Junjie Luo
Ph.D. candidate @Johns Hopkins University, School of Medicine
I am a Ph.D. Candidate in Biomedical Informatics and Data Science at the School of Medicine, Johns Hopkins University. My research focuses on designing and developing healthcare AI using real-world data, with special emphasis on human factors and behaviors to enhance health and wellness.
My research program centers on large foundation models for healthcare, particularly continuous glucose monitoring (CGM) for diabetes management:
- Research & Development: I develop novel AI systems tailored to real-world health problems, including large sensor models, agentic AI for patient messaging, and multi-modal sensor data modeling.
- Implementation & Evaluation: I deploy these models in production environments and evaluate them for both performance and human values such as fairness and robustness.
My research methodology distinctively combines the full AI lifecycle from development to deployment with strong industry collaborations (WellDoc, DrFirst, MedStar Health). My passion is conducting research that delivers tangible benefits to both patients and healthcare systems.
Education
- 2022-2026 (expected)PhD in Biomedical Informatics and Data Science
Johns Hopkins School of Medicine - 2020-2022PhD in Information Systems (transferred)
University of Maryland, College Park - 2018-2020M.S. in Data Science
The Chinese University of Hong Kong, Shenzhen - 2014-2018B.B.A. in Applied Economics
The Chinese University of Hong Kong, Shenzhen
Research Interests
- TopicsHealthcare AI, Foundation Models for Sensor Data, Continuous Glucose Monitoring, Agentic AI, Human-Centered AI, Health Disparities
- MethodologiesFoundation Models, Natural Language Processing, Multi-Modal Sensor Modeling, Causal Inference, MLOps
Papers
- Working Papers
-
Working PaperFrom Testing to Learning: Distilling Experimental Data for Patient Messaging Design Using Agentic AIWorking Paper, targeting Management Science
-
Working PaperA Time-Event Foundation Model Integrating Sensors and Behaviors for Diabetes ManagementWorking Paper
-
Working PaperMasked Excellence: Rating Disparity Reversal Reveals Physician Evaluation BiasWorking Paper
-
Working PaperPhysician Personality and Behavior: How LLM-Revealed Traits Shape Opioid OverprescriptionsWorking Paper
-
Working PaperFairGlucose: A Benchmark for Fair and Robust Forecasting of Continuous Glucose Monitoring Time-Series DataWorking Paper
- Under Review
-
Under ReviewGroup-Sensitive Offline Contextual BanditUnder Review at AISTATS 2026
-
Under ReviewMapping Patient-Perceived Physician Traits from Nationwide Online Reviews with LLMsUnder Review at NPJ Digital Medicine
- Publications
-
PublishedRisk Stratification at Prediabetes Onset and Association with Diabetes Outcomes Using EHR DataNPJ Metabolic Health and Disease (2025)
-
PublishedPAME-AI: Patient Messaging Creation and Optimization using Agentic AINeurIPS 2025 Workshop on GenAI for Health
-
PublishedA large sensor foundation model pretrained on continuous glucose monitor data for diabetes managementnpj Health Systems, 2(1):35 (2025)
-
PublishedImpact of Food on A Transformer Based Glucose Prediction Model to Predict Glucose Trajectories at Different Time HorizonsMachine Learning for Healthcare (2025)
-
PublishedCGM-GPT: A Transformer Based Glucose Prediction Model to Predict Glucose Trajectories at Different Time HorizonsMachine Learning for Healthcare (2024)
-
PublishedRecency in Online Physician RatingsJAMA Internal Medicine, 182(8):881-883 (2022)
-
PublishedQuantifying the Effect of Public Activity Intervention Policies on COVID-19 Pandemic Containment Using Epidemiologic Data From 145 CountriesValue in Health, 25(5):699-708 (2022)
-
PublishedSocioeconomic privilege and political ideology are associated with racial disparity in COVID-19 vaccinationPNAS 118, no. 33 (2021)
-
PublishedField Embedding: A Unified Grain Based Framework for Word RepresentationNAACL, 1754-1762 (2021)
Awards
Grants
- 2025 Samsung Research America (SRA) START Program Grant - $150,000 CGM- and Behavior-based Large Health Model for Just-in-Time Diabetes Management PI: Guodong (Gordon) Gao; Co-PIs: Brian Hasselfeld, Nestoras Mathioudakis; Research Lead: Junjie Luo
- 2023-24 Hopkins Business of Health Initiative (HBHI) Grant - $25,000 A Double-Edged Sword? Examining the Relationship Between Patient Satisfaction and Physician Overprescription PI: Gordon Gao; Co-PI: Ritu Agarwal; Research Team Member: Junjie Luo
Patents
- 2025 Systems and methods for metabolic outcome predictions Junjie Luo, Abhimanyu Kumbara, Anand Iyer, Mansur Shomali, Guodong (Gordon) Gao US Patent 12,293,841
- 2025 Systems and methods for continuous glucose monitoring outcome predictions Anand Iyer, Abhimanyu Kumbara, Mansur Shomali, Junjie Luo, Guodong (Gordon) Gao US Patent 12,283,381
- 2023 Systems and methods for score based clusters Anand Iyer, Malinda Peeples, Mansur Shomali, Abhimanyu Kumbara, Michelle Dugas, Guodong (Gordon) Gao, Junjie Luo US Patent App. 18/052,094
Honors and Awards
- 2020 R.H.Smith School Graduate Assistant Scholarship
- 2020 Academic Performance Scholarship
-
2018
Admission Scholarship for Masters Program
The Chinese University of Hong Kong, Shenzhen - 2015-18 Dean's List Honor
- 2016-17 Academic Performance Scholarship
- 2016-17 Undergraduate Research Award
-
2014
Admission Scholarship for Undergraduate Program
The Chinese University of Hong Kong, Shenzhen
Teaching
Johns Hopkins University
-
2025 Spring
AI Essentials for Business, Carey Business School
Lecturer, Content Co-designer, 47 students, 4.51/5.00 -
2024 Fall
AI Essentials for Business, Carey Business School
AI Tech Lead -
2024 Fall
Clinical Data Analysis with Python, School of Medicine
Teaching Assistant -
2024 Spring
Big Data Machine Learning, Carey Business School
AI Tech Lead
University of Maryland, College Park
-
2021 Summer
AI Leadership in Healthcare Exec Ed program, R.H.Smith School of Business
Head of AI Specialists
The Chinese University of Hong Kong, Shenzhen
-
2017 Spring
DMS2051: IT in Business Application, School of Management and Economics
Teaching Assistant
Conference Presentations
-
2024.9
Mansur Shomali, Junjie Luo, Abhimanyu Kumbara, Anand Iyer, Guodong (Gordon) Gao. Evaluating a state-of-the-art generative pre-trained transformer model to predict continuous glucose monitoring values at different time intervals.
