When Brains Disagree: Biological Ambiguity Underlies the Challenge of Amyloid PET Synthesis from Structural MRI

2026-05-12

AI Summary

This study addresses the instability in synthesizing amyloid PET images from structural MRI, which stems from the biological temporal decoupling between neurodegeneration and amyloid pathology, resulting in an ill-posed one-to-many mapping. The work provides the first scientific validation that this intrinsic ill-posedness arises from such biological ambiguity. Rather than relying on complex model architectures, the authors propose mitigating this ambiguity by controlling the training data distribution and integrating orthogonal information—specifically plasma biomarkers—through multimodal fusion. Experimental results demonstrate that models can learn effectively within non-ambiguous subsets of the data, and that incorporating plasma biomarkers significantly improves synthesis performance and restores stability. These findings confirm that multimodal biomarkers are essential for alleviating mapping ambiguity and enhancing the accuracy of cross-modality image synthesis.

Abstract

Structural MRI-to-amyloid PET synthesis has been proposed as a non-invasive alternative for amyloid assessment in Alzheimer's disease (AD). However, reported performance of identical models varies widely across studies, and increasingly complex architectures have not led to consistent gains. This inconsistency is thought to be caused by a fundamental biological ambiguity: MRI captures neurodegeneration, while PET measures amyloid pathology - two processes that are often temporally decoupled in AD. As a result, similar MRI patterns may correspond to different amyloid states, creating ambiguous one-to-many mappings. MRI-to-amyloid PET synthesis may therefore be intrinsically ill-posed; however, this idea has yet to be tested scientifically. The aim of this work is to test this hypothesis through two controlled experiments. We first control the training distribution by stratifying paired MRI-PET data by amyloid and neurodegeneration status. Using two standard synthesis models under a controlled design, we show that biologically unambiguous mappings are learnable in isolation, but performance collapses when data ambiguity is introduced. This demonstrates that ambiguity in the data distribution, rather than architectural capacity, constrains performance. Second, we show that introducing orthogonal biological information in the form of plasma biomarkers resolves this ambiguity. When multimodal inputs are incorporated, performance improves and stability is restored. Together, these findings suggest that limited and inconsistent performance in MRI-to-amyloid PET synthesis is explained by intrinsic biological ambiguity, and that stable, meaningful progress requires multimodal integration rather than architectural complexity.

Problem

• biological ambiguity
• MRI-to-PET synthesis
• Alzheimer's disease
• amyloid pathology
• neurodegeneration

Innovation

• biological ambiguity
• MRI-to-PET synthesis
• multimodal integration
• Alzheimer's disease
• plasma biomarkers