The authors of the open access paper linked here are training neural networks on large numbers of blood samples in an effort to produce biomarkers of aging. This is an interesting approach, primarily because it should in theory answer the question of whether a given data set - such as the data from blood tests - has any useful correlation with age. Biomarkers of biological age, how damaged an individual happens to be, are a necessary development in the field of aging research. At present the only reliable way to see how well a possible rejuvenation therapy works is to wait and see, which is slow and expensive in mice and out of the question in humans. What is needed is a quick measurement that accurately reflects biological age and thus remaining life expectancy. Given that, many more potential approaches to treating aging could be assessed and compared for a given level of funding and time, as the need to wait and see could be eliminated.
One of the major impediments in human aging research is the absence of a comprehensive and actionable set of biomarkers that may be targeted and measured to track the effectiveness of therapeutic interventions. In this study, we designed a modular ensemble of 21 deep neural networks (DNNs) of varying depth, structure and optimization to predict human chronological age using a basic blood test. To train the DNNs, we used over 60,000 samples from common blood biochemistry and cell count tests from routine health exams performed by a single laboratory and linked to chronological age and sex. To make this deep network ensemble available to the public, we placed our system online (at www.aging.ai), allowing any patient with blood test data to predict their age and sex.
The best performing DNN in the ensemble demonstrated 6.07 years mean absolute error in predicting chronological age within a 10 year frame, while the entire ensemble achieved 5.55 years mean absolute error. The analysis of relative feature importance within the DNNs helped deduce the most important features that may shed light on the contribution of these systems to the aging process, ranked in the following order: metabolic, liver, renal system and respiratory function. The five markers related to these functions were previously associated with aging and used to predict human biological age. Another interesting finding was the extraordinarily high importance of albumin, which primarily controls the oncotic pressure of blood. Albumin declines during aging and is associated with sarcopenia. The second marker by relative importance is glucose, which is directly linked to metabolic health. Cardiovascular diseases associated with diabetes mellitus are major causes of death within the general population. Current and future directions of this work include adding other sources of features including transcriptomic and metabolomics markers from blood, urine, individual organ biopsies and even imaging data as well as testing the system using data from patients with accelerated aging syndromes, multiple diseases and performing gender-specific analysis.