Daoyi Zhu

Daoyi Zhu

Ph.D. Student in AI for Health · Washington University in St. Louis

I am a third-year Ph.D. student in Computer Science at Washington University in St. Louis, advised by Dr. Chenyang Lu. My research focuses on clinical AI chatbots and the safety of large language models in healthcare, real-time perioperative risk prediction, and EHR-based clinical machine learning.

daoyi@wustl.edu (314) 614-6385 oscarzhu142857.github.io

Daoyi Zhu

Phone: (314) 614-6385  |  Email: daoyi@wustl.edu  |  Website: oscarzhu142857.github.io

Education

Washington University in St. Louis St. Louis, MO
Ph.D. in Computer Science (AI and Internet of Things for Healthcare) — Advisor: Dr. Chenyang Lu Expected 2028
  • Currently a third-year doctoral student advised by Dr. Chenyang Lu. Research focus: clinical AI chatbots and large language model safety in healthcare, real-time perioperative and intra-operative risk prediction, oncology EHR modeling, and multimodal clinical machine learning.
Washington University in St. Louis St. Louis, MO
B.S. in Computer Science — GPA: 3.96/4.00 — Honors: Dean's List, Dean Select Ph.D. Fellowship Aug 2020 – May 2023

Research Interests

Publications & Manuscripts

Adam Calderon, Daoyi Zhu, Nur Hani Zainal, Chenyang Lu, Ellen E. Fitzsimmons-Craft, Denise E. Wilfley, Daniel Eisenberg, Barr Taylor, Michelle G. Newman. “Baseline prediction of two-year prevention and remission of anxiety, depression, and eating disorders in college students receiving a digital guided self-help intervention: derivation and external validation of a machine learning algorithm across 26 population-based cohorts.” Submitted to Psychotherapy and Psychosomatics, 2026. Under major revision.
Daoyi Zhu, Bing Xue, Chenyang Lu, Joanna Abraham. “A Real-Time Machine Learning Framework for Improved Intra-operative Risk Predictions in Cardiac Surgery Patients.” Submitted to Journal of the American Medical Informatics Association, 2026. Under major revision.
Daoyi Zhu, Hanyang Liu, Bing Xue, Chenyang Lu, Mark Williams, Patricia Litkowski. “Predicting Discharge Readiness in Hospitalized Patients with Cancer Using a Machine Learning-Derived Clinical Stability Score.” 2026.
Daoyi Zhu, Bing Xue, Neel Shah, Philip Richard Orrin Payne, Chenyang Lu, Ahmed Sameh Said. “Multi-modal prediction of extracorporeal support — a resource-intensive therapy, utilizing a large national database.” 2026.

Research Experience

WashU Cyber Physical System Laboratory St. Louis, MO
Machine Learning Research Assistant Sep 2021 – Present
  • Developed and maintained machine learning pipelines for clinical risk prediction and ML-assisted handoff support, including feature updates, model reliability checks, SHAP-based explanations, and clinician-facing interpretation materials.
  • Designed a real-time intra-operative risk prediction framework that integrates preoperative variables with dense intra-operative time-series and medication data for cardiac surgery patients; prepared the manuscript and major-revision responses for the JAMIA submission.
  • Built EHR-based machine learning models for oncology discharge readiness using a clinical stability score framework; contributed to cohort construction, feature engineering, validation, calibration and evaluation, and manuscript preparation.
  • Contributed to a 26-cohort mental health prediction study for two-year prevention and remission of anxiety, depression, and eating disorders; supported semi-supervised pseudo-labeling design for missing outcomes and external validation reporting.
  • Conducted literature review and system design for clinical chatbot safety, including benchmark construction, clinician labeling workflows, retrieval-augmented generation baselines, and guardrail-oriented safety evaluation.
  • Developed pediatric COVID-19 early-warning models using large-scale NIH clinical data; built PySpark data pipelines, compared ML baselines, and applied SHAP analysis for clinical interpretability.

Professional Experience

Got Sure Remote
Software Development Engineer Intern Jun 2022 – Aug 2022
  • Built a customer-facing SaaS insurance website using React, React Router, and Ant Design; refactored class components into React Hooks and improved page navigation and maintainability.
  • Designed and formalized a MySQL database for insurance order tracking and deployed application components on AWS EC2.
Darwin et al Remote
Software Engineer Intern Nov 2021 – Jan 2022
  • Developed React and Ionic web application features for medical researchers, including discussion posts, reusable profile and project components, project search, and autocomplete functionality.
Bank of China International Wuhan, China
Summer Intern Jun 2020 – Jul 2020
  • Collected and analyzed historical Chinese A-share healthcare-sector stock data; evaluated a quantitative trading strategy using Python and reported findings in monthly group meetings.

Skills

Programming Languages: Python, R, SQL, Java, C/C++, JavaScript, HTML/CSS

Machine Learning & Statistics: scikit-learn, PyTorch, TensorFlow, XGBoost, LightGBM, CatBoost, SHAP, calibration analysis, decision-curve analysis

Data & Clinical Informatics: pandas, NumPy, Matplotlib, PySpark, EHR data processing, EPIC-derived data, large-scale clinical data pipelines

LLM/NLP & Software: RAG, prompt engineering, local LLM deployment, Flask, React, Git, Docker, Linux, AWS, MySQL, PostgreSQL

Awards & Honors