Deep generative computed perfusion-deficit mapping of ischaemic stroke

2025-02-03

AI Summary

Precise early functional localization in acute ischemic stroke remains challenging due to the lack of reliable, annotation-free neuroanatomical mapping methods. Method: We propose a novel computational neuroanatomical mapping paradigm that requires no lesion segmentation. Leveraging ultra-acute-phase CTA images from 1,393 patients, we train a deep generative model to synthesize high-fidelity computed perfusion deficit maps directly from raw vascular imaging, and subsequently infer the neural substrates underlying individual NIHSS subitems. Contribution/Results: This is the first study to reconstruct canonical lesion–symptom mappings without lesion priors, revealing novel patterns of neural dependency. We demonstrate that CTA-derived perfusion maps exhibit exceptional anatomical fidelity and strong phenotype–genotype associations even in the ultra-acute phase. Our framework achieves high-accuracy anatomical localization for NIHSS sub-scores, providing an interpretable, generalizable computational basis for pre-intervention, individualized functional assessment.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Machine Learning: Neuro-Symbolic Learning
• Computer Vision: Segmentation

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Focal deficits in ischaemic stroke result from impaired perfusion downstream of a critical vascular occlusion. While parenchymal lesions are traditionally used to predict clinical deficits, the underlying pattern of disrupted perfusion provides information upstream of the lesion, potentially yielding earlier predictive and localizing signals. Such perfusion maps can be derived from routine CT angiography (CTA) widely deployed in clinical practice. Analysing computed perfusion maps from 1,393 CTA-imaged-patients with acute ischaemic stroke, we use deep generative inference to localise neural substrates of NIHSS sub-scores. We show that our approach replicates known lesion-deficit relations without knowledge of the lesion itself and reveals novel neural dependents. The high achieved anatomical fidelity suggests acute CTA-derived computed perfusion maps may be of substantial clinical-and-scientific value in rich phenotyping of acute stroke. Using only hyperacute imaging, deep generative inference could power highly expressive models of functional anatomical relations in ischaemic stroke within the pre-interventional window.

Problem

• Stroke Diagnosis
• Brain Vasculature Analysis
• Medical Imaging Technology

Innovation

• Advanced Computer Techniques
• Stroke Patient Brain Imaging
• Early Condition Assessment