Research

Dissertation: Big Data-Based Health Risk Analytics: A Deep Learning Approach


Value-based healthcare is an emerging healthcare delivery model which incentivizes and rewards physicians for improved patient outcomes and quality of care, rather than the amount of services. The objective of value-based healthcare is to move the healthcare delivery system from reactive disease treatment to a proactive care approach. Health risk analytics serves as the analytical foundation for such proactive healthcare by actively assessing major health risks. My research develops novel deep learning methods for health risk analytics leveraging big data, including clinical claims, wearable sensor signals, and social media. The proposed methods provide find-grained analytical capabilities for three levels of critically important health risks:

  • Patient behavioral risk: understand risk factors of medication nonadherence and opioid addiction behaviors
  • Disease risk: predict major disease risks using wearable sensor signals and clinical claims
  • Policy risk: predict readmission risk for patients and hospitals

Published Journal Articles

Mining E-cigarette Adverse Events in Social Media Using Bi-LSTM Recurrent Neural Network with Word Embedding Representation

Jiaheng Xie, Xiao Liu, Daniel Zeng
Journal of the American Medical Informatics Association (JAMIA) (IF: 4.27), 2017
Volume 25, Issue 1, Pages 72–80

Using Deep Learning to Improve Medication Safety: the Untapped Potential of Social Media

Jiaheng Xie, Daniel Zeng, Zachary Marcum
Therapeutic Advances in Drug Safety (IF: 2.84), 2017
Volume 8, Issue 12, Pages 375-377

Papers Under Review

Understanding Medication Nonadherence from Social Media: A Sentiment-Enriched Deep Learning Approach

Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
Revising for the 3rd Round of Review at MIS Quarterly
Best Paper Runner-Up (ICSH 2019)

Readmission Prediction for Patients with Heterogeneous Hazard: A Trajectory-Based Deep Learning Approach

Jiaheng Xie, Bin Zhang, Jian Ma, Daniel Zeng, Jenny Lo-Ciganic
Revising for the 2nd Round of Review at Information Systems Research
Best Paper Runner-Up (ICSH 2018)

Write Like a Pro or an Amateur? The Effect of Medical Language Formality in Senior Care: A Multi-Method Approach

Jiaheng Xie, Bin Zhang, Susan Brown, Daniel Zeng
Revising for the 2nd Round of Review at Information Systems Research

[Pre-Ph.D.] The Effect of Web Page Background Color on the Uniqueness of Customized Products

Jiaheng Xie, Wangsheng Zhu, Kanliang Wang, Jun Pang
Under Review at Information & Management

Completed Papers

The Impact of FBI Shutdown Operations on Cybercriminals and Product Sales in Dark Net Markets

Mohammadreza Ebrahimi, Jiaheng Xie, Wei Chen, Hsinchun Chen
In Preparation for Submission to MIS Quarterly

Discovering Barriers to Opioid Addiction Treatment from Social Media: A Similarity Network-Based Deep Learning Approach

Jiaheng Xie, Zhu Zhang, Xiao Liu, Daniel Zeng
In Preparation for Submission to Journal of Management Information Systems

Research in Progress

Predicting Parkinson’s Disease Risk Using Wearable Sensor Data: A Multi-View Attention Convolutional Neural Network Approach

Jiaheng Xie, Daniel Zeng
Method Development, Targeted at Information Systems Research

Customers at Fingertips: A Large-Scale Field Experiment Using Tap Stream Data on Mobile Apps

Wei Chen, Karen Xie, Dong Jing, Jiaheng Xie
Preparing Field Experiment, Targeted at Management Science

Bridging the Vocabulary Gap in Online Knowledge Community: A Graph Convolutional Network Approach

Jiaheng Xie, Xiao Liu, Daniel Zeng
Method Development, Targeted at Information Systems Research

Modeling Parkinson’s Disease Progression Using Wearable Sensor Data: A Generative Adversarial Network Approach

Jiaheng Xie, Zhu Zhang, Daniel Zeng
Method Development, Targeted at MIS Quarterly

Predicting Opioid Overdose Using Recurrent Neural Networks

Jiaheng Xie, Daniel Zeng, Jenny Lo-Ciganic
Data Analysis, Targeted at Journal of the American Medical Association (JAMA)

Predicting Prescription Opioid Misuse Using Deep Multi-Task Learning

Jiaheng Xie, Yongcheng Zhan, Jenny Lo-Ciganic
Data Analysis, Targeted at Proceedings of the National Academy of Sciences (PNAS)

