Dataset-Agnostic Biological Age Calculator and Predictions of Mortality in NHANES (1999-2010), UK Biobank (2006-2010), and Dalian Heart Failure (2015-2023) Cohorts: Development and Validation Study.
5 affiliations
- 1Cardiovascular Analytics Group, Kowloon, China (Hong Kong).
- 2School of Nursing and Health Sciences, Hong Kong Metropolitan University, 1 Sheung Shing Street, Ho Man Tin, Kowloon, China (Hong Kong), +852 3970 2942.
- 3Department of Biomedical Engineering and Health Sciences, McMaster University, Hamilton, ON, Canada.
- 4School of Life Sciences, Chinese University of Hong Kong, Shatin, China (Hong Kong).
- 5Tianjin Key Laboratory of Ionic-Molecular Function of Cardiovascular disease, Department of Cardiology, Tianjin Institute of Cardiology, Second Hospital of Tianjin Medical University, Tianjin, China.
Abstract
Background
A positive difference between biological age (BA) and chronological age indicates accelerated aging. The application of existing BA models is limited by high technical demands, a lack of alignment across diverse populations, and difficulty in updating them to accommodate expanding cohorts.Objective
To address the issues in existing tools, this project aims to develop and validate an automated toolkit for adaptive BA estimation.Methods
The proposed toolkit, named BACalc, facilitates the processing of raw clinical datasets stored in CSV format. Within the graphical user interface, users can select (1) the chronological age column, (2) binary variables (eg, sex, smoking, and alcohol use), and (3) continuous variables (eg, creatinine levels) for BA estimation, using the principal component analysis (PCA) method with the Dubina correction or the Klemera-Doubal method (KDM). All numerical operations are implemented using the NumPy and Pandas libraries, and linear modeling components are provided by scikit-learn and SciPy. The program generates three main output files: (1) the row-wise predictions, (2) the regression statistics, and (3) the explicit linear equations and PCA loadings. Example outputs were generated using the National Health and Nutrition Examination Survey (NHANES, 1999-2010), UK Biobank (2006-2010), and Dalian Heart Failure (2015-2023). The results were compared with the reference BioAge R package.Results
Using the KDM model, the BACalc-derived and BioAge-derived estimates exhibit a near-perfect correlation (r=0.9937, mean absolute error 2.02 y). The PCA-Dubina model also shows a strong correlation with BioAge (r=0.8871), within a ±20-year window of accelerated aging residuals. In the UK Biobank, a 1-year increase in PCA-Dubina-derived age acceleration was associated with a 1.77% higher hazard of all-cause mortality (hazard ratio [HR] 1.018, 95% CI 1.017-1.018; P<.001). Similarly, KDM-derived age acceleration was associated with a 1.59% higher hazard of mortality per year of accelerated aging (HR 1.016, 95% CI 1.016-1.016; P<.001). In the Dalian cohort, a 1-year increase in age acceleration derived from the PCA-Dubina model (BAc) independently predicted mortality with a 2.84% increased hazard per year (HR 1.028, 95% CI 1.022-1.035; P<.001). Similarly, KDM-derived age acceleration (BA_EC) was associated with a 1.26% increased hazard of mortality per year of accelerated aging (HR 1.013, 95% CI 1.007-1.018; P<.001).Conclusions
BACalc provides a reliable platform for BA estimation, consistent with previously validated computational frameworks and convenient for real-time updating. BA acceleration predicted mortality in 2 independent cohorts.
