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

arXiv: https://arxiv.org/abs/2109.12065 · HTML: https://mlsanity.com/paper/2109.12065/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 98 references that are in the map.
- First posted on arXiv 24 Sep 2021.

## Builds on (most cited first, 19 of 19)

- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/1511.06434/) (2015)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Least Squares Generative Adversarial Networks](https://mlsanity.com/paper/1611.04076/) (2016)
- [Network In Network](https://mlsanity.com/paper/1312.4400/) (2013)
- [Deep Speech 2: End-to-End Speech Recognition in English and Mandarin](https://mlsanity.com/paper/1512.02595/) (2015)
- [One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing](https://mlsanity.com/paper/2011.15126/) (2020)
- [Deep Speech: Scaling up end-to-end speech recognition](https://mlsanity.com/paper/1412.5567/) (2014)
- [Diverse Image-to-Image Translation via Disentangled Representations](https://mlsanity.com/paper/1808.00948/) (2018)
- [ECO: Efficient Convolution Operators for Tracking](https://mlsanity.com/paper/1611.09224/) (2016)
- [Audiovisual SlowFast Networks for Video Recognition](https://mlsanity.com/paper/2001.08740/) (2020)
- [Gait Recognition via Disentangled Representation Learning](https://mlsanity.com/paper/1904.04925/) (2019)
- [Multimodal Deep Learning for Robust RGB-D Object Recognition](https://mlsanity.com/paper/1507.06821/) (2015)
- [SegAN: Adversarial Network with Multi-scale L1 Loss for Medical Image Segmentation](https://mlsanity.com/paper/1706.01805/) (2017)
- [Unsupervised Domain-Specific Deblurring via Disentangled Representations](https://mlsanity.com/paper/1903.01594/) (2019)
- [Exploring Disentangled Feature Representation Beyond Face Identification](https://mlsanity.com/paper/1804.03487/) (2018)
- [Cascaded V-Net using ROI masks for brain tumor segmentation](https://mlsanity.com/paper/1812.11588/) (2018)
- [EmoNets: Multimodal deep learning approaches for emotion recognition in video](https://mlsanity.com/paper/1503.01800/) (2015)
- [Identity-Free Facial Expression Recognition using conditional Generative Adversarial Network](https://mlsanity.com/paper/1903.08051/) (2019)

## Built on by (most cited first, 7 of 7)

- [SafeTriage: Facial Video De-identification for Privacy-Preserving Stroke Triage](https://mlsanity.com/paper/2506.16578/) (2025)
- [DeepGraviLens: a Multi-Modal Architecture for Classifying Gravitational Lensing Data](https://mlsanity.com/paper/2205.00701/) (2022)
- [FAST-CAD: A Fairness-Aware Framework for Non-Contact Stroke Diagnosis](https://mlsanity.com/paper/2511.08887/) (2025)
- [Incorporating simulated spatial context information improves the effectiveness of contrastive learning models](https://mlsanity.com/paper/2401.15120/) (2024)
- [Digital FAST: An AI-Driven Multimodal Framework for Rapid and Early Stroke Screening](https://mlsanity.com/paper/2601.11896/) (2026)
- [StrokeNeXt: A Siamese-encoder Approach for Brain Stroke Classification in Computed Tomography Imagery](https://mlsanity.com/paper/2602.15087/) (2026)
- [StrokeGuard: A Multi-Agent Guided System for Prehospital Stroke Assessment](https://mlsanity.com/paper/2608.24555/) (2026)

## Abstract (arXiv)

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.