Conference Preceedings and Workshops (*Presenting Author)

Discovering Barriers to Opioid Addiction Treatment from Social Media: A Similarity Network-Based Deep Learning Approach

*Jiaheng Xie, Zhu Zhang, Xiao Liu, Daniel Zeng
International Conference on Information Systems (ICIS) 2019, Munich, Germany

Understanding Medication Nonadherence from Social Media: A Sentiment-Enriched Deep Learning Approach

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
Conference on Information Systems and Technology (CIST) 2019, Seattle, USA

Discovering Barriers to Opioid Addiction Treatment Using Similarity Network-Based Deep Learning

*Jiaheng Xie, Zhu Zhang, Xiao Liu, Daniel Zeng
China Summer Workshop on Information Management (CSWIM) 2019, Shenzhen, China

Understanding Medication Nonadherence Using Sentiment-Enriched Deep Learning

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
China Summer Workshop on Information Management (CSWIM) 2019, Shenzhen, China

Understanding Opioid Addiction with Similarity Network-Based Deep Learning

*Jiaheng Xie, Zhu Zhang, Xiao Liu, Daniel Zeng
International Conference for Smart Health (ICSH) 2019, Shenzhen, China

Extracting Medication Nonadherence Reasons with Sentiment-Enriched Deep Learning

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
International Conference for Smart Health (ICSH) 2019, Shenzhen, China

Readmission Risk Prediction for Patients with Heterogeneous Hazard: A Trajectory-Aware Deep Learning Approach

*Jiaheng Xie and Bin Zhang
International Conference on Information Systems (ICIS) 2018, San Francisco, USA

Write Like a Pro or Amateur? The Effect of Online Caregiver Forum Writing Professionalism

*Jiaheng Xie, Bin Zhang, Daniel Zeng
Conference on Information Systems and Technology (CIST) 2018, Pheonix, USA

Discovering Medication Nonadherence Reasons with Sentiment-Enriched Deep Learning Approach

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
INFORMS Workshop on Data Science 2018, Pheonix, USA

Predicting Hospital Readmission Risk Using Trajectory-Based Deep Learning Approach

*Jiaheng Xie, Bin Zhang, Daniel Zeng
INFORMS Workshop on Data Science 2018, Pheonix, USA

Readmission Prediction Using Trajectory-Based Deep Learning Approach

Jiaheng Xie, Bin Zhang, Daniel Zeng
International Conference for Smart Health (ICSH) 2018, Wuhan, China
Best Paper Runner-Up

Predicting Hospital Readmission Risk Using Trajectory-Based Deep Learning Approach

*Jiaheng Xie, Bin Zhang, Daniel Zeng
Conference on Health IT and Analytics (CHITA) 2018, Washington, D.C., USA

Discovering Medication Nonadherence Reasons with Sentiment-Enriched Deep Learning Approach

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
Conference on Health IT and Analytics (CHITA) 2018, Washington, D.C., USA

Predicting Hospital Readmission with Deep Learning

Jiaheng Xie, Bin Zhang, Daniel Zeng
China Summer Workshop on Information Management (CSWIM) 2018, Qingdao, China

Understanding Reasons for Medication Nonadherence: An Exploration in Social Media Using Sentiment-Enriched Deep Learning Approach

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
International Conference on Information Systems (ICIS) 2017, Seoul, South Korea

Understanding Medication Nonadherence from Social Media: A Sentiment-Enriched Deep Learning Approach

*Jiaheng Xie, Xiao Liu, Daniel Zeng, Xiao Fang
INFORMS Annual Meeting 2017, Houston, USA

Mining E-cigarette Adverse Events in Social Media Using Bi-LSTM Recurrent Neural Network with Word Embedding Representation

*Jiaheng Xie, Xiao Liu, Daniel Zeng
INFORMS Annual Meeting 2016, Nashiville, USA

[Pre-Ph.D.] Consumers’ Purchase Intention of Online Product Customization Using Different Terminals with/without Default Template

Jiaheng Xie, Wangsheng Zhu, Kanliang Wang
International Conference on HCI in Business 2015, Los Angeles, USA

[Pre-Ph.D.] An Improvement to E-Commerce Recommendation using Product Network Analysis

*Jiaheng Xie, Wangsheng Zhu, Kanliang Wang
Pacific-Asian Conference on Information Systems (PACIS) 2014, Chengdu, China