Development and Validation of Deep Learning‐ and Ensemble Learning‐Based Biological Ages in the NHANES Study
Open accessAbstract
Introduction: Conventional machine learning (ML) approaches for constructing biological age (BA) have predominantly relied on blood-based markers, limiting their scope. This study aims to develop and validate novel ML-based BA models using a comprehensive set of clinical, behavioral, and socioeconomic factors and evaluate their predictive performance for mortality. Methods: We analyzed data from 24,985 participants in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2010, with follow-up extending to 31 December 2019, or until death or loss to follow-up. Thirty features, including blood and urine biochemistry, physical examination data, behavioral traits, and socioeconomic factors, were selected using the Least Absolute Shrinkage and Selection Operator (LASSO). These features were utilized to train deep neural networks (DNN) and ensemble learning models, specifically the Deep Biological Age (DBA) and Ensemble Biological Age (EnBA), with chronological age (CA) as the reference label. Model performance was assessed using mean absolute error (MAE), while interpretability was explored using Shapley Additive exPlanation (SHAP). Predictive accuracy of DBA and EnBA for mortality was compared with Phenotypic Age (PhenoAge) using the area under the curve (AUC) derived from Cox proportional hazards models and hazard ratios (HR), adjusted for demographics and lifestyle factors. Sensitivity analyses were performed to ensure robustness. Results: DBA and EnBA accurately predicted actual age (MAE = 2.98 and 3.58 years, respectively) and demonstrated strong predictive capability for all-cause mortality, with AUCs of 0.896 (95% CI: 0.891–0.898) for DBA and 0.889 (95% CI: 0.884–0.894) for EnBA. Higher DBA and EnBA accelerations were significantly associated with increased mortality risk (HR = 1.059 and 1.039, respectively). SHAP analysis highlighted prescription medication usage, hepatitis B surface antibody status, and vigorous physical activity as the most influential features contributing to DBA predictions. Furthermore, BA acceleration was linked to elevated risk of death from specific chronic conditions, including cardiovascular and cerebrovascular diseases and cancer. Discussion: Our study successfully developed and validated two ML-based BA models capable of accurately predicting both all-cause and cause-specific mortality. These findings suggest that the DBA and EnBA models hold promise for early identification of high-risk individuals, potentially facilitating timely preventive interventions and improving population health outcomes.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
ML models incorporating socioeconomic variables consistently outperform traditional clinical markers in predicting risk.
“ML models incorporating socioeconomic variables consistently outperform traditional clinical markers in predicting risk.”
From Artificial Intelligence and Machine Learning for Identifying Social Determinants of Health in Low‐Income Populations Within United States Health Systems: A Scoping Review (2026) · cites this paper as [15]
The Results (§3.2) report that adding behavioral and socioeconomic variables improved mortality-prediction AUC for both ML models compared with their biometric-only versions. But the Discussion (§4) reports that the full-feature DBA model performed slightly worse than PhenoAge V2, so the paper does not support 'consistently outperform' across traditional comparators. (LLM, conf=0.88)
Experimental — accuracy not yet measuredDeep learning models trained on behavioral and socioeconomic features show strong predictive performance for mortality outcomes.
