logo
ResearchBunny Logo
Unsupervised learning of aging principles from longitudinal data

Medicine and Health

Unsupervised learning of aging principles from longitudinal data

K. Avchaciov, M. P. Antoch, et al.

Discover the groundbreaking research from Konstantin Avchaciov, Marina P. Antoch, and their colleagues that unveils a revolutionary 'dynamic frailty indicator' (dFI) using advanced machine learning. This study reveals how dFI can predict lifespan and respond to both life-shortening and life-extending interventions, providing a crucial new marker for biological age.

00:00
00:00
Playback language: English
Abstract
Age is the leading risk factor for prevalent diseases and death. This study combined analytical and machine learning tools to describe the aging process using longitudinal measurements. A deep artificial neural network, incorporating auto-encoder and auto-regression components, generated a "dynamic frailty indicator" (dFI). In mouse blood tests, dFI increased exponentially, predicted remaining lifespan, and correlated with aging hallmarks. It responded to life-shortening and life-extending treatments, suggesting its utility as a biological age marker.
Publisher
Nature Communications
Published On
Nov 01, 2022
Authors
Konstantin Avchaciov, Marina P. Antoch, Ekaterina L. Andrianova, Andrei E. Tarkhov, Leonid I. Menshikov, Olga Burmistrova, Andrei V. Gudkov, Peter O. Fedichev
Tags
aging
dynamic frailty indicator
machine learning
lifespan prediction
biological age
health markers
longitudinal measurements

Related Publications

Explore these studies to deepen your understanding of the subject.

Deep learning from "passive feeding" to "selective eating" of real-world data
Medicine and Health

Deep learning from "passive feeding" to "selective eating" of real-world data

Z. Li, C. Guo, et al.

Learning interpretable dynamics of stochastic complex systems from experimental data
Biology

Learning interpretable dynamics of stochastic complex systems from experimental data

T. Gao, B. Barzel, et al.

Extracting structural motifs from pair distribution function data of nanostructures using explainable machine learning
Chemistry

Extracting structural motifs from pair distribution function data of nanostructures using explainable machine learning

A. S. Anker, E. T. S. Kjær, et al.

Reliability of Supervised Machine Learning Using Synthetic Data in Health Care: Model to Preserve Privacy for Data Sharing
Computer Science

Reliability of Supervised Machine Learning Using Synthetic Data in Health Care: Model to Preserve Privacy for Data Sharing

D. Rankin, M. Black, et al.

Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny