Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA

Morteza Naghavi; Amir Azimi; Kyle Atlas; Anthony P. Reeves; Chenyu Zhang; Jakob Wasserthal; Seyed Reza Mirjalili; Mohammadhossein Mozafarybazargany; Ali Hashemi; Thomas Atlas; Claudia I. Henschke; David F. Yankelevitz; Jeffrey I. Mechanick; Andrea D. Branch; Susan K. Fried; Khurram Nasir; Zahi A. Fayad; Michael V. McConnell; Rozemarijn Vliegenthart; David J. Maron; Jagat Narula; Kim A. Williams; Prediman K. Shah; Matthew J. Budoff; Daniel Levy; Emelia J. Benjamin; Robert A. Kloner; Nathan D. Wong

doi:10.1186/s13098-025-01970-8

Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA

Morteza Naghavi, Amir Azimi, Kyle Atlas, Anthony P. Reeves, Chenyu Zhang, Jakob Wasserthal, Seyed Reza Mirjalili, Mohammadhossein Mozafarybazargany, Ali Hashemi, Thomas Atlas, Claudia I. Henschke, David F. Yankelevitz, Jeffrey I. Mechanick, Andrea D. Branch, Susan K. Fried, Khurram Nasir, Zahi A. Fayad, Michael V. McConnell, Rozemarijn Vliegenthart, David J. MaronJagat Narula, Kim A. Williams, Prediman K. Shah, Matthew J. Budoff, Daniel Levy, Emelia J. Benjamin, Robert A. Kloner, Nathan D. Wong

Research output: Contribution to journal › Article › peer-review

Abstract

Background: The AI-CVD initiative aims to maximize the value of coronary artery calcium (CAC) scans for cardiometabolic risk prediction by extracting opportunistic screening information. We investigated whether artificial intelligence (AI)-derived measures from CAC scans are associated with new-onset Type 2 diabetes mellitus (T2DM) in adults without obesity or hyperglycemia. Methods: Baseline CAC scans and up to 23 years of follow-up data were analyzed for participants without obesity (body mass index < 30 kg/m²) and hyperglycemia (fasting plasma glucose < 100 mg/dL) from the Multi-Ethnic Study of Atherosclerosis (MESA). AI-derived measures included liver attenuation index (LAI), subcutaneous fat index (SFI), total visceral fat index (TVFI), epicardial fat index (EFI), skeletal muscle index, and skeletal muscle mean density. Cox regression models compared highest vs. lowest quartiles of each AI-derived metric for T2DM risk. Multivariable models assessed adjusted predictive value using Wald chi-squared statistics. Subgroup analyses stratified participants by demographic and clinical factors. Results: During a median follow-up of 19.7 years among 2,993 participants (baseline mean age 61.9 ± 10.5 years, 53% women), 257 participants (8.6%) developed T2DM. Key predictors included LAI (HR: 3.13, 95% CI: 2.15–4.55), SFI (HR: 2.85, 95% CI: 1.93–4.21), TVFI (HR: 2.49, 95% CI: 1.72–3.60), and EFI (HR: 1.59, 95% CI: 1.09–2.32). LAI remained the most robust predictor after adjusting for all metrics (Wald χ² = 38.24). Subgroup analyses confirmed LAI’s consistent predictive performance. Conclusion: AI-derived adiposity measures from CAC scans—especially liver fat—can identify adults without obesity or hyperglycemia at elevated risk for developing T2DM. These findings underscore the potential of AI-enabled opportunistic screening during CAC imaging to support early T2DM risk stratification in individuals not captured by current clinical guidelines.

Original language	English (US)
Article number	419
Journal	Diabetology and Metabolic Syndrome
Volume	17
Issue number	1
DOIs	https://doi.org/10.1186/s13098-025-01970-8
State	Published - Dec 2025

Keywords

AI-CVD
Artificial intelligence
Coronary artery calcium (CAC) scans
Opportunistic screening
Type 2 diabetes mellitus

ASJC Scopus subject areas

Internal Medicine
Endocrinology, Diabetes and Metabolism

Access to Document

10.1186/s13098-025-01970-8

Cite this

Naghavi, M., Azimi, A., Atlas, K., Reeves, A. P., Zhang, C., Wasserthal, J., Mirjalili, S. R., Mozafarybazargany, M., Hashemi, A., Atlas, T., Henschke, C. I., Yankelevitz, D. F., Mechanick, J. I., Branch, A. D., Fried, S. K., Nasir, K., Fayad, Z. A., McConnell, M. V., Vliegenthart, R., ... Wong, N. D. (2025). Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA. Diabetology and Metabolic Syndrome, 17(1), Article 419. https://doi.org/10.1186/s13098-025-01970-8

Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA. / Naghavi, Morteza; Azimi, Amir; Atlas, Kyle et al.
In: Diabetology and Metabolic Syndrome, Vol. 17, No. 1, 419, 12.2025.

