Collecting Public Health Data at a Global Scale D. Cenk Erdil, PhD Marist College DataCloud 2015 November 15, 2015
Outline Background A Public Health CPS Data Collection Marist MAgIC Summary Q&A 2
Background A Computer Scientist Grid and Cloud Computing Research Associate Research Scientist in Public Health Informatics Global Health Initiative focusing HIV/AIDS Data Collection for Monitoring and Evaluation Resource-limited settings Assistant Professor at Marist College Applied Informatics Center for Cloud Computing and mhealth 3
Cloud Infrastructure @ Marist Marist-IBM Joint Study A $10m mainframe computer in 1988, and ongoing Data Analytics, SDN, Cloud New York State Cloud Computing and Analytics Center $3 million grant in 2012 Center for Collaborative and On-Demand Computing $4 million NYSTAR grant in 2013 Amazon (thru AWS Educate), and other industry partners 4
A Cyber-Physical System 5
Stakeholders US Government - PEPFAR Other Governments Country, Regional, District Third-party Governments (indirect support & grants) Reporting Agencies - CDC Atlanta, WHO, Others Requirements and Oversight Intervention Programs - Plans Indicators - Data Elements 6
Stakeholders Funding Agencies Bill and Melinda Gates Foundation Robert Wood Johnson Foundation Others NGOs and faith-based organizations Open-Source Community: OpenMRS and DHIS Medical Professionals Patients (Populations) 7
Biomedical Informatics Public Health Informatics Population Level Medical Informatics Patient Level Bioinformatics Molecular Level 8
Columbia ICAP Working under CDC Guidelines ICAP is a CDC PEPFAR implementing partner PEPFAR: The President s Emergency Plan for AIDS Relief 2003-2013: Clinical, Monitoring and Implementation 2013-2023: an AIDS-free generation ICAP works: 20 countries; 2,000 health facilities 200,000 patients; 2,000,000 people 9
CDC Guidelines Maintenance Scope of Public Health Informatics Evaluation Implementation Development Design Conceptualization Disease Reporting Outbreak Management Cancer Registries Risk Factor Surveillance Newborn Screening Health Information Systems 10
Public Health Informatics Monitoring and Evaluation (Routine Reporting) Responsive Websites and Optimized Systems Cloud Implementations Mobile Public Health Apps Data Analytics GIS Systems - Mapping Visualizations 11
Health Informatics Cloud Initiated with an Amazon Research and Education Grant PHI 2 - Public Health Informatics Infrastructure 12
Monitoring and Evaluation PHI 2 - Public Health Informatics Infrastructure 13
ICAP Dashboard 14
ICAP Dashboard 15
ICAP Dashboard 16
Unified Reporting System 17
Country Mapper 18
State-of-the-Cloud 19
Biomedical Informatics Public Health Informatics Population Level Medical Informatics Patient Level Bioinformatics Molecular Level 20
Marist Applied Informatics Center for Cloud Computing and mhealth mhealth on Cloud Computing ACASI tools Android-based enterprise mobile apps Text Highlighting and Text-to-Speech Using JavaRosa - OpenDataKit LinuxONE - IBM Enterprise Linux on z Systems (pending funding!) Photo Courtesy pedaids.org 21
Nature of Public Health Data Monitoring and Evaluation Quarterly Data - Aggregated Relatively manageable data Epidemiology How bacteria change forms How viruses respond to certain populations HIV EVD 22
Nature of Public Health Data (cont d) Genetics - Proteomics High-dimensional data Other factors such as: Complexity of social and sexual networks Matching with other datasets 23
From Data to Actions Image courtesy of CDC CDC Health Protection Framework 24
Simon Hay, et al., Big Data Opportunities for Global Infectious Disease Surveillance 25
Summary Slow transition into Big Data era for Public Health Translational research How to collect data on disease epidemics How to make it available for rapid comparison Robust cloud infrastructures for data collection Routine activities Emergency response Digital Disease Detection 26
Thank You! We are seeking collaborators cenk.erdil@marist.edu ccac.marist.edu 27