DeepStroke: An Efficient Stroke Screening Framework for Emergency Rooms with Multimodal Adversarial Deep Learning

TL;DR

DeepStroke is a multimodal deep learning framework that rapidly assesses stroke presence in ER patients using facial videos and audio, outperforming traditional methods and clinicians in sensitivity and accuracy within six minutes.

Contribution

The paper introduces DeepStroke, a novel multimodal adversarial deep learning framework that enhances stroke screening accuracy and speed in emergency settings.

Findings

DeepStroke achieves 10.94% higher sensitivity than traditional triage.

It maintains 7.37% higher accuracy at similar specificity levels.

Assessment time is less than six minutes.

Abstract

In an emergency room (ER) setting, stroke triage or screening is a common challenge. A quick CT is usually done instead of MRI due to MRI's slow throughput and high cost. Clinical tests are commonly referred to during the process, but the misdiagnosis rate remains high. We propose a novel multimodal deep learning framework, DeepStroke, to achieve computer-aided stroke presence assessment by recognizing patterns of minor facial muscles incoordination and speech inability for patients with suspicion of stroke in an acute setting. Our proposed DeepStroke takes one-minute facial video data and audio data readily available during stroke triage for local facial paralysis detection and global speech disorder analysis. Transfer learning was adopted to reduce face-attribute biases and improve generalizability. We leverage a multi-modal lateral fusion to combine the low- and high-level features and provide mutual regularization for joint training. Novel adversarial training is introduced to obtain identity-free and stroke-discriminative features. Experiments on our video-audio dataset with actual ER patients show that DeepStroke outperforms state-of-the-art models and achieves better performance than both a triage team and ER doctors, attaining a 10.94% higher sensitivity and maintaining 7.37% higher accuracy than traditional stroke triage when specificity is aligned. Meanwhile, each assessment can be completed in less than six minutes, demonstrating the framework's great potential for clinical translation.