Tongan Cai

Tongan Cai [CV][Google Scholar][LinkedIn]

Email (Reveal):

Summary: I'm an Applied Scientist at Samsara. I obtained my Ph.D. in Informatics from The Pennsylvania State University. I have extensive research experience in Computer Vision, Deep Learning, and Data Science, especially their application in the IoT, healthcare, and medical domains.

Work Experience

	Samsara, 2024/11 - Current Applied Scientist II. ML Science.
	Nationwide Mutual Insurance Company, 2023/5 - 2024/8 Data Scientist Intern. Enterprise Analytics.
	Tencent USA, 2021/5 - 2021/8 Research Scientist Intern. Medical AI Lab.

Research

My thesis research is on the DeepStroke project that aims to advance stroke emergency triage and patient self-assessment with mobile AI. I have lead several publications in major venues and participate in NIH proposal writing.
I have 7 years of academic and industrial research expertise in Computer Vision and Data Science, with their applications in healthcare, finances, and education.
Hands-on experience with state-of-the-art Multimodal Deep Learning, Medical AI, and Foundation Models. Previously, I collaborated on (1) vision-language understanding of placenta and (2) bodily expression of emotion projects.
I have strong passion about methods and techniques for data to “talk” themselves, for models to improve fairness and usability, and for AI products to make positive impacts to the society.

Selected Publications (Full Publication List)

	SafeTriage: Facial Video De-identification for Privacy-Preserving Stroke Triage Tongan Cai, HaomiaoNi, Wenchao Ma, Yuan Xue, Qian Ma, Rachel Leicht, Kelvin Wong, John Volpi and Stephen T.C. Wong, James Z. Wang, Xiaolei Huang Information Processing in Medical Imaging 2025 (IPMI'25)*, May 25-30, Kos Island, Greece. `[ Gen AI \| Video Retargeting \| Privacy ]`
	M³Stroke: Advancing Stroke Triage with MultiModal Mobile AI for Accurate and Timely Identification of Mild to Moderate Acute Strokes in Emergency Care Tongan Cai, Kelvin Wong, James Z. Wang, Xiaolei Huang, Xiaohui Yu, John Volpi and Stephen T.C. Wong IEEE-EMBS 2024 Conference of Biomedical and Health Informatics (BHI'24), Nov 11-13, Houston, TX, USA `[ Transformers \| State-Space Model \| Mobile ]` [Paper]
	Enhancing Automatic Placenta Analysis Through Distributional Feature Recomposition in Vision-Language Contrastive Learning Yimu Pan, Tongan Cai, Manas Mehta, Alison D Gernand, Jeffery A Goldstein, Leena Mithal, Delia Mwinyelle, Kelly Gallagher, James Z. Wang Proceedings of the International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2023). pp. 116-126. `[ Vision-Language Model \| Digital Pathology ]` [Paper]
	DeepStroke: An efficient stroke screening framework for emergency rooms with multimodal adversarial deep learning Tongan Cai, HaomiaoNi, Mingli Yu, Xiaolei Huang, Kelvin Wong, John Volpi, James Z. Wang, Stephen T.C. Wong Medical Image Analysis, Volume 80, August 2022, p.102522 (IF=13.8) `[ Video-Audio Fusion \| Adversarial \| Fairness Embedding ]` [Paper]
	What Makes A Good Reference Manager? A Quantitative Analysis of Bibliography Management Applications Tongan Cai, Chacha Chen, Kenneth Huang, and Frank E. Ritter Proceedings of the Asian CHI Symposium 2021 (Best Paper) `[ Human Factor Design \| Behavior \| Cognitive ]` [Paper]
	Toward Rapid Stroke Diagnosis with Multimodal Deep Learning Mingli Yu, Tongan Cai, Kelvin Wong, Xiaolei Huang, John Volpi, James Z. Wang and Stephen T.C. Wong Proceedings of the International Conference on Medical Image Computing and Computer Assisted Interventions. (MICCAI2020). pp. 616-626 `[ Multimodal \| Face Video \| Speech Text ]` [Paper][Code]

Misc

I have many genius friends and This Link will redirect you to one of them.

	SafeTriage: Facial Video De-identification for Privacy-Preserving Stroke Triage Tongan Cai, HaomiaoNi, Wenchao Ma, Yuan Xue, Qian Ma, Rachel Leicht, Kelvin Wong, John Volpi and Stephen T.C. Wong, James Z. Wang, Xiaolei Huang Information Processing in Medical Imaging 2025 (IPMI'25)*, May 25-30, Kos Island, Greece. `[ Gen AI \| Video Retargeting \| Privacy ]`
	M³Stroke: Advancing Stroke Triage with MultiModal Mobile AI for Accurate and Timely Identification of Mild to Moderate Acute Strokes in Emergency Care Tongan Cai, Kelvin Wong, James Z. Wang, Xiaolei Huang, Xiaohui Yu, John Volpi and Stephen T.C. Wong IEEE-EMBS 2024 Conference of Biomedical and Health Informatics (BHI'24), Nov 11-13, Houston, TX, USA `[ Transformers \| State-Space Model \| Mobile ]` [Paper]
	Enhancing Automatic Placenta Analysis Through Distributional Feature Recomposition in Vision-Language Contrastive Learning Yimu Pan, Tongan Cai, Manas Mehta, Alison D Gernand, Jeffery A Goldstein, Leena Mithal, Delia Mwinyelle, Kelly Gallagher, James Z. Wang Proceedings of the International Conference on Medical Image Computing and Computer Assisted Interventions (MICCAI 2023). pp. 116-126. `[ Vision-Language Model \| Digital Pathology ]` [Paper]
	DeepStroke: An efficient stroke screening framework for emergency rooms with multimodal adversarial deep learning Tongan Cai, HaomiaoNi, Mingli Yu, Xiaolei Huang, Kelvin Wong, John Volpi, James Z. Wang, Stephen T.C. Wong Medical Image Analysis, Volume 80, August 2022, p.102522 (IF=13.8) `[ Video-Audio Fusion \| Adversarial \| Fairness Embedding ]` [Paper]
	What Makes A Good Reference Manager? A Quantitative Analysis of Bibliography Management Applications Tongan Cai, Chacha Chen, Kenneth Huang, and Frank E. Ritter Proceedings of the Asian CHI Symposium 2021 (Best Paper) `[ Human Factor Design \| Behavior \| Cognitive ]` [Paper]
	Toward Rapid Stroke Diagnosis with Multimodal Deep Learning Mingli Yu, Tongan Cai, Kelvin Wong, Xiaolei Huang, John Volpi, James Z. Wang and Stephen T.C. Wong Proceedings of the International Conference on Medical Image Computing and Computer Assisted Interventions. (MICCAI2020). pp. 616-626 `[ Multimodal \| Face Video \| Speech Text ]` [Paper][Code]