Machine Learning in Stem Cell Research: From Biological Data to Clinical Translation

Description

Anna Nicolaou leads a review of how machine learning is moving stem cell research from the bench toward the clinic. Stem cells promise disease models and replacement tissues, yet their behaviour varies, their maturation stalls, and quality checks strain under mountains of molecular, imaging, and clinical data. Algorithms that learn patterns in that data now touch every stage of the pipeline.

The review organises the field three ways: by data source, by model family, and by job. The jobs run from reprogramming adult cells back to stemness, through guided differentiation and maturation, to quality control, disease modelling, drug development, and bedside decisions. Transformer-based foundation models and methods that fuse several data types get special attention. The authors set criteria for choosing benchmark datasets, stressing biological diversity, methods, complete records, and public access. Their closing call is for models that clinicians can interpret and trust, not accurate black boxes.

Catalog record

Source: https://europepmc.org/article/MED/PMC13624364