Individualized Mapping of Aberrant Cortical Thickness via Stochastic Cortical Self-Reconstruction

2024-03-11

AI Summary

Existing cortical thickness reference models suffer from site-specific biases and regional averaging, limiting their sensitivity to focal abnormalities. To address this, we propose the Stochastic Cortical Self-Reconstruction (SCSR) framework—the first unsupervised, subject-specific, vertex-wise deep generative model capable of high-resolution cortical thickness self-reconstruction from a single MRI scan, without requiring auxiliary subject metadata or cross-site harmonization. Trained on >25,000 healthy individuals, SCSR substantially reduces reconstruction error and, for the first time, detects previously missed focal cortical thinning in preterm infants. In neurodegenerative disease classification, it improves diagnostic accuracy and generates fine-grained cortical deviation maps for clinical dementia patients. By jointly overcoming limitations in multi-site generalizability and local sensitivity, SCSR establishes a new benchmark for individualized cortical morphometry.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Other Foundations of Computer Vision

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Understanding individual differences in cortical structure is key to advancing diagnostics in neurology and psychiatry. Reference models aid in detecting aberrant cortical thickness, yet site-specific biases limit their direct application to unseen data, and region-wise averages prevent the detection of localized cortical changes. To address these limitations, we developed the Stochastic Cortical Self-Reconstruction (SCSR), a novel method that leverages deep learning to reconstruct cortical thickness maps at the vertex level without needing additional subject information. Trained on over 25,000 healthy individuals, SCSR generates highly individualized cortical reconstructions that can detect subtle thickness deviations. Our evaluations on independent test sets demonstrated that SCSR achieved significantly lower reconstruction errors and identified atrophy patterns that enabled better disease discrimination than established methods. It also hints at cortical thinning in preterm infants that went undetected by existing models, showcasing its versatility. Finally, SCSR excelled in mapping highly resolved cortical deviations of dementia patients from clinical data, highlighting its potential for supporting diagnosis in clinical practice.

Problem

• Detecting localized cortical thickness deviations in individual brains
• Overcoming site-specific biases in cortical thickness reference models
• Identifying subtle cortical changes for improved disease discrimination

Innovation

• Deep learning reconstructs cortical thickness without subject information
• Generates individualized cortical maps from healthy training data
• Detects subtle thickness deviations for improved disease discrimination