Generative Modeling of Complex-Valued Brain MRI Data

2026-04-16

AI Summary

This work addresses the common oversight in MRI reconstruction methods, which typically discard phase data—rich in tissue pathology information—and rely solely on magnitude images. The study proposes a novel generative framework that jointly models both magnitude and phase components of complex-valued brain MRI. By employing a conditional variational autoencoder to compress data while preserving phase consistency, followed by flow matching to generate high-fidelity synthetic samples, the method achieves exceptional phase coherence (>0.997) and realistic image generation (AUROC 0.50–0.66). Notably, an anomaly detection classifier trained exclusively on synthetic data attains an AUROC of 0.880 on public benchmarks, surpassing the baseline trained on real data (AUROC 0.842), and demonstrates strong generalization on external test sets.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Objective. Standard Magnetic Resonance Imaging (MRI) reconstruction pipelines discard phase information captured during acquisition, despite evidence that it encodes tissue properties relevant to tumor diagnosis. Current machine learning approaches inherit this limitation by operating exclusively on reconstructed magnitude images. The aim of this study is to build a generative framework which is capable of jointly modeling magnitude and phase information of complex-valued MRI scans. Approach. The proposed generative framework combines a conditional variational autoencoder, which compresses complex-valued MRI scans into compact latent representations while preserving phase coherence, with a flow-matching-based generative model. Synthetic sample quality is assessed via a real-versus-synthetic classifier and by training downstream classifiers on synthetic data for abnormal tissue detection. Main results. The autoencoder preserves phase coherence above 0.997. Real-versus-synthetic classification yields low AUROC values between 0.50 and 0.66 across all acquisition sequences, indicating generated samples are nearly indistinguishable from real data. In downstream normal-versus-abnormal classification, classifiers trained entirely on synthetic data achieve an AUROC of 0.880, surpassing the real-data baseline of 0.842 on a publicly available dataset (fastMRI). This advantage persists on an independent external test set from a different institution with biopsy-confirmed labels. Significance. The proposed framework demonstrates the feasibility of jointly modeling magnitude and phase information for normal and abnormal complex-valued brain MRI data. Beyond synthetic data generation, it establishes a foundation for the usage of complete brain MRI information in future diagnostic applications and enables systematic investigation of how magnitude and phase jointly encode pathology-specific features.

Problem

• complex-valued MRI
• phase information
• generative modeling
• brain tumor diagnosis
• magnitude-phase joint modeling

Innovation

• complex-valued MRI
• phase coherence
• generative modeling
• flow matching
• synthetic data