Research output: Contribution to journal › Article › peer-review

Naghavi, M, Azimi, A, Atlas, K, Reeves, AP, Zhang, C, Wasserthal, J, Mirjalili, SR, Mozafarybazargany, M, Hashemi, A, Atlas, T, Henschke, CI, Yankelevitz, DF, Mechanick, JI, Branch, AD, Fried, SK, Nasir, K, Fayad, ZA, McConnell, MV, Vliegenthart, R, Maron, DJ, Narula, J, Williams, KA, Shah, PK, Budoff, MJ, Levy, D, Benjamin, EJ, Kloner, RA & Wong, ND 2025, 'Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA', Diabetology and Metabolic Syndrome, vol. 17, no. 1, 419. https://doi.org/10.1186/s13098-025-01970-8

@article{d224078699f74cd99338632157ba0606,

title = "Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA",

abstract = "Background: The AI-CVD initiative aims to maximize the value of coronary artery calcium (CAC) scans for cardiometabolic risk prediction by extracting opportunistic screening information. We investigated whether artificial intelligence (AI)-derived measures from CAC scans are associated with new-onset Type 2 diabetes mellitus (T2DM) in adults without obesity or hyperglycemia. Methods: Baseline CAC scans and up to 23 years of follow-up data were analyzed for participants without obesity (body mass index < 30 kg/m²) and hyperglycemia (fasting plasma glucose < 100 mg/dL) from the Multi-Ethnic Study of Atherosclerosis (MESA). AI-derived measures included liver attenuation index (LAI), subcutaneous fat index (SFI), total visceral fat index (TVFI), epicardial fat index (EFI), skeletal muscle index, and skeletal muscle mean density. Cox regression models compared highest vs. lowest quartiles of each AI-derived metric for T2DM risk. Multivariable models assessed adjusted predictive value using Wald chi-squared statistics. Subgroup analyses stratified participants by demographic and clinical factors. Results: During a median follow-up of 19.7 years among 2,993 participants (baseline mean age 61.9 ± 10.5 years, 53\% women), 257 participants (8.6\%) developed T2DM. Key predictors included LAI (HR: 3.13, 95\% CI: 2.15–4.55), SFI (HR: 2.85, 95\% CI: 1.93–4.21), TVFI (HR: 2.49, 95\% CI: 1.72–3.60), and EFI (HR: 1.59, 95\% CI: 1.09–2.32). LAI remained the most robust predictor after adjusting for all metrics (Wald χ² = 38.24). Subgroup analyses confirmed LAI{\textquoteright}s consistent predictive performance. Conclusion: AI-derived adiposity measures from CAC scans—especially liver fat—can identify adults without obesity or hyperglycemia at elevated risk for developing T2DM. These findings underscore the potential of AI-enabled opportunistic screening during CAC imaging to support early T2DM risk stratification in individuals not captured by current clinical guidelines.",

keywords = "AI-CVD, Artificial intelligence, Coronary artery calcium (CAC) scans, Opportunistic screening, Type 2 diabetes mellitus",

author = "Morteza Naghavi and Amir Azimi and Kyle Atlas and Reeves, \{Anthony P.\} and Chenyu Zhang and Jakob Wasserthal and Mirjalili, \{Seyed Reza\} and Mohammadhossein Mozafarybazargany and Ali Hashemi and Thomas Atlas and Henschke, \{Claudia I.\} and Yankelevitz, \{David F.\} and Mechanick, \{Jeffrey I.\} and Branch, \{Andrea D.\} and Fried, \{Susan K.\} and Khurram Nasir and Fayad, \{Zahi A.\} and McConnell, \{Michael V.\} and Rozemarijn Vliegenthart and Maron, \{David J.\} and Jagat Narula and Williams, \{Kim A.\} and Shah, \{Prediman K.\} and Budoff, \{Matthew J.\} and Daniel Levy and Benjamin, \{Emelia J.\} and Kloner, \{Robert A.\} and Wong, \{Nathan D.\}",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2025.",

year = "2025",

month = dec,

doi = "10.1186/s13098-025-01970-8",

language = "English (US)",

volume = "17",

journal = "Diabetology and Metabolic Syndrome",

issn = "1758-5996",

publisher = "BioMed Central",

number = "1",

}

TY - JOUR

T1 - Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia

T2 - insights from the AI-CVD study within MESA

AU - Naghavi, Morteza

AU - Azimi, Amir

AU - Atlas, Kyle

AU - Reeves, Anthony P.

AU - Zhang, Chenyu

AU - Wasserthal, Jakob

AU - Mirjalili, Seyed Reza

AU - Mozafarybazargany, Mohammadhossein

AU - Hashemi, Ali

AU - Atlas, Thomas

AU - Henschke, Claudia I.

AU - Yankelevitz, David F.

AU - Mechanick, Jeffrey I.

AU - Branch, Andrea D.

AU - Fried, Susan K.

AU - Nasir, Khurram

AU - Fayad, Zahi A.

AU - McConnell, Michael V.

AU - Vliegenthart, Rozemarijn

AU - Maron, David J.

AU - Narula, Jagat

AU - Williams, Kim A.

AU - Shah, Prediman K.

