INDIANAPOLIS - The IUPUI team of Wen-Hao Chiang, Junfeng Liu, Ziwei Fan and Bo Peng under the guidance of Computer Science Assistant Professor, Dr. Xia Ning presented a comprehensive pattern analysis for new insight to better support future Medicaid program and public health decisions. They integrated the aggregated Medicaid information data of the past five years with third-party data from multiple sources, including the local census and economic data on county levels and city levels and incorporated machine-learning techniques in their analysis. Based on this data, they revealed the underlying clustering patterns among counties and cities, respectively, in the State of Indiana. They also built models to analyze the dominating factors that affect the total Medicaid cost and high-cost Medicaid claims in each year.
"We found that unemployment rate, per capita income, and birth rates are the dominating factors that affect the Medicaid claims," said Junfeng Liu. "Interestingly, from 2012, the unemployment rate was becoming less important, probably because the economy was recovering. And per capita income started to play an important role in recent years."
These findings are particularly helpful to predict the future Medicaid trends based on future census trends and time series analysis and inspire further research on recommendations for the healthcare resource allocation as well as personalized prediction of healthcare needs for future strategic planning.
The 2017 Indiana Medicaid Data Challenge is an event designed by Indiana Chapter of HIMSS. Teams were presented with the de-identified Indiana Medicaid healthcare information with the purpose of demonstrating its usefulness in Population Health after data analysis for correlations, patterns, trends and other indicators to identify risk factors and potential interventions. The team received the data on October 16th and worked tirelessly to create, develop and demonstrate innovative and impactful utilization of this data for their presentation and on-site judging on October 21st.
Bios:
Wen-Hao Chiang:
Wen-Hao Chiang received his B.S. degree from the Department of Electronics Engineering, National Chiao Tung University, Taiwan, in 2012, and his M.S. degree from the Graduate Institute of Electronics Engineering, National Taiwan University, Taiwan, in 2014. He is currently working toward his Ph.D. degree at the Department of Computer & Information Science, Indiana University-Purdue University Indianapolis. His research interests include medical informatics, data mining and machine learning.
Junfeng Liu:
Junfeng Liu received his B.S. degrees from Kelley School of Business, Indiana University, Indiana, and from Sun Yat-sen Business School, Sun Yat-sen University, China, in 2015. He is currently a graduate student at the Department of Computer & Information Science, Indiana University-Purdue University Indianapolis and a University Fellowship at Indiana University. His research is on data mining and machine learning with applications on drug discovery, bioinformatics and healthcare informatics.
Ziwei Fan:
Ziwei Fan received his Bachelor's degree in Network Engineering from South China Agricultural University. He is currently pursuing a Master's degree in the Department of Computer & Information Science in IUPUI. His research interest is applying machine learning algorithms to different domains, including e-commercial recommender system and healthcare system.
Bo Peng:
Bo Peng received his B.E. degree from the Department of Mechanical Engineering, Beijing Institute of Technology, China, in 2017. He is currently working toward his Master degree at the Department of Computer & Information Science, Indiana University-Purdue University Indianapolis with Fellowship. His research interests include Data Mining and Machine Learning.
Written by Lori Vanatsky (lvanatsk@iupui.edu).