← All research

Cellular senescence & machine-learning biomarkers

Data collection
Data preparation
Model selection
Model training
Model evaluation
Model improvement
Prediction

The machine-learning workflow behind our senescence biomarkers — from data collection to prediction.

As cells age they can enter senescence — a durable state in which they permanently stop dividing yet remain metabolically active, secreting a cocktail of inflammatory signals that disturbs the surrounding tissue. These cells accumulate with age and drive a wide range of age-related pathologies, from fibrosis to neurodegeneration — but they have long been notoriously difficult to identify reliably and at scale.

We develop computational tools to change that. Using deep learning, we read the signatures of senescence and biological age directly from cell-nucleus morphology and routine imaging data, without the need for specialised staining. Trained across large image datasets, these models detect senescent cells, quantify how tissues age, and reveal how candidate interventions shift that trajectory over time.

Coupled with artificial intelligence for target identification and platforms such as mitodb.com for classifying disease mechanisms, these data-driven methods scale far beyond what manual analysis allows. Together they help move the field from simply cataloguing the hallmarks of aging toward measuring — and ultimately modulating — them in living systems, and toward therapies that clear or silence senescent cells.

Deep-learning pipeline detecting senescent cells from DAPI-stained nuclei
From a microscopy image of DAPI-stained nuclei, a deep neural network detects and segments individual nuclei, normalises each one for background, size and shape, and a second network reads the nuclear morphology to predict whether the cell is senescent.

Selected reading

  1. Heckenbach I, Mkrtchyan GV, Ezra MB, et al. & Scheibye-Knudsen M. Nuclear morphology is a deep learning biomarker of cellular senescence. Nature Aging (2022). doi:10.1038/s43587-022-00263-3
  2. Bakula D, Scheibye-Knudsen M. MitophAging: Mitophagy in aging and disease. Frontiers in Cell and Developmental Biology (2020). doi:10.3389/fcell.2020.00239
  3. Heckenbach I, Powell M, Fuller S, et al. & Scheibye-Knudsen M. Deep learning assessment of senescence-associated nuclear morphologies in mammary tissue predicts future risk of breast cancer. The Lancet Digital Health (2024). doi:10.1016/S2589-7500(24)00150-X

See the full list of the lab's publications on Google Scholar.