Purpose of Review: Machine learning (ML) enables high-throughput analysis of multimodal data generated from stem cell experiments such as gene expression data, images of cells, or proteomic data. In this review, we analyse the progression of ML adaptation in advancing the field of stem cell research. Recent Findings: On the one hand, the field of stem cell phenotypic characterisation is experiencing a significant growth, largely due to the successful implementation of deep networks in domains with similar problem characteristics (i.e., rapid advances of the image recognition field). On the other hand, genotypic characterisation is gradually gaining traction as researchers are beginning to apply ML to understand the genetic and molecular mechanisms behind stem cell behaviour. Summary: The use of advanced machine learning techniques, such as deep networks, is demonstrating promising results in phenotypic stem cell characterisation, although it is still lagging slightly in genotypic characterisation. Despite this progress, significant challenges persist, including ensuring the interