Mission-Oriented Public Health Informatics Consortium
About the Consortium
Mission and Vision
The Mission-Oriented Public Health Informatics and Technology Consortium (M-PHIT), launched in October 2024, is a research and development initiative focused on addressing urgent, high-priority issues in public health, population health, and healthcare. These challenges are selected based on their potential to significantly strengthen the nation’s ability to respond to public health threats and healthcare concerns. The program begins by identifying and prioritizing the most critical problems to target. Advancing mission-oriented research and development in public health and healthcare informatics requires innovative thinking and the development of novel approaches, making this work both essential and highly impactful.
Public Health Challenges
01
Problem:
Public health data is siloed across states, agencies, and healthcare providers, making real-time data sharing and aggregation difficult. Surveillance systems are not equipped to handle real-time cases, hospitalizations, or outcomes. Data often lacked demographic detail, leading to blind spots in understanding the impact of public health outbreaks and pandemics on vulnerable populations.
02
Impact:
Slow detection of outbreaks, delayed response efforts, and hindered national coordination. Inconsistent data reporting and inability to accurately track the spread and severity of the disease. Inadequate response to racial, ethnic, and socio-economic disparities in infection rates and outcomes.
03
Informatics:
Building interoperable, standardized systems that can communicate across jurisdictions and sectors. Developing advanced, scalable surveillance tools that can ensure data quality, timeliness, and completeness. Ensuring equity-focused data collection and integrating social determinants of health.
04
Workforce:
Many public health agencies lack personnel trained in modern informatics and have outdated IT infrastructure. Limited capacity to adapt quickly or leverage advanced technologies like pipeline processing in real time, big data analytics, knowledge and experience working with real raw data, predictive modeling, AI, and machine learning. Training and expanding the public health informatics workforce and modernizing infrastructure
M-PHIT is built on modern technologies and recent breakthroughs in health informatics
Recent breakthroughs that significantly advance public health and healthcare: FHIR and PM GenAI.
FHIR developed by Grahame Grieve stands for Fast Healthcare Interoperability Resources. FHIR uses RESTful APIs, making it easier for different healthcare systems to connect, share data, and work together. It revolutionized public health and healthcare data exchange and is used on the web, mobile, cloud, EHRs, and wearable devices.
PM GenAI, developed by Philip de Melo, stands for Principal Model Generative AI. It is regarded as one of the most advanced innovations in modern data science. PM GenAI transforms observed data into a posterior probabilistic data volume and leverages artificial intelligence to filter out low-probability data, significantly enhancing data quality and boosting the accuracy of predictive models. It is widely used in public health and healthcare for precise diagnostics of diseases such as diabetes and lung cancer.
Fast Healthcare Interoperability Resources (FHIR)
Grahame Grieve
Grahame Grieve is an Australian health data scientist with 30 years of experience in healthcare, spanning laboratory science, clinical research, enterprise healthcare system development, national program architecture, data exchange standards development, and open-source community development. Grahame has spent the last decade building FHIR, which is both an open community dedicated to developing open and simple ways to exchange healthcare data between systems, and is the basis for most new work building integrated healthcare systems and processes. As the community lead, Grahame’s interactions span the breadth of the healthcare system, from academic interests, including clinical trials and research, through working with vendors, providers, and government programs to dealing with politicians concerned with the future of the healthcare system. The systems that we know and fail to love around the world are based on institutions as the primary means by which healthcare is provided, funded, and the outcomes are assessed. Grahame’s long-term interest is in how to recast the healthcare system so that the individuals – patients, their families and carers, and the care providers – become the focus of the system, not the institutions that bind them. We’ll always need institutions, but they should serve the individuals.
Principal Model Generative Artificial Intelligence (PM GenAI)
Philip de Melo
Dr. Philip de Melo is an American researcher and academic specializing in computer science, applied mathematics, and public health informatics. He holds both a Ph.D. and a D.Sc. in biostatistics, applied mathematics, informatics, and computer science, and is widely regarded as one of the leading experts in public health informatics. With over 20 years of experience in health informatics, Dr. de Melo has served in several prestigious roles, including as a UNESCO professor in public health, principal investigator (PI) of the International Consortium of HIV Informatics, and PI of the Workforce Development Program funded by the Office of the National Coordinator for Health Information Technology (ONC).He gained international recognition for his work in big data analytics, and more recently, for developing PM GenAI—a generative AI algorithm with the potential to revolutionize public health and healthcare analytics. In practice, much of the recorded health data is of poor quality, which compromises the accuracy and reliability of even the most advanced machine learning models. Dr. de Melo’s innovation addresses this challenge by shifting the paradigm of generative AI: instead of using input text to generate content, PM GenAI accepts a desired accuracy level as input. The system then analyzes the existing (observed) data, generates stochastic data volumes, and filters out outcomes with low probability, thereby enhancing data quality and model accuracy and performance.