Siyi Tang

siyitang.ai@gmail.com

I am an Machine Learning Scientist at ArteraAI, where my research focuses on developing multimodal deep learning models for predicting cancer patient outcomes and personalizing treatment strategies.

I obtained my PhD Degree from Stanford University, where I was advised by Prof. Daniel Rubin. At Stanford, I worked on developing deep learning methods for modeling medical time series data, with a focus on graph-based modeling approaches.

I co-organized the Stanford MedAI Group Exchange Sessions, a weekly seminar series where researchers around the world are invited to present the most recent advances in medical AI research. Check out the YouTube Channel!

I received my Bachelor's Degree in Electrical Engineering (Highest Distinction Honors) from National University of Singapore, where I was fortunate to be advised by Prof. Nitish Thakor and Prof. Thomas Yeo.

Google Scholar  /  LinkedIn  /  Twitter  /  GitHub

News

• 2024-02: Our paper on trustworthy seizure onset detection is now published on npj Digital Medicine!

• 2023-06: Our paper on a predictive AI biomarker of androgen deprivation therapy benefit in prostate cancer is now published on NEJM Evidence!

• 2023-06: Our paper on modeling multivariate biosignals with GNNs and S4 won the Best Paper Award at CHIL 2023!

• 2023-04: Our paper on modeling multivariate biosignals with GNNs and S4 have been accepted as oral presentations at ICLR 2023 TSRL4H Workshop and CHIL 2023!

• 2023-04: I received my PhD Degree from Stanford University and will be joining Artera as an Machine Learning Scientist in April 2023!

• 2023-02: I defended my PhD dissertation!

• 2023-01: Our paper on multimodal graph neural networks for hospital readmission prediction is now published online on IEEE Journal of Biomedical and Health Informatics!

• 2022-07: Our paper on multimodal fusion for atrial fibrillation ablation outcome prediction is now published online on Circulation: Arrythmia and Electrophysiology!

• 2022-04: I presented our work on deep-learning-based multimodal fusion for atrial fibrillation ablation outcome prediction at Heart Rhythm 2022. The abstract has received the Highest Scoring Abstract in Digital Health Award!

• 2022-01: Our paper on self-supervised graph neural network for EEG seizure analysis has been accepted to ICLR 2022!

• 2021-12: I obtained my MS Degree!

• 2021-04: Our paper on data valuation for chest X-rays is now out on Scientific Reports.

• 2021-02: I will be joining the medical AI team at Salesforce Research for a summer internship!

• 2020-12: I presented our work on transfer and meta-learning for EEG analysis at AES 2020.

• 2020-06: Our paper on Autism Spectrum Disorder subtyping with a Bayesian model is now out on Biological Psychiatry.

• 2020-01: I passed my PhD qualifying exam and advanced to PhD candidacy!

Selected Publications

Towards trustworthy seizure onset detection using workflow notes

Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer

Modeling Multivariate Biosignals With Graph Neural Networks and Structured State Space Models (Best Paper Award)

Predicting 30-day all-cause hospital readmission using multimodal spatiotemporal graph neural networks

Machine Learning–Enabled Multimodal Fusion of Intra-Atrial and Body Surface Signals in Prediction of Atrial Fibrillation Ablation Outcomes

Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis

Data Valuation for Medical Imaging Using Shapley Value and Application to a Large-Scale Chest X-Ray Dataset

Comparison of Segmentation-Free and Segmentation-Dependent Computer-Aided Diagnosis of Breast Masses on a Public Mammography Dataset

Reconciling dimensional and categorical models of autism heterogeneity: a brain connectomics and behavioral study

Somatosensory-motor dysconnectivity spans multiple transdiagnostic dimensions of psychopathology

Conference Presentations

Machine Learning-Enabled Multimodal Fusion of Intra-Atrial and Body Surface Signals in Prediction of Atrial Fibrillation Ablation Outcomes (Highest Scoring Abstract in Digital Health)

Self-Supervised Graph Neural Networks for Improved Electroencephalographic Seizure Analysis

From Adults to Neonates: Transfer and Meta-learning Approaches for Knowledge Generalization in Deep Networks for Electroencephalographic Analysis

Latent ASD Factors with Dissociable Functional Connectivity Patterns and Behavioral Symptoms

Live demonstration: Real-time orientation estimation and grasping of household objects for upper limb prostheses with a dynamic vision sensor (Honorable Mention Award)

Teaching

• Fall 2021: Teaching assistant of CS 330 Deep Multi-task and Meta Learning
• Spring 2022: Teaching assistant of BIOMEDIN 260 Computational Methods for Biomedical Image Analysis and Interpretation

About Me

My Chinese name is 汤斯怡. I grew up in a beautiful city, Zhang Zhou, in Fujian Province of China. I lived in Singapore for 7+ years before coming to Stanford.

Besides research, I also enjoy photography, cooking and baking, violin and traveling around the world. You can find some of my photography works here.

Design and source code from Jon Barron's and Leonid Keselman's websites