Predicting age by mining electronic medical records with deep learning characterizes differences between chronological and physiological age

Abstract

Determining the discrepancy between chronological and physiological age of patients is central to preventative and personalized care. Electronic medical records (EMR) provide rich information about the patient physiological state, but it is unclear whether such information can be predictive of chronological age. Here we present a deep learning model the uses vital signs and lab tests contained within the EMR of Mount Sinai Health System (MSHS) to predict chronological age. The model is trained on 377,686 EMR from patients of ages 18–85 years old. The discrepancy between the predicted and recorded chronological age is then used as a proxy to estimate physiological age. Overall, the model can predict the chronological age of patients with a standard deviation error of ~7 years. The ages of the youngest and oldest patients were more accurately predicted, while patients of ages ranging between 40 and 60 years were the least accurately predicted. Patients with the largest discrepancy between their physiological and chronological age were further inspected. The patients predicted to be significantly older than their chronological age have higher systolic blood pressure, higher cholesterol, damaged liver, and anemia. In contrast, patients predicted to be younger than their chronological age have lower blood pressure and shorter stature among other indicators; both groups display lower weight than the population average. Using information from ~10,000 patients from the entire cohort who have been also profiled with SNP arrays, genome-wide association study (GWAS) uncover several novel genetic variants associated with aging. In particular, significant variants were mapped to genes known to be associated with inflammation, hypertension, lipid metabolism, height, and increased lifespan in mice. Several genes with missense mutations were identified as novel candidate aging genes. In conclusion, we demonstrate how EMR data can be used to assess overall health via a scale that is based on deviation from the patient’s predicted chronological age.

Prediction of biological age and all-cause mortality by 12-lead electrocardiogram in patients without structural heart disease

“…There have been a number of recent investigations regarding the prediction of BA using medical records, vital signs and laboratory data, 14 or epigenetic changes. 15 These investigations indicated a gap between predicted BA and actual CA, and the gap was deemed to represent epigenetic age acceleration, because it was shown to be associated with higher risks of all-cause mortality, 16,17 cardiovascular disease, 15,18 and cross-sectionally with obesity, 19 earlier menopause, 20 and frailty.…”

Predicting opioid dependence from electronic health records with machine learning

“…EHR data include demographics, diagnoses, laboratory tests, vital signs, clinical notes, prescriptions, and procedures data. Examples of previous predictive studies that utilized EHR systems implementing machine learning methods include predicting the incidence of cardiovascular disease in patients with severe schizophrenia, bipolar disorder, or other non-organic psychosis [15]; length of hospital stay and time to readmission based on Research Domain Criteria in psychiatric patients [16]; unplanned readmission after discharge [17]; in-hospital mortality [18]; patient physiological age [19]; and many more. Here we describe the application of a machine learning classifier to predict substance dependence based on lab tests and vital signs using patient data derived from the Mount Sinai Medical Center (MSMC) EHR system.…”

The Kidney–Gut–Muscle Axis in End-Stage Renal Disease is Similarly Represented in Older Adults

“…Similarly, a kidney-gut-muscle axis may be present in older adults that do not have ESRD. Reductions in muscle mass and physical function, and poor muscle composition are prevalent in older adults [44,45], and kidney function decreases during aging from average eGFR values of 120 mL/min/1.73 m 2 in 20-year-olds to less than 60 mL/min/1.73 m 2 in adults older than 70 y [46]. Correspondingly, circulating levels of urea (i.e., blood urea nitrogen) increase during aging [46], which may provide the stimulus for an increase in urea-degrading intestinal bacteria, while impairing intestinal barrier function.…”

“…Reductions in muscle mass and physical function, and poor muscle composition are prevalent in older adults [44,45], and kidney function decreases during aging from average eGFR values of 120 mL/min/1.73 m 2 in 20-year-olds to less than 60 mL/min/1.73 m 2 in adults older than 70 y [46]. Correspondingly, circulating levels of urea (i.e., blood urea nitrogen) increase during aging [46], which may provide the stimulus for an increase in urea-degrading intestinal bacteria, while impairing intestinal barrier function. In support of this, Enterobacteriaceae are higher in subjects older than 60 y, when compared with younger subjects (19-45 y) [47][48][49], an important finding because Enterobacteriaceae are associated with frailty, a reduced percentage of whole body lean mass, and worse physical function in older adults [50][51][52].…”