Extracting biological age from biomedical data via deep learning: too much of a good thing?

Abstract

Aging-related physiological changes are systemic and, at least in humans, are linearly associated with age. Therefore, linear combinations of physiological measures trained to estimate chronological age have recently emerged as a practical way to quantify aging in the form of biological age. Aging acceleration, defined as the difference between the predicted and chronological age was found to be elevated in patients with major diseases and is predictive of mortality. In this work, we compare three increasingly accurate biological age models: metrics derived from unsupervised Principal Components Analysis (PCA), alongside two supervised biological age models; a multivariate linear regression and a state-of-the-art deep convolution neural network (CNN). All predictions were made using one-week long locomotor activity records from a 2003-2006 National Health and Nutrition Examination Survey (NHANES) dataset. We found that application of the supervised approaches improves the accuracy of the chronological age estimation at the expense of a loss of the association between the aging acceleration predicted by the model and all-cause mortality. Instead, we turned to the NHANES death register and introduced a novel way to train parametric proportional hazards models in a form suitable for out-of-the-box implementation with any modern machine learning software. Finally, we characterized a proof-of-concept example, a separate deep CNN trained to predict mortality risks that outperformed any of the biological age or simple linear proportional hazards models. Our findings demonstrate the emerging potential of combined wearable sensors and deep learning technologies for applications involving continuous health risk monitoring and real-time feedback to patients and care providers.

Determination of the phenotypic age in residents of Mexico City: effect of accelerated ageing on lung function and structure

“…In this context, numerous studies are indicating that ageing contributes to pathogenesis of several chronic lung diseases, some of them virtually exclusive to older adults, whereas others may occur at any age, usually with a worse prognosis in adults over 65 years [23]. However, these studies have been carried out considering chronological age, which offers limited information regarding the complex processes driving biological ageing [9,24]. Thus, it is the phenotypic age, usually influenced by a complex combination of genetic, lifestyle, and environmental factors that plays an important role in susceptibility to morbidity and mortality [10,25].…”

Self-supervised Pretraining and Transfer Learning Enable Flu and COVID-19 Predictions in Small Mobile Sensing Datasets

“…To answer this question, we removed the transformer blocks from our model and passed the CNN's final output directly to a linear layer. 1D CNNs are frequently used in timeseries classification [27,52], and have been applied to data from wearable devices before [36,49,56] (Delong). Note that while substantial class imbalance precludes statistically significant results on some tasks ("Flu Positivity", "Severe Fever", and "Severe Cough"), Our Model performs better than all baselines and ablations when jointly evaluating performance across all tasks to increase statistical power (Figure 2).…”

Paleodemography: From archaeology and skeletal age estimation to life in the past

“…Although for blood alone the clock's correlation with age was lower than what was reported in the original study, it still predicts mortality better than a variety of environmental and individual characteristics associated with increased risks of dying (Marioni et al, 2015). Other biological estimators exist, including a model based on physical activity where there was a correlation of 0.75 with chronological age (Pyrkov et al, 2018). For paleodemographers, these estimates provide a level of accuracy that one might strive to achieve with skeletal markers of age, as long as reasonably good morphological features can be identified and analyzed appropriately.…”