AU - Budoff, Matthew J.

AU - Levy, Daniel

AU - Benjamin, Emelia J.

AU - Kloner, Robert A.

AU - Wong, Nathan D.

PY - 2025/12

Y1 - 2025/12

N2 - Background: The AI-CVD initiative aims to maximize the value of coronary artery calcium (CAC) scans for cardiometabolic risk prediction by extracting opportunistic screening information. We investigated whether artificial intelligence (AI)-derived measures from CAC scans are associated with new-onset Type 2 diabetes mellitus (T2DM) in adults without obesity or hyperglycemia. Methods: Baseline CAC scans and up to 23 years of follow-up data were analyzed for participants without obesity (body mass index < 30 kg/m²) and hyperglycemia (fasting plasma glucose < 100 mg/dL) from the Multi-Ethnic Study of Atherosclerosis (MESA). AI-derived measures included liver attenuation index (LAI), subcutaneous fat index (SFI), total visceral fat index (TVFI), epicardial fat index (EFI), skeletal muscle index, and skeletal muscle mean density. Cox regression models compared highest vs. lowest quartiles of each AI-derived metric for T2DM risk. Multivariable models assessed adjusted predictive value using Wald chi-squared statistics. Subgroup analyses stratified participants by demographic and clinical factors. Results: During a median follow-up of 19.7 years among 2,993 participants (baseline mean age 61.9 ± 10.5 years, 53% women), 257 participants (8.6%) developed T2DM. Key predictors included LAI (HR: 3.13, 95% CI: 2.15–4.55), SFI (HR: 2.85, 95% CI: 1.93–4.21), TVFI (HR: 2.49, 95% CI: 1.72–3.60), and EFI (HR: 1.59, 95% CI: 1.09–2.32). LAI remained the most robust predictor after adjusting for all metrics (Wald χ² = 38.24). Subgroup analyses confirmed LAI’s consistent predictive performance. Conclusion: AI-derived adiposity measures from CAC scans—especially liver fat—can identify adults without obesity or hyperglycemia at elevated risk for developing T2DM. These findings underscore the potential of AI-enabled opportunistic screening during CAC imaging to support early T2DM risk stratification in individuals not captured by current clinical guidelines.

AB - Background: The AI-CVD initiative aims to maximize the value of coronary artery calcium (CAC) scans for cardiometabolic risk prediction by extracting opportunistic screening information. We investigated whether artificial intelligence (AI)-derived measures from CAC scans are associated with new-onset Type 2 diabetes mellitus (T2DM) in adults without obesity or hyperglycemia. Methods: Baseline CAC scans and up to 23 years of follow-up data were analyzed for participants without obesity (body mass index < 30 kg/m²) and hyperglycemia (fasting plasma glucose < 100 mg/dL) from the Multi-Ethnic Study of Atherosclerosis (MESA). AI-derived measures included liver attenuation index (LAI), subcutaneous fat index (SFI), total visceral fat index (TVFI), epicardial fat index (EFI), skeletal muscle index, and skeletal muscle mean density. Cox regression models compared highest vs. lowest quartiles of each AI-derived metric for T2DM risk. Multivariable models assessed adjusted predictive value using Wald chi-squared statistics. Subgroup analyses stratified participants by demographic and clinical factors. Results: During a median follow-up of 19.7 years among 2,993 participants (baseline mean age 61.9 ± 10.5 years, 53% women), 257 participants (8.6%) developed T2DM. Key predictors included LAI (HR: 3.13, 95% CI: 2.15–4.55), SFI (HR: 2.85, 95% CI: 1.93–4.21), TVFI (HR: 2.49, 95% CI: 1.72–3.60), and EFI (HR: 1.59, 95% CI: 1.09–2.32). LAI remained the most robust predictor after adjusting for all metrics (Wald χ² = 38.24). Subgroup analyses confirmed LAI’s consistent predictive performance. Conclusion: AI-derived adiposity measures from CAC scans—especially liver fat—can identify adults without obesity or hyperglycemia at elevated risk for developing T2DM. These findings underscore the potential of AI-enabled opportunistic screening during CAC imaging to support early T2DM risk stratification in individuals not captured by current clinical guidelines.

KW - AI-CVD

KW - Artificial intelligence

KW - Coronary artery calcium (CAC) scans

KW - Opportunistic screening

KW - Type 2 diabetes mellitus

UR - https://www.scopus.com/pages/publications/105021036456

UR - https://www.scopus.com/inward/citedby.url?scp=105021036456&partnerID=8YFLogxK

U2 - 10.1186/s13098-025-01970-8

DO - 10.1186/s13098-025-01970-8

M3 - Article

AN - SCOPUS:105021036456

SN - 1758-5996

VL - 17

JO - Diabetology and Metabolic Syndrome

JF - Diabetology and Metabolic Syndrome

IS - 1

M1 - 419

ER -

Opportunistic AI-derived adiposity measures from coronary artery calcium scans predict new-onset type 2 diabetes in adults without obesity or hyperglycemia: insights from the AI-CVD study within MESA

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this