Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Methodological Approaches to and Reported Performance of Applications of Automated Machine Learning in Diabetes Risk

Alexandre Castonguay¹, Sandrine Hegg-Deloye^1,2, Arthur Chatton^2,3

¹Faculté des sciences infirmières, Université de Montréal, Pavillon Marguerite d'Youville, 2375, Chemin de la Côte-Sainte-Catherine, Montréal, QC, H3T 1A8, Canada, 1 4182626594.

|May 12, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Implementation of a Multidevice Telemonitoring Program for Home-Based Nursing Care in Quebec: Qualitative Report.

JMIR medical informatics·2026

Same author

Experiences of Older Adults and Caregivers With Home Telemonitoring for Heart Failure in Canada: Qualitative Study.

JMIR aging·2026

Same author

AI-assisted teams outperform AI-led teams but not human-only teams in assessing research reproducibility in quantitative social science.

Proceedings of the National Academy of Sciences of the United States of America·2026

Same author

Correction: Perceptions and Intentions of Nursing Students Regarding Digital Health: Cross-Sectional Study.

JMIR nursing·2026

Same author

Perceptions and Intentions of Nursing Students Regarding Digital Health: Cross-Sectional Study.

JMIR nursing·2026

Same author

Is Checking for Sequential Positivity Violations Getting You Down? Try sPoRT!

Epidemiology (Cambridge, Mass.)·2025

Automated machine learning (AutoML) shows promise for type 2 diabetes (T2D) risk prediction. However, limited external validation and data integration hinder clinical use, necessitating prospective data and open science principles for future models.

Area of Science:

Computational medicine
Health informatics
Artificial intelligence in healthcare

Background:

Type 2 diabetes (T2D) presents a significant healthcare burden, emphasizing the need for effective prevention and early detection strategies.
Automated machine learning (AutoML) offers potential for personalized T2D risk prediction and intervention, but clinical deployment is hindered by data limitations and validation issues.

Purpose of the Study:

To systematically review and map the application of AutoML techniques for T2D risk prediction.
To assess the capability of AutoML models in integrating diverse data types, including clinical, behavioral, environmental, and genomic information.

Main Methods:

A rapid systematic review guided by PRISMA standards was conducted across six major databases.
Included studies published between 2015-2025 utilizing AutoML for T2D prediction with at least two data types.

Keywords:

AI AutoML artificial intelligence automated machine learning explainable artificial intelligence machine learning validation multimodal data integration type 2 diabetes risk prediction

Related Experiment Videos

Systematic screening, data extraction, and synthesis were performed by two independent reviewers, with AI assistance for arbitration.

Main Results:

Thirteen studies met the inclusion criteria, demonstrating methodological diversity in AutoML pipelines.
Reported model performance varied (AUC 0.74-0.99), but external validation was infrequent.
Integration of behavioral and environmental data was partial, and genomic data was notably absent; transparency and reproducibility were lacking.

Conclusions:

AutoML has substantial potential to enhance T2D risk prediction through automation and explainability.
Future AutoML development requires prospective, multicenter datasets and integration of diverse data, including genomics.
Adherence to open science principles (transparency, reproducibility, interpretability) is crucial for clinical adoption and generalizability of AutoML in T2D risk prediction.