-uniformly sampled imaging data. To achieve this, we advance the training strategy by enabling adaptive loss computation and define the iterative nature of the method to resemble a physical progression in time. This conditions the model to learn a physiologically plausible evolution of contrast enhancement. We rigorously train and test TeNCA on a diverse breast MRI dataset and demonstrate its effectiveness, surpassing the performance of existing methods in generation of images that align with ground truth post-contrast sequences.

Problem

• Modeling temporal contrast enhancement in breast MRI
• Improving consistency in synthetic contrast generation
• Adapting NCA for sparse non-uniform imaging data

Innovation

• Temporal Neural Cellular Automata for MRI
• Adaptive loss computation strategy
• Physiologically plausible contrast evolution