“Deep learning models trained on behavioral and socioeconomic features show strong predictive performance for mortality outcomes [15].”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707 · see citing works
“Why Should I Trust You?”: Explaining the Predictions of Any Classifier
Ribeiro, M., Singh, S., & Guestrin, C. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations · doi:10.18653/v1/n16-3020 · Cited by 6,682 · see citing works
Hallmarks of aging: An expanding universe
López-Otín, C., Blasco, M. A., Partridge, L., Serrano, M., & Kroemer, G. (2023) · Cell · doi:10.1016/j.cell.2022.11.001 · Cited by 6,549 · see citing works
Ageing populations: the challenges ahead
Christensen, K., Doblhammer, G., Rau, R., & Vaupel, J. W. (2009) · The Lancet · doi:10.1016/s0140-6736(09)61460-4 · Cited by 3,764 · see citing works
An epigenetic biomarker of aging for lifespan and healthspan
Levine, M. E., Lu, A. T., Quach, A., Chen, B. H., Assimes, T. L., Bandinelli, S., Hou, L., Baccarelli, A. A., Stewart, J. D., Li, Y., et al. (2018) · Aging · doi:10.18632/aging.101414 · Cited by 3,675 · see citing works
Hepatitis B Virus Infection — Natural History and Clinical Consequences
Ganem, D., & Prince, A. M. (2004) · New England Journal of Medicine · doi:10.1056/nejmra031087 · Cited by 2,361 · see citing works
A survey on deep learning and its applications
Dong, S., Wang, P., & Abbas, K. (2021) · Computer Science Review · doi:10.1016/j.cosrev.2021.100379 · Cited by 1,573 · see citing works
Facing up to the global challenges of ageing
Partridge, L., Deelen, J., & Slagboom, P. E. (2018) · Nature · doi:10.1038/s41586-018-0457-8 · Cited by 1,475 · see citing works
Zhang, Y., Chen, C., Pan, X., Guo, J., Li, Y., Franco, O. H., Liu, G., & Pan, A. (2021) · BMJ · doi:10.1136/bmj.n604 · Cited by 933 · see citing works
A new approach to the concept and computation of biological age.
Klemera, P., & Ďoubal, S. (2006) · Mechanisms of Ageing and Development · doi:10.1016/j.mad.2005.10.004 · Cited by 727 · see citing works
Biomarkers of aging for the identification and evaluation of longevity interventions
Moqri, M., Herzog, C. M. S., Poganik, J. R., Justice, J. N., Belsky, D. W., Higgins‐Chen, A., Moskalev, A., Fuellen, G., Cohen, A. A., Bautmans, I., et al. (2023) · Cell · doi:10.1016/j.cell.2023.08.003 · Cited by 697 · see citing works
Prediction of the development of acute kidney injury following cardiac surgery by machine learning
Tseng, P.-Y., Chen, Y., Wang, C.-H., Chiu, K., Peng, Y.-S., Hsu, S., Chen, K.-L., Yang, C., & Lee, O. K. (2020) · Critical Care · doi:10.1186/s13054-020-03179-9 · Cited by 448 · see citing works
Kwon, D., & Belsky, D. W. (2021) · GeroScience · doi:10.1007/s11357-021-00480-5 · Cited by 410 · see citing works
Deep biomarkers of human aging: Application of deep neural networks to biomarker development
Putin, E., Mamoshina, P., Aliper, A. M., Korzinkin, M., Moskalev, A., Kolosov, A., Ostrovskiy, A., Cantor, C. R., Vijg, J., & Zhavoronkov, A. A. (2016) · Aging · doi:10.18632/aging.100968 · Cited by 373 · see citing works
Xue, B., Li, D., Lu, C., King, C. R., Wildes, T. S., Avidan, M. S., Kannampallil, T., & Abraham, J. (2021) · JAMA Network Open · doi:10.1001/jamanetworkopen.2021.2240 · Cited by 373 · see citing works
Epigenetic clock: A promising biomarker and practical tool in aging
