Trustworthy Longitudinal Brain MRI Completion: A Deformation-Based Approach with KAN-Enhanced Diffusion Model

2026-01-14

AI Summary

Longitudinal brain MRI studies often suffer from high attrition rates, leading to substantial missing data. Existing generative models rely heavily on image intensity information, resulting in anatomically implausible outputs and limited generalization across imaging modalities. To address these limitations, this work proposes DF-DiffCom, a diffusion model that integrates deformation field guidance with Kolmogorov–Arnold Networks (KANs). By constraining the generative process with deformation fields and leveraging KANs to enhance nonlinear modeling capacity, the method significantly improves anatomical fidelity and achieves modality-agnostic generalization. Experiments on the OASIS-3 dataset demonstrate that DF-DiffCom yields a 5.6% improvement in PSNR and a 0.12 gain in SSIM over existing approaches, and it effectively extends to diverse MRI modalities and auxiliary tasks such as brain tissue segmentation map synthesis.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Longitudinal brain MRI is essential for lifespan study, yet high attrition rates often lead to missing data, complicating analysis. Deep generative models have been explored, but most rely solely on image intensity, leading to two key limitations: 1) the fidelity or trustworthiness of the generated brain images are limited, making downstream studies questionable; 2) the usage flexibility is restricted due to fixed guidance rooted in the model structure, restricting full ability to versatile application scenarios. To address these challenges, we introduce DF-DiffCom, a Kolmogorov-Arnold Networks (KAN)-enhanced diffusion model that smartly leverages deformation fields for trustworthy longitudinal brain image completion. Trained on OASIS-3, DF-DiffCom outperforms state-of-the-art methods, improving PSNR by 5.6% and SSIM by 0.12. More importantly, its modality-agnostic nature allows smooth extension to varied MRI modalities, even to attribute maps such as brain tissue segmentation results.

Problem

• longitudinal brain MRI
• missing data
• image fidelity
• trustworthiness
• flexible guidance

Innovation

• deformation-based
• KAN-enhanced diffusion model
• longitudinal brain MRI completion
• trustworthy generation
• modality-agnostic