DIABETOLOGIA'2024 -
2024.5
Junjie Luo, Abhimanyu Kumbara, Anand Lyer, Mansur Shomali and Guodong (Gordon) Gao. Using an Automated, Real-time Data Enabled Feature Engineering Process to Predict Future Weight Outcomes.
Conference on Health IT and Analytics'2024 -
2023.7
Junjie Luo, Gordon Gao, Ritu Agarwal. Assessing Dataset Quality to AI Bias and Healthcare Disparity: A Study on EHR-Based Prediabetes to Diabetes Risk Prediction Models.
INFORMS Healthcare'2023 -
2023.7
Junjie Luo, Mansur Shomali, Abhimanyu Kumbara, Anand Lyer, and Guodong (Gordon) Gao. Using Early Engagement Data from a Digital Health to Predict Future Health Outcomes.
INFORMS Healthcare'2023 -
2023.5
Junjie Luo, Aishwarya Shukla, Jie Mein Goh, Guodong (Gordon) Gao, Ritu Agarawl. Relationships Between Patient Satisfaction and Physician Overprescription: Evidence from Online Reviews and CMS Claim Data.
Production and Operations Management Society'2023 -
2023.5
Junjie Luo, Abhimanyu Kumbara, Anand Lyer, Mansur Shomali, and Guodong (Gordon) Gao. Using Early Engagement and CGM Data from a Digital Health Solution to Predict Future Engagement Patterns.
Conference on Health IT and Analytics'2023 -
2022.11
Junjie Luo, Casey Kindall, Sara-Laure Faraji, Guodong (Gordon) Gao, Ritu Agarawl, Sally Simpson. Foretelling the Physician's Future: Healthcare Fraud Prediction with Behavioral Big Data.
American Society of Criminology Annual Meeting'2022 -
2022.6
Junjie Luo, Guodong (Gordon) Gao, Ritu Agarwal, and Sally Simpson. "Minority Report" Comes True: Using Behavioral Big Data to Predict Physician Fraud.
Academy Health'2022 -
2022.4
Junjie Luo, Shiping Liu, Guodong (Gordon) Gao, Ritu Agarwal. Racial Disparity in the Availability of Social Determinants of Health Information in EHR Data.
Production and Operations Management Society'2022 -
2022.3
Junjie Luo, Guodong (Gordon) Gao, Ritu Agarwal, and Sally Simpson. "Minority Report" Comes True: Using Behavioral Big Data to Predict Physician Fraud.
Conference on Health IT and Analytics'2022 -
2021.6
Junjie Luo, Xi Chen, Jichao Sun, Yuejia Xiang, Ningyu Zhang and Xiang Wan. Field Embedding: A Unified Grain Based Framework for Word Representation.
North American Association for Computational Linguistics'2021
Industry Collaboration & Experience
Industry Collaboration
- I have been collaborating with several industry partners to develop and deploy healthcare AI solutions for real-world clinical and business problems.
- • WellDoc Inc. (Digital Health - Diabetes Management)
- • DrFirst Inc. (Healthcare Technology - Medication Adherence)
- • MedStar Health (Healthcare System - EHR Analytics)
- • Inovalon Inc. (Healthcare Data Analytics)
- • Johns Hopkins Hospital (Academic Medical Center)
Professional Experience
-
2020
Algorithm Engineer at Jarvis AI Lab, Healthcare Group, Tencent Ltd
Shenzhen, China -
2017-2020
Research Assistant at Shenzhen Research Institute of Big Data
Shenzhen, China -
2017
Data Mining Engineer at Ping An of China, Ping An Technology Co. Ltd
Shenzhen, China -
2016-2017
Research Assistant at School of Management and Economics, CUHK, Shenzhen
Shenzhen, China