Duan, R., Fu, Q., Sun, Y., & Li, Q. (2022) · Ageing Research Reviews · doi:10.1016/j.arr.2022.101743 · Cited by 353 · see citing works
Validation of biomarkers of aging
Moqri, M., Herzog, C. M. S., Poganik, J. R., Ying, K., Justice, J. N., Belsky, D. W., Higgins‐Chen, A., Chen, B. H., Cohen, A. A., Fuellen, G., et al. (2024) · Nature Medicine · doi:10.1038/s41591-023-02784-9 · Cited by 348 · see citing works
Baker, G. T., & Sprott, R. L. (1988) · Experimental Gerontology · doi:10.1016/0531-5565(88)90025-3 · Cited by 341 · see citing works
Stamatakis, E., Ahmadi, M. N., Gill, J. M. R., Thøgersen-Ntoumani, C., Gibala, M. J., Doherty, A., & Hamer, M. (2022) · Nature Medicine · doi:10.1038/s41591-022-02100-x · Cited by 336 · see citing works
Gao, X., Geng, T., Jiang, M., Huang, N., Zheng, Y., Belsky, D. W., & Huang, T. (2023) · Nature Communications · doi:10.1038/s41467-023-38013-7 · Cited by 306 · see citing works
Risk and Predictors of Mortality Associated With Chronic Hepatitis B Infection
Iloeje, U. H., Yang, H., Jen, C., Su, J., Wang, L., You, S., & Chen, C. (2007) · Clinical Gastroenterology and Hepatology · doi:10.1016/j.cgh.2007.06.015 · Cited by 231 · see citing works
Integrating the environmental and genetic architectures of aging and mortality
Argentieri, M. A., Amin, N., Nevado-Holgado, A. J., Sproviero, W., Collister, J. A., Keestra, S. M., Kuilman, M. M., Ginos, B. N. R., Ghanbari, M., Doherty, A., et al. (2025) · Nature Medicine · doi:10.1038/s41591-024-03483-9 · Cited by 211 · see citing works
Mamoshina, P., Kochetov, K., Lane, E., Cortese, F., Aliper, A. M., Lee, W., Ahn, S., Lee, U., Skjodt, N. M., Kovalchuk, O., et al. (2018) · The Journals of Gerontology Series A · doi:10.1093/gerona/gly005 · Cited by 198 · see citing works
Vigorous physical activity, incident heart disease, and cancer: how little is enough?
Ahmadi, M. N., Clare, P. J., Katzmarzyk, P. T., del Pozo Cruz, B., Lee, I. M., & Stamatakis, E. (2022) · European Heart Journal · doi:10.1093/eurheartj/ehac572 · Cited by 175 · see citing works
Clinical biomarker-based biological aging and risk of cancer in the UK Biobank
Mak, J. K. L., McMurran, C. E., Kuja‐Halkola, R., Hall, P. N., Czene, K., Jylhävä, J., & Hägg, S. (2023) · British Journal of Cancer · doi:10.1038/s41416-023-02288-w · Cited by 169 · see citing works
Dynamic and static maintenance of epigenetic memory in pluripotent and somatic cells
Shipony, Z., Mukamel, Z., Cohen, N. M., Landan, G., Chomsky, E., Zeliger, S. R., Fried, Y. C., Ainbinder, E., Friedman, N., & Tanay, A. (2014) · Nature · doi:10.1038/nature13458 · Cited by 159 · see citing works
DeepMAge: A Methylation Aging Clock Developed with Deep Learning
Galkin, F., Mamoshina, P., Kochetov, K., Sidorenko, D., & Zhavoronkov, A. A. (2021) · Aging and Disease · doi:10.14336/ad.2020.1202 · Cited by 153 · see citing works
Assessment of biological age by principal component analysis
Nakamura, E., Miyao, K., & Ozeki, T. (1988) · Mechanisms of Ageing and Development · doi:10.1016/0047-6374(88)90109-1 · Cited by 127 · see citing works
PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging
Bobrov, E., Georgievskaya, A., Kiselev, K., Sevastopolsky, A., Zhavoronkov, A. A., Gurov, S., Владимирович, Р. К., Tobar, M. D. P. B., Jaspers, S., & Clemann, S. (2018) · Aging · doi:10.18632/aging.101629 · Cited by 126 · see citing works
