Structured feature disentanglement for high-fidelity unified brain MRI synthesis

Multisequence Magnetic Resonance Imaging (MRI) provides a more reliable diagnosis in clinical applications through complementary information across sequences. However, in practice, the absence of certain MR sequences is a common problem that can lead to inconsistent analysis results. In this work, we propose a novel unified framework for synthesizing multisequence MR images, called hybrid-fusion GAN (HF-GAN). The fundamental mechanism of this work is structured feature disentanglement, which aligns the design of the architecture with the complexity of the features. A powerful many-to-one stream is constructed for the extraction of complex complementary features, while utilizing parallel, one-to-one streams to process modality-specific information. A channel attention-based fusion module (CAFF) dynamically integrates these disentangled features as a universal anatomical map into a common latent space. Then, a modality infuser transforms the features to generate the target sequence. We validated our framework on public datasets of both healthy and pathological brain MRI. Quantitative and qualitative results show that HF-GAN achieves state-of-the-art performance, with our 2D slice-based framework notably yielding highly competitive and efficient results compared to a leading 3D volumetric model. Furthermore, the utilization of HF-GAN for data imputation substantially improves the performance of the downstream brain tumor segmentation task, demonstrating its clinical relevance. Our source code is available on https://github.com/SSTDV-Project/HF-GAN.