Outcome prediction and individualized treatment effect estimation in patients with large vessel occlusion stroke

2025-07-03

AI Summary

This study addresses large-vessel occlusion (LVO) stroke, aiming to jointly model clinical functional outcome prediction and individualized treatment effect estimation. Method: We propose a novel, interpretable multimodal deep learning architecture that integrates non-contrast CT and CT angiography (CTA) with clinical variables, enabling simultaneous binary outcome classification and heterogeneous treatment effect estimation via the C-statistic for Benefit (C-for-Benefit). Results: Clinical variables alone achieve an AUC of 0.737 for outcome prediction; adding CTA yields only marginal improvement. While the individualized treatment benefit estimation is well-calibrated, its discriminative capacity remains limited (C-for-Benefit = 0.55). Our key contribution is the first unified multimodal framework jointly modeling imaging (non-contrast CT + CTA) and clinical data to produce concurrently interpretable predictions of both prognosis and treatment response—substantially enhancing clinical applicability and decision-support utility.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Mechanical thrombectomy has become the standard of care in patients with stroke due to large vessel occlusion (LVO). However, only 50% of successfully treated patients show a favorable outcome. We developed and evaluated interpretable deep learning models to predict functional outcomes in terms of the modified Rankin Scale score alongside individualized treatment effects (ITEs) using data of 449 LVO stroke patients from a randomized clinical trial. Besides clinical variables, we considered non-contrast CT (NCCT) and angiography (CTA) scans which were integrated using novel foundation models to make use of advanced imaging information. Clinical variables had a good predictive power for binary functional outcome prediction (AUC of 0.719 [0.666, 0.774]) which could slightly be improved when adding CTA imaging (AUC of 0.737 [0.687, 0.795]). Adding NCCT scans or a combination of NCCT and CTA scans to clinical features yielded no improvement. The most important clinical predictor for functional outcome was pre-stroke disability. While estimated ITEs were well calibrated to the average treatment effect, discriminatory ability was limited indicated by a C-for-Benefit statistic of around 0.55 in all models. In summary, the models allowed us to jointly integrate CT imaging and clinical features while achieving state-of-the-art prediction performance and ITE estimates. Yet, further research is needed to particularly improve ITE estimation.

Problem

• Predict functional outcomes in LVO stroke patients
• Estimate individualized treatment effects using deep learning
• Integrate clinical and imaging data for improved predictions

Innovation

• Interpretable deep learning models for stroke outcome prediction
• Novel foundation models integrating NCCT and CTA scans
• Combined clinical and imaging features for ITE estimation