Stroke outcome and evolution prediction from CT brain using a spatiotemporal diffusion autoencoder

2026-02-28

AI Summary

Accurately predicting lesion evolution and clinical outcomes in stroke patients is crucial for personalized treatment, yet remains challenging with current approaches. This work proposes the first spatiotemporal diffusion autoencoder, which leverages a diffusion probabilistic model to construct self-supervised semantic representations that effectively integrate multiphase CT scans with symptom onset time. The model enables unsupervised modeling of brain tissue fate without requiring labeled data. Evaluated on a large multicenter dataset comprising 3,573 patients and 5,824 CT images, the method significantly outperforms existing techniques, achieving state-of-the-art performance in predicting both next-day stroke severity and functional outcomes at discharge.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Diffusion Models for Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Stroke is a major cause of death and disability worldwide. Accurate outcome and evolution prediction has the potential to revolutionize stroke care by individualizing clinical decision-making leading to better outcomes. However, despite a plethora of attempts and the rich data provided by neuroimaging, modelling the ultimate fate of brain tissue remains a challenging task. In this work, we apply recent ideas in the field of diffusion probabilistic models to generate a self-supervised semantically meaningful stroke representation from Computed Tomography (CT) images. We then improve this representation by extending the method to accommodate longitudinal images and the time from stroke onset. The effectiveness of our approach is evaluated on a dataset consisting of 5,824 CT images from 3,573 patients across two medical centers with minimal labels. Comparative experiments show that our method achieves the best performance for predicting next-day severity and functional outcome at discharge.

Problem

• stroke outcome prediction
• stroke evolution
• CT brain imaging
• neuroimaging
• clinical decision-making

Innovation

• spatiotemporal diffusion autoencoder
• stroke outcome prediction
• self-supervised representation
• longitudinal CT imaging
• diffusion probabilistic models