Explainable machine learning framework to predict personalized physiological aging

Abstract

Attaining personalized healthy aging requires accurate monitoring of physiological changes and identifying subclinical markers that predict accelerated or delayed aging. Classic biostatistical methods most rely on supervised variables to estimate physiological aging and do not capture the full complexity of inter‐parameter interactions. Machine learning (ML) is promising, but its black box nature eludes direct understanding, substantially limiting physician confidence and clinical usage. Using a broad population dataset from the National Health and Nutrition Examination Survey (NHANES) study including routine biological variables and after selection of XGBoost as the most appropriate algorithm, we created an innovative explainable ML framework to determine a Personalized physiological age (PPA). PPA predicted both chronic disease and mortality independently of chronological age. Twenty‐six variables were sufficient to predict PPA. Using SHapley Additive exPlanations (SHAP), we implemented a precise quantitative associated metric for each variable explaining physiological (i.e., accelerated or delayed) deviations from age‐specific normative data. Among the variables, glycated hemoglobin (HbA1c) displays a major relative weight in the estimation of PPA. Finally, clustering profiles of identical contextualized explanations reveal different aging trajectories opening opportunities to specific clinical follow‐up. These data show that PPA is a robust, quantitative and explainable ML‐based metric that monitors personalized health status. Our approach also provides a complete framework applicable to different datasets or variables, allowing precision physiological age estimation.

Interpretable machine learning framework to predict gout associated with dietary fiber and triglyceride-glucose index

“…This was a cross-sectional study, and all data for this study were obtained from the Centers for Disease Control and Prevention NHANES dataset for 7 cycles from 2005 to 2018. Demographic, lifestyle, anthropometric, laboratory analysis, questionnaire interview, and dietary data from each NHANES cycle were merged according to SEQN (participant ID) using the Bernard D et al method to generate a participant dataset containing all study variables [ 17 ]. The NHANES study was approved by the National Center for Health Statistics (NCHS) Ethics Review Board under approval no.…”

Association of red cell distribution width to albumin ratio with risk of all-cause and cause-specific mortality: two prospective cohort studies

“…Additionally, RDW is positively correlated with chronological age, while serum albumin is negatively correlated with chronological age[17]. RDW and albumin were enrolled in the construction of phenotypic and biological age, and contributed a lot to individual variances of aging pace[18–20].…”

Ratio of Red Blood Cell Distribution Width to Albumin Level and Risk of Mortality

“… 14 , 15 , 16 Additionally, RDW is positively correlated with chronological age, while serum albumin is negatively correlated with chronological age, 17 and RDW and albumin are associated with phenotypic and biological age and contribute greatly to individual variances in aging pace. 18 , 19 , 20 …”

“…[14][15][16] Additionally, RDW is positively correlated with chronological age, while serum albumin is negatively correlated with chronological age, 17 and RDW and albumin are associated with phenotypic and biological age and contribute greatly to individual variances in aging pace. [18][19][20] Both RDW and the albumin concentration have been suggested as integrative biomarkers for a multidimensional dysfunctional physiological status that relates to inflammation, oxidative stress, and nutrition, [10][11][12] but they represent these pathological aspects from different perspectives. Therefore, considering their essential roles in physiological function, disease, and assessment of aging, the integration of these 2 markers may be valuable for predicting mortality.…”

“…Several studies have reported that the serum albumin concentration is inversely associated with the incidence of dysfunction, disease, and mortality . Additionally, RDW is positively correlated with chronological age, while serum albumin is negatively correlated with chronological age, and RDW and albumin are associated with phenotypic and biological age and contribute greatly to individual variances in aging pace …”