MED-AI-01: Hospital-Local Multimodal Artificial Intelligence for Predictive Medicine Using Electronic Health Records, Genomics, and Biometric Monitoring

Abstract MED-AI-01 proposes a privacy-focused, hospital-local artificial intelligence architecture designed to investigate the prediction of future illnesses through the integration of electronic health records (EHRs), family medical histories, genomic information, lifestyle factors, and longitudinal biometric monitoring. The proposed system utilizes locally hosted large language models (LLMs), including models deployed through Ollama, alongside conventional machine-learning algorithms to analyze medical information and estimate disease risk. Unlike architectures that depend on external cloud-based AI inference, MED-AI-01 investigates whether hospitals can maintain independent, locally controlled AI infrastructure while protecting sensitive patient information through restricted network access, authentication, encryption, and institutional data governance. A central feature is the continuous integration of newly available patient information. As authorized clinical records, laboratory results, and biometric measurements are updated, the system would automatically reassess relevant disease-risk estimates. Initial research will investigate predictive models for type 2 diabetes, cardiovascular disease, and chronic kidney disease, with the long-term objective of expanding to additional medical conditions. The project emphasizes physician oversight, transparent communication of clinically significant findings, patient autonomy, and the ethical use of predictive medical technologies. Research Objectives Investigate multimodal AI architectures for predicting multiple illnesses using longitudinal medical information. Evaluate whether genomic, familial, and biometric information improves established clinical prediction methods. Explore secure hospital-local AI deployment using Ollama and open-weight language models. Develop a framework for automatically updating disease-risk estimates when new patient information becomes available. Investigate privacy-preserving collaboration between independently operated hospital AI systems. Research Status MED-AI-01 is an early-stage, independently developed research proposal. No clinical prediction system has been validated or approved for patient care, and no real patient medical records have been used in this proposed study. Initial development is intended to use synthetic medical datasets, with future clinical research contingent upon appropriate institutional collaboration, authorization, and ethical oversight. Open Science and Future Development This project supports open scientific collaboration, reproducible research, and transparent evaluation of artificial intelligence in healthcare. Scientific findings and selected research materials may be published openly, while future software implementations, specialized technical innovations, and commercial applications may be developed under separate licensing arrangements. The long-term objective is to investigate whether secure, institutionally controlled AI systems can improve preventive healthcare and support earlier identification of disease risk. Researcher: Peter McGrathResearch Type: Independent Undergraduate Research ProposalDate: November 11, 2026Status: Preliminary proposal — Not peer reviewed

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-12
DOI
https://doi.org/10.5281/zenodo.24148991
Primary Topic
Machine Learning in Healthcare
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

MED-AI-01: Hospital-Local Multimodal Artificial Intelligence for Predictive Medicine Using Electronic Health Records, Genomics, and Biometric Monitoring

Peter McGrath
Zenodo (CERN European Organization for Nuclear Research)
Machine Learning in Healthcare
article

MED-AI-01: Hospital-Local Multimodal Artificial Intelligence for Predictive Medicine Using Electronic Health Records, Genomics, and Biometric Monitoring

Peter McGrath
article en

Abstract

Abstract MED-AI-01 proposes a privacy-focused, hospital-local artificial intelligence architecture designed to investigate the prediction of future illnesses through the integration of electronic health records (EHRs), family medical histories, genomic information, lifestyle factors, and longitudinal biometric monitoring. The proposed system utilizes locally hosted large language models (LLMs), including models deployed through Ollama, alongside conventional machine-learning algorithms to analyze medical information and estimate disease risk. Unlike architectures that depend on external cloud-based AI inference, MED-AI-01 investigates whether hospitals can maintain independent, locally controlled AI infrastructure while protecting sensitive patient information through restricted network access, authentication, encryption, and institutional data governance. A central feature is the continuous integration of newly available patient information. As authorized clinical records, laboratory results, and biometric measurements are updated, the system would automatically reassess relevant disease-risk estimates. Initial research will investigate predictive models for type 2 diabetes, cardiovascular disease, and chronic kidney disease, with the long-term objective of expanding to additional medical conditions. The project emphasizes physician oversight, transparent communication of clinically significant findings, patient autonomy, and the ethical use of predictive medical technologies. Research Objectives Investigate multimodal AI architectures for predicting multiple illnesses using longitudinal medical information. Evaluate whether genomic, familial, and biometric information improves established clinical prediction methods. Explore secure hospital-local AI deployment using Ollama and open-weight language models. Develop a framework for automatically updating disease-risk estimates when new patient information becomes available. Investigate privacy-preserving collaboration between independently operated hospital AI systems. Research Status MED-AI-01 is an early-stage, independently developed research proposal. No clinical prediction system has been validated or approved for patient care, and no real patient medical records have been used in this proposed study. Initial development is intended to use synthetic medical datasets, with future clinical research contingent upon appropriate institutional collaboration, authorization, and ethical oversight. Open Science and Future Development This project supports open scientific collaboration, reproducible research, and transparent evaluation of artificial intelligence in healthcare. Scientific findings and selected research materials may be published openly, while future software implementations, specialized technical innovations, and commercial applications may be developed under separate licensing arrangements. The long-term objective is to investigate whether secure, institutionally controlled AI systems can improve preventive healthcare and support earlier identification of disease risk. Researcher: Peter McGrathResearch Type: Independent Undergraduate Research ProposalDate: November 11, 2026Status: Preliminary proposal — Not peer reviewed

Zenodo (CERN European Organization for Nuclear Research)
Appalachian State University (US)
Good health and well-being
Machine Learning in Healthcare
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon