GENOMICS ORCHESTRATE
POPULACE HEALTH WITH HUMAN PHEONOTYPE ONTOLOGY
ROADMAP NATIONAL AIM
QUALITY ASSURANCE BACKGROUND
ENGINEER THE
NATIONAL SAFETY STRUCTURE
FIVE POINT PHM MISSION FOR FIT LIFECYCLES IN ANALYTICS
PUBLIC HEALTH AND PATIENT SAFETY
ASSESS BALANCE & BIAS
DATA BALANCE AND BIAS IN REAL WORLD INSTANCES
ASSESS PROCESS
DATA STRUCTURE AND CLEANING IN AN OPEN-SOURCE ECOSYSTEM
HIGHER
EXPERT MEDICAL SCIENCE SAFETY
ASSESS THEMES
DATA ALLIANCES AND SCEINTIFIC THEMES
SCIENCE DATA Fig 2. Science and technology foundations which steward PHM.
for enhanced primary healthcare,1-3 this UN, UK and US HPO Policy Brief proposal aims to accelerate Sustainable Development Goals (SDGs) 3, 8, and 17, contributing to global patient safety and economic growth by 2030.4-7
The UK’s
Department of Health and Social Care, the Department for Science, Innovation and Technology, and Genomics AU Network are key stakeholders in stewarding an HPO standard vocabulary, aligning with international efforts for responsible AI while upholding human rights.8-11 To address critical clinical needs, this
proposal recommends the UK’s Digital Regulation Service (AIDRS) steward the HPO policy’s development, authorisation, and adoption, initially focusing on Primary Care Networks (PCNs) and guided by AI principles derived from scientific themes.12,13
Healthcare Informatics and
Management System Society (HIMSS) maturity should incorporate Higher Expert Medical Science Safety (HEMSS) principles to steward PHM adoption.14 Figure 1 shows a schematic roadmap for HPO Policy to steward data training, bias mitigation, and Explainable AI (XAI), while adhering to ethical, secure, privacy, and informed consent safeguards. The HPO policy will guide adoption of biological modelling (BM) pre- eXam predictors and eXam intercepts, using Generative AI classifications to enhance each citizen’s lifecycle. Figure 1 illustrates a PHM mission for good health and economic growth and depicts
The Pre-eXam Predictor
Biological genomics predictive health modelling
HPO standard predictive health modelling
Algorithmic bias detection and mitigation
Population-specific risk assessment
Socioeconomic vulnerability screening
Geospatial or trend screening
The eXam Intercept
Biological disease intercepts for model wellbeing
HPO disease intercepts for trait wellbeing
Personalised care plans with transparent reasoning
Culturally sensitive communication and care delivery
Resource allocation based on social determinants of health
Location based with health interventions on escalations
X is Fair, Transparent BM in standard HPO
Fair and transparent Biological Models
Fair and transparent HPO Standards
Algorithmic Fairness (implicit/explicit)
Equity in interpretation and public inclusion
Social Fairness with community inclusion
Interpretive Accuracy for Actions
Table 1. A roadmap to steward authorised pre-eXam and eXams classifications for adoption.
WWW.PATHOLOGYINPRACTICE.COM MAY 2025 39
the proposed UN programme for HPO policy with HEMSS stewardship of these classifications.
Background to a human phenotype ontology proposal A unified HPO guide and policy across the UN – as instanced for the UK, and US – is crucial for training high-quality data-driven predictors and intercepts. This policy should incorporate metrics that rigorously assess bias and performance, spanning from initial research to value- based care implementation across points of need.15,16
Figure 2 illustrates the
author’s proposal for national stewardship in biological modelling (BM), using scientific data themes to advance HPO- driven AI for a standardised vocabulary within PHM. Furthermore, an HPO
policy overseen by the DHSC, AIDRS, Genomics Network and CQC provides a roadmap for BM disease risk stratification and segmentation to enable targeted interventions. Figure 2 emphasises the role of open-source data science themes and strategies in effectively managing risks during predictor and intercept training, ultimately ensuring fair HPO classifications. Table 1 outlines the collaborative arrangements within the user and citizen ecosystem, highlighting mechanisms for continuous improvement and accountability in addressing bias and ensuring interpretability. These elements offer insights into AI comparative analysis, Explainable AI (XAI), and the development of robust metrics for predictor and intercept classifiers, as detailed in paper 5.
POPULATION HEALTH
MANAGEMENT MISSION
1 PUBLIC INCLUSIVENESS
NATIONAL PHM QUALITY ASSURANCE
2
STAKEHOLDER ENGAGEMENT
EVALUATE THE
VARIABLES FOR DATA TRAINING
EVALUATE VARIABLES
ASSESS PRIMARY CARE FOR HPO NEIGHBOURHOOD TRIALS
SAFETY INSITUTE
RESEARCH RESOURCE
POPULATION HEALTH
MANAGEMENT
HUMAN PHENOTYPE ONTOLOGY
AI DIGITAL REGULATION
HPO POLICY
DISCUSS
THE PHM MISSION AS A TEN YEAR INFRASTRUCTURE PLAN
4 GOVERN
FIT LIFECYCLES IN ANALYTICS ADOPT THE PHM MISSION
3 DEVELOP PHM
ARCHITECTURE ASSURED PERSONAL CLASSIFIERS
AIDRS-AISI AUTHORITY
PLAN AI ARCHITECTURE WITH MONIITORING
AI
TECHNICAL ADOPTION MISSION
DEVELOP
STEWARDS DISCUSS
DISCUSS FUTURE NHS
FEDERATED LEARNING WITH NEW TECH
AIDRS GOVERNANCE AND AISI EVALUATION OF PHM
DISCUSS CONTINOUS
IMPROVEMENT AND ACCOUNTABILITY
DIGITAL HIMSS
SERVICES ADOPT
GUARDRAILS AND ETHICS FOR A PHM MISSION
ADOPT
NATIONAL GENERATIVE CLASSIFICATION FOR PREDICTOR & INTERCEPTS
EXPLAINABLE MODEL VISION AND LANGUAGE PERFORMANCE
AI TECHNOLOGY
© copyright James Henry
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