Epigenetic Age and the Risk of Incident Atrial Fibrillation
Roberts, J. D., Vittinghoff, E., Lu, A. T.-H., Alonso, Á., Wang, B., Sitlani, C. M., Mohammadi-Shemirani, P., Fornage, M., Kornej, J., Brody, J. A., et al. (2021) · Circulation · doi:10.1161/circulationaha.121.056456 · Cited by 121 · see citing works
Accelerated biological aging elevates the risk of cardiometabolic multimorbidity and mortality
Jiang, M., Tian, S., Liu, S., Wang, Y., Guo, X., Huang, T., Lin, X., Belsky, D. W., Baccarelli, A. A., & Gao, X. (2024) · Nature Cardiovascular Research · doi:10.1038/s44161-024-00438-8 · Cited by 117 · see citing works
Blood Biochemistry Analysis to Detect Smoking Status and Quantify Accelerated Aging in Smokers
Mamoshina, P., Kochetov, K., Cortese, F., Kovalchuk, A., Aliper, A. M., Lane, E., Scheibye‐Knudsen, M., Cantor, C. R., Skjodt, N. M., Kovalchuk, O., et al. (2019) · Scientific Reports · doi:10.1038/s41598-018-35704-w · Cited by 113 · see citing works
Association of Neighborhood Deprivation With Epigenetic Aging Using 4 Clock Metrics
Lawrence, K. G., Kresovich, J. K., O’Brien, K. M., Hoang, T. T., Xu, Z., Taylor, J. A., & Sandler, D. P. (2020) · JAMA Network Open · doi:10.1001/jamanetworkopen.2020.24329 · Cited by 100 · see citing works
Murabito, J. M., Zhao, Q., Larson, M. G., Rong, J., Lin, H., Benjamin, E. J., Levy, D. A., & Lunetta, K. L. (2017) · The Journals of Gerontology Series A · doi:10.1093/gerona/glx144 · Cited by 84 · see citing works
BoostTree and BoostForest for Ensemble Learning
Zhao, C., Wu, D., Huang, J., Yuan, Y., Zhang, H., Peng, R., & Shi, Z. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3227370 · Cited by 66 · see citing works
Gialluisi, A., Castelnuovo, A. F. D., Costanzo, S., Bonaccio, M., Persichillo, M., Magnacca, S., Curtis, A. D., Cerletti, C., Donati, M. B., de Gaetano, G., et al. (2021) · European Journal of Epidemiology · doi:10.1007/s10654-021-00797-7 · Cited by 56 · see citing works
Wang, C., Koutrakis, P., Gao, X., Baccarelli, A., & Schwartz, J. D. (2019) · Environmental Pollution · doi:10.1016/j.envpol.2019.113690 · Cited by 43 · see citing works
Epigenetic aging and perceived psychological stress in old age
Vetter, V. M., Drewelies, J., Sommerer, Y., Kalies, C. H., Regitz‐Zagrosek, V., Bertram, L., Gerstorf, D., & Demuth, I. (2022) · Translational Psychiatry · doi:10.1038/s41398-022-02181-9 · Cited by 40 · see citing works
Wang, G. C., Han, C. S., Detrick, B., Casolaro, V., Levine, D. M., Fried, L. P., & Walston, J. D. (2016) · Journal of the American Geriatrics Society · doi:10.1111/jgs.14090 · Cited by 26 · see citing works
Biological Age Acceleration and Motoric Cognitive Risk Syndrome
Sathyan, S., Ayers, E. I., Adhikari, D., Gao, T., Milman, S., Barzilai, N., & Verghese, J. (2023) · Annals of Neurology · doi:10.1002/ana.26624 · Cited by 24 · see citing works
Medication use and kidney cancer survival: A population‐based study
Nayan, M., Juurlink, D. N., Austin, P. C., Macdonald, E. M., Finelli, A., Kulkarni, G. S., & Hamilton, R. J. (2017) · International Journal of Cancer · doi:10.1002/ijc.31204 · Cited by 14 · see citing works
Wang, S., Wen, C. P., Li, W., Li, S., Sun, M., Xu, A., Tsai, M. K., Chu, D. T., Tsai, S. P., Tu, H., et al. (2022) · The Journals of Gerontology Series A · doi:10.1093/gerona/glac161 · Cited by 11 · see citing works
Self-citationClinical biomarker-based biological age predicts deaths in Brazilian adults: the ELSA-Brasil study
Machado, A. V., Silva, J. F. D. M. E., Colosimo, E. A., Needham, B. L., Maluf, C. B., Giatti, L., Camelo, L. D. V., & Barreto, S. M. (2024) · GeroScience · doi:10.1007/s11357-024-01186-0 · Cited by 8 · see citing works
Jia, Q., Chen, C., Xu, A., Wang, S., He, X., Shen, G., Luo, Y., Tu, H., Sun, T., & Wu, X. (2024) · iScience · doi:10.1016/j.isci.2024.108891 · Cited by 7 · see citing works
Cabral, D. F., Bigliassi, M., Cattaneo, G., Rundek, T., Pascual‐Leone, Á., Cahalin, L. P., & Gomes‐Osman, J. (2022) · Autonomic Neuroscience · doi:10.1016/j.autneu.2022.103023 · Cited by 5 · see citing works
Huang, S.-W., Li, X.-T., Chen, C., Ning, Q., & Huang, J. (2023) · Infectious Diseases and Therapy · doi:10.1007/s40121-023-00766-5 · Cited by 4 · see citing works
Machine Learning and Mathematical Models for Prediction of Structural Aging Process
Mudabbiruddin, M., & Mosavi, A. A. (2023) · IEEE International Symposium on Applied Computational Intelligence and Informatics (SACI) · doi:10.1109/saci58269.2023.10158652 · Cited by 4 · see citing works
Huang, Y., Da, L., Dong, Y., Li, Z., Liu, Y., Li, Z., Wu, X., & Li, W. (2025) · GeroScience · doi:10.1007/s11357-025-01577-x · Cited by 3 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Bafei, S. E. C., Xie, H., Yang, S., Sun, J., Liu, Y., Fan, Y., Tang, W., Liu, J., Chen, C., & Shen, C. (2025) · GeroScience · doi:10.1007/s11357-025-01866-5 · Cited by 2 · see citing works
Razzano, A. V. (2026) · Health Science Reports · doi:10.1002/hsr2.73083
Deep sequence learning with multi-task supervision for scalable population health monitoring
Ashraf, I., Nasir, I. M., Awais, M., Bouchelligua, W., Mansour, S., Alyounis, E., & Nawaz, M. (2026) · Frontiers in Public Health · doi:10.3389/fpubh.2026.1837769
- Authors
- Huang, Y., Yang, X., Wang, Q., Abula, A., Dong, Y., & Li, W.
- Affiliations
- Second Affiliated Hospital of Zhejiang University (CN)
- Corresponding author
- Li, Wenyuan
- Venue
- Frontiers in Aging Neuroscience
- Publisher
- Frontiers Media SA
- Published
- 2025-07-16
- Volume
- 17
- Pages
- 1532884-1532884
- ISSN
- 1663-4365 (ISSN-L)
- PMC
- PMC12307447
- OpenAlex
- W4412465853
- Cited by
- 3
- References
- 48
- Citations by year
- 2026: 2 · 2025: 1
- FWCI
- 1.20
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Frontiers in Aging Neuroscience (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Natural Science Foundation of China (82203984)
- Key Research and Development Program of Zhejiang Province
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
- Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
- Health Sciences › Medicine › Geriatrics and Gerontology › Frailty in Older Adults
- Keywords
- biological age, all-cause mortality, mortality prediction, deep neural network, ensemble learning, NHANES, blood and urine biomarkers, socioeconomic factors, prescription drug use, hepatitis B surface antibody, high-intensity exercise, phenotypic age, SHAP, cardiovascular mortality, cancer mortality, cerebrovascular disease mortality, accelerated aging, LASSO feature selection, model validation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