Related
- Development and validation of an interpretable machine learning model for early prediction of mortality in ICU patients with chronic obstructive pulmonary disease and heart failure
- Development and validation of a novel metabolic-inflammatory health-related nomogram for predicting in-hospital mortality in older patients with type 2 diabetes mellitus combined with cardiovascular disease
- Admission-time prediction of 7-day mortality in clinician-attributed hypertensive intracerebral hemorrhage: development and external validation against the ICH score
- Development and validation of an interpretable machine learning model for predicting in-hospital mortality among critically ill patients with liver cirrhosis
- Development and validation of a nomogram for predicting 28-day mortality in critically ill liver cirrhotic patients with coagulopathy
BibTeX
@article{liu2026datasetagnostic,
title = {{Dataset-Agnostic Biological Age Calculator and Predictions of Mortality in NHANES (1999-2010), UK Biobank (2006-2010), and Dalian Heart Failure (2015-2023) Cohorts: Development and Validation Study.}},
author = {Haipeng Liu and Angelina Amabi Perez Aranda and Anselm Au and Derek Wu and Haixin Zhou and Wing Tak Wong and Tong Liu and Mehrdad Shahmohammadi Beni and Gary Tse},
journal = {JMIR formative research},
year = {2026},
month = oct,
doi = {10.2196/95321},
url = {https://europepmc.org/article/MED/42954807},
note = {CC BY},
}
RIS
TY - JOUR
TI - Dataset-Agnostic Biological Age Calculator and Predictions of Mortality in NHANES (1999-2010), UK Biobank (2006-2010), and Dalian Heart Failure (2015-2023) Cohorts: Development and Validation Study.
AU - Haipeng Liu
AU - Angelina Amabi Perez Aranda
AU - Anselm Au
AU - Derek Wu
AU - Haixin Zhou
AU - Wing Tak Wong
AU - Tong Liu
AU - Mehrdad Shahmohammadi Beni
AU - Gary Tse
PY - 2026
DA - 2026/10/20
JO - JMIR formative research
DO - 10.2196/95321
UR - https://europepmc.org/article/MED/42954807
AB - Background A positive difference between biological age (BA) and chronological age indicates accelerated aging. The application of existing BA models is limited by high technical demands, a lack of alignment across diverse populations, and difficulty in updating them to accommodate expanding cohorts.Objective To address the issues in existing tools, this project aims to develop and validate an automated toolkit for adaptive BA estimation.Methods The proposed toolkit, named BACalc, facilitates the processing of raw clinical datasets stored in CSV format. Within the graphical user interface, users can select (1) the chronological age column, (2) binary variables (eg, sex, smoking, and alcohol use), and (3) continuous variables (eg, creatinine levels) for BA estimation, using the principal component analysis (PCA) method with the Dubina correction or the Klemera-Doubal method (KDM). All numerical operations are implemented using the NumPy and Pandas libraries, and linear modeling components are provided by scikit-learn and SciPy. The program generates three main output files: (1) the row-wise predictions, (2) the regression statistics, and (3) the explicit linear equations and PCA loadings. Example outputs were generated using the National Health and Nutrition Examination Survey (NHANES, 1999-2010), UK Biobank (2006-2010), and Dalian Heart Failure (2015-2023). The results were compared with the reference BioAge R package.Results Using the KDM model, the BACalc-derived and BioAge-derived estimates exhibit a near-perfect correlation (r=0.9937, mean absolute error 2.02 y). The PCA-Dubina model also shows a strong correlation with BioAge (r=0.8871), within a ±20-year window of accelerated aging residuals. In the UK Biobank, a 1-year increase in PCA-Dubina-derived age acceleration was associated with a 1.77% higher hazard of all-cause mortality (hazard ratio [HR] 1.018, 95% CI 1.017-1.018; P<.001). Similarly, KDM-derived age acceleration was associated with a 1.59% higher hazard of mortality per year of accelerated aging (HR 1.016, 95% CI 1.016-1.016; P<.001). In the Dalian cohort, a 1-year increase in age acceleration derived from the PCA-Dubina model (BAc) independently predicted mortality with a 2.84% increased hazard per year (HR 1.028, 95% CI 1.022-1.035; P<.001). Similarly, KDM-derived age acceleration (BA_EC) was associated with a 1.26% increased hazard of mortality per year of accelerated aging (HR 1.013, 95% CI 1.007-1.018; P<.001).Conclusions BACalc provides a reliable platform for BA estimation, consistent with previously validated computational frameworks and convenient for real-time updating. BA acceleration predicted mortality in 2 independent cohorts.
SN - 2561-326X
ER -