Summary:
UCLA researchers in the Department of Anesthesiology have introduced a novel machine learning model that accurately predicts physiological aging by analyzing features extracted from patients’ arterial blood pressure (ABP) waveforms. This approach enables efficient, data-driven assessment of biological aging in clinical settings to facilitate timely health interventions.
Background:
While chronological age is a widely used benchmark for assessing health risks and predicting mortality, it often fails to account for significant variability in how individuals biologically age. Two people of the same chronological age may exhibit vastly different physiological conditions, rendering age-based clinical assessments insufficient for personalized care. To address this gap, biological age estimation methods have emerged as tools to better capture an individual’s true physiological state. Among these, epigenetic clocks have gained prominence. However, they typically rely on advanced genomic analyses that require costly equipment, trained personnel, and extended processing times, making them impractical for routine clinical use. Other available alternatives, such as vascular aging assessments, often depend on specialized cardiovascular imaging devices and are constrained in scope, limiting their scalability and broader clinical applicability. Consequently, there is a critical need for a more accurate, scalable, and clinically accessible approach to estimating physiological age—one that leverages data already routinely collected in many hospital settings. Such a method would not only improve personalized health risk profiling but also enable earlier interventions aimed at slowing or reversing the effects of aging.
Innovation:
Dr. Ravi Pal and his research team have developed a cutting-edge machine learning (ML) technology that predicts an individual's physiological age with high accuracy using features derived from arterial blood pressure (ABP) waveforms—data that is already commonly collected in clinical environments. Unlike traditional biological age assessments, this method is non-invasive, rapidly deployable, and compatible with existing monitoring infrastructure. Validated on data from over 17,000 patients, the model achieves a mean prediction deviation of less than five years, underscoring its robustness and clinical relevance across a diverse population. The scale and consistency of these results position the technology as a reliable and generalizable tool for real-world healthcare settings. This innovation has strong potential to enhance risk stratification, enable early detection of age-related conditions, and drive precision medicine strategies by integrating seamlessly into patient monitoring systems. Commercial partners in digital health, diagnostics, wearables, and clinical decision support tools may find valuable opportunities to integrate or license this technology as part of a next-generation aging and wellness platform.
.png)
Credit: Dr. Ravi Pal
Potential Applications:
● Continuous monitoring of healthy aging trajectories
● Integration into patient monitoring systems
● Early detection of physiological decline
● Personalized risk profiling
● Preventive medicine
● Wearable device integration
● Holistic healthcare monitoring0
Advantages:
● Accurate, real-time estimation of physiological age
● Eliminates need for specialized lab testing
● Applicable to non-cardiovascular patients
● Scalable and compatible with existing clinical workflows
● Machine learning enhanced optimization and adaptability
State of Development:
Successful demonstration completed 02/27/25.
Related Papers:
1. Jylhava J, Pedersen NL, Hagg S. Biological Age Predictors. EBioMedicine. 2017; 21:29-36.
2. Sun ED, Qian Y, Oppong R, Butler TJ, Zhao J, Chen BH, Tanaka T, Kang J, Sidore C, Cucca F, Bandinelli S, Abecasis GR, Gorospe M, Ferrucci L, Schlessinger D, Goldberg I, Ding J. Predicting physiological aging rates from a range of quantitative traits using machine learning. Aging (Albany NY). 2021 Oct 29;13(20):23471-23516.
3. Horvath S, Raj K. DNA methylation-based biomarkers and the epigenetic clock theory of ageing. Nat Rev Genet 2018; 19, 371–384.
4. Bernard D, Doumard E, Ader I, Kemoun P, Pages JC, Galinier A, Cussat- Blanc S, Furger F, Ferrucci L, Aligon J, Delpierre, C, Penicaud L, Monsarrat P, Casteilla L. Explainable machine learning framework to predict personalized physiological aging. Aging Cell. 2023: e13872.
5. Kim S, Kwon S, Rudas A, Pal R, Markey MK, Bovik AC, Cannesson M. Machine Learning of Physiologic Waveforms and Electronic Health Record Data: A Large Perioperative Data Set of High-Fidelity Physiologic Waveforms. Crit Care Clin. 2023.
6. Pal R, Rudas A, Kim S, Chiang JN, Cannesson M. A signal processing tool for extracting features from arterial blood pressure and photoplethysmography waveforms. 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 2024, pp. 1-5.
7. Attia ZI, Friedman PA, Noseworthy PA, Lopez-Jimenez F, Ladewig DJ, Satam G, Pellikka PA, Munger TM, Asirvatham SJ, Scott CG, Carter RE, Kapa S. Age and Sex Estimation Using Artificial Intelligence From Standard 12-Lead ECGs. Circ Arrhythm Electrophysiol. 2019 Sep;12(9):e007284.
8. Wang Z, Li L, Glicksberg BS, Israel A, Dudley JT, Ma'ayan A. Predicting age by mining electronic medical records with deep learning characterizes differences between chronological and physiological age. J Biomed Inform. 2017 Dec;76:59-68.
9. Pal R, Rudas A, Kim S, Chiang JN, Cannesson M. A signal processing tool for extracting features from arterial blood pressure and photoplethysmography waveforms. 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Orlando, FL, USA, 2024, pp. 1-5. https://ucla.technologypublisher.com/technology/53910
Reference:
UCLA Case No. 2025-264
Lead Inventor:
Maxime Cannesson, Ravi Pal