End-to-end Cortical Surface Reconstruction from Clinical Magnetic Resonance Images

2025-05-20

AI Summary

Current cortical surface reconstruction tools require high-resolution (≥1 mm) and specific-contrast (e.g., T1-weighted) MRI scans, limiting generalizability to heterogeneous clinical acquisitions. To address this, we propose the first end-to-end, retraining-free explicit cortical reconstruction method adaptable to arbitrary clinical MR sequences—including multi-contrast, anisotropic, and low-resolution data. Our approach leverages synthetically domain-randomized training data, integrates differentiable template mesh deformation, and employs a two-stage joint estimation of white-matter and gray-matter surfaces—ensuring strict topological correctness. Evaluated on ADNI and a real-world clinical cohort of 1,332 subjects, our method reduces cortical thickness estimation error by 50% compared to recon-all-clinical (0.50 → 0.24 mm). It achieves significantly improved accuracy, robustly recapitulates age-related cortical thinning patterns, and demonstrates both computational robustness and biological validity.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Computer Vision: Adversarial Attacks & Robustness
• Cognitive Modeling & Cognitive Systems: (Computational) Cognitive Architectures

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Surface-based cortical analysis is valuable for a variety of neuroimaging tasks, such as spatial normalization, parcellation, and gray matter (GM) thickness estimation. However, most tools for estimating cortical surfaces work exclusively on scans with at least 1 mm isotropic resolution and are tuned to a specific magnetic resonance (MR) contrast, often T1-weighted (T1w). This precludes application using most clinical MR scans, which are very heterogeneous in terms of contrast and resolution. Here, we use synthetic domain-randomized data to train the first neural network for explicit estimation of cortical surfaces from scans of any contrast and resolution, without retraining. Our method deforms a template mesh to the white matter (WM) surface, which guarantees topological correctness. This mesh is further deformed to estimate the GM surface. We compare our method to recon-all-clinical (RAC), an implicit surface reconstruction method which is currently the only other tool capable of processing heterogeneous clinical MR scans, on ADNI and a large clinical dataset (n=1,332). We show a approximately 50 % reduction in cortical thickness error (from 0.50 to 0.24 mm) with respect to RAC and better recovery of the aging-related cortical thinning patterns detected by FreeSurfer on high-resolution T1w scans. Our method enables fast and accurate surface reconstruction of clinical scans, allowing studies (1) with sample sizes far beyond what is feasible in a research setting, and (2) of clinical populations that are difficult to enroll in research studies. The code is publicly available at https://github.com/simnibs/brainnet.

Problem

• Reconstruct cortical surfaces from diverse clinical MR scans
• Overcome limitations of contrast and resolution variability
• Improve accuracy and speed for clinical neuroimaging studies

Innovation

• Uses synthetic domain-randomized data for training
• Deforms template mesh for topological correctness
• Processes heterogeneous clinical MR scans without retraining