Towards Personalized Cardiometabolic Risk Prediction: A Fusion of Exposome and AI DOI Creative Commons
Zeinab Shahbazi, Sławomir Nowaczyk

Heliyon, Journal Year: 2024, Volume and Issue: 11(1), P. e40859 - e40859

Published: Dec. 20, 2024

The influence of the exposome on major health conditions like cardiovascular disease (CVD) is widely recognized. However, integrating diverse factors into predictive models for personalized assessments remains a challenge due to complexity and variability environmental exposures lifestyle factors. A machine learning (ML) model designed predicting CVD risk introduced in this study, relying easily accessible This approach particularly novel as it prioritizes non-clinical, modifiable exposures, making applicable broad public screening assessments. Assessments were conducted using both internal external validation groups from multi-center cohort, comprising 3,237 individuals diagnosed with South Korea within twelve years their baseline visit, along an equal number participants without these control group. Examination 109 variables participants' visits spanned physical measures, factors, choices, mental events, early-life For prediction, Random Forest classifier was employed, performance compared integrative ML clinical variables. Furthermore, data preprocessing involved normalization handling missing values enhance accuracy. model's decision-making process advanced explainability method. Results indicated comparable between exposome-based model, achieving AUC 0.82(+/-)0.01, 0.70(+/-)0.01, 0.73(+/-)0.01. study underscores potential leveraging early intervention strategies. Additionally, significant identifying pinpointed, including daytime naps, completed full-time education, past tobacco smoking, frequency tiredness/unenthusiasm, current work status.

Language: Английский

Integrating the exposome and one health approach to national health surveillance: an opportunity for Latin American countries in health preventive management DOI Creative Commons

Patricia Matus,

Cinthya Urquidi, Marcela Cárcamo

et al.

Frontiers in Public Health, Journal Year: 2024, Volume and Issue: 12

Published: Aug. 14, 2024

The exposome approach, emphasizing lifelong environmental exposures, is a holistic framework exploring the intricate interplay between genetics and environment in shaping health outcomes. Complementing this, one approach recognizes interconnectedness of human ecological within shared ecosystem, extending to planetary health, which encompasses entire planet. Integrating Disease Surveillance Systems with exposome, signifies paradigm shift management, fostering comprehensive public framework. This publication advocates for combining traditional surveillance health/planetary proposing three-step approach: analysis, territorial intervention identified issues, an analytical phase assessing interventions. Particularly relevant Latin American countries facing double burden diseases, integrating into proves cost-effective by leveraging existing data measurements. In conclusion, integration approaches presents robust monitoring population especially regions like America complex challenges. innovative enables tailored interventions, disease outbreak predictions, understanding links environment, offering substantial benefits prevention despite

Language: Английский

Citations

2

Weight of evidence evaluation of the metabolism disrupting effects of triphenyl phosphate using an expert knowledge elicitation approach DOI Creative Commons
Claire Beausoleil, Anne Thébault, Patrik L. Andersson

et al.

Toxicology and Applied Pharmacology, Journal Year: 2024, Volume and Issue: 489, P. 116995 - 116995

Published: June 11, 2024

Identification of Endocrine-Disrupting Chemicals (EDCs) in a regulatory context requires high level evidence. However, lines evidence (e.g. human, vivo, vitro or silico) are heterogeneous and incomplete for quantifying the adverse effects mechanisms involved. To date, appraisal metabolism-disrupting chemicals (MDCs), no harmonised guidance to assess weight has been developed at EU international level. explore how develop this, we applied formal Expert Knowledge Elicitation (EKE) approach within European GOLIATH project. EKE captures expert judgment quantitative manner provides an estimate uncertainty final opinion. As proof principle, selected one suspected MDC -triphenyl phosphate (TPP) - based on its related endpoints (obesity/adipogenicity) relevant metabolic disruption putative Molecular Initiating Event (MIE): activation peroxisome proliferator activated receptor gamma (PPARγ). We conducted systematic literature review assessed quality with two independent groups experts GOLIATH, objective categorising properties TPP, by applying approach. Having followed entire process separately, both arrived same conclusion, designating TPP as "suspected MDC" overall agreement exceeding 85%, indicating robust reproducibility. The method be important way bring together scientists diverse expertise is recommended future work this area.

Language: Английский

Citations

2

Chemical exposome and children health: identification of dose-response relationships from meta-analyses and epidemiological studies DOI
Audrey Rocabois,

Margaux Sanchez,

Claire Philippat

et al.

Environmental Research, Journal Year: 2024, Volume and Issue: 262, P. 119811 - 119811

Published: Aug. 17, 2024

Language: Английский

Citations

2

Science evolves but outdated testing and static risk management in the US delay protection to human health DOI Creative Commons
Maricel V. Maffini, Laura N. Vandenberg

Frontiers in Toxicology, Journal Year: 2024, Volume and Issue: 6

Published: Aug. 13, 2024

Science evolves but outdated testing and static risk management in the US delay protection to human health

Language: Английский

Citations

0

Towards Personalized Cardiometabolic Risk Prediction: A Fusion of Exposome and AI DOI Creative Commons
Zeinab Shahbazi, Sławomir Nowaczyk

Heliyon, Journal Year: 2024, Volume and Issue: 11(1), P. e40859 - e40859

Published: Dec. 20, 2024

The influence of the exposome on major health conditions like cardiovascular disease (CVD) is widely recognized. However, integrating diverse factors into predictive models for personalized assessments remains a challenge due to complexity and variability environmental exposures lifestyle factors. A machine learning (ML) model designed predicting CVD risk introduced in this study, relying easily accessible This approach particularly novel as it prioritizes non-clinical, modifiable exposures, making applicable broad public screening assessments. Assessments were conducted using both internal external validation groups from multi-center cohort, comprising 3,237 individuals diagnosed with South Korea within twelve years their baseline visit, along an equal number participants without these control group. Examination 109 variables participants' visits spanned physical measures, factors, choices, mental events, early-life For prediction, Random Forest classifier was employed, performance compared integrative ML clinical variables. Furthermore, data preprocessing involved normalization handling missing values enhance accuracy. model's decision-making process advanced explainability method. Results indicated comparable between exposome-based model, achieving AUC 0.82(+/-)0.01, 0.70(+/-)0.01, 0.73(+/-)0.01. study underscores potential leveraging early intervention strategies. Additionally, significant identifying pinpointed, including daytime naps, completed full-time education, past tobacco smoking, frequency tiredness/unenthusiasm, current work status.

Language: Английский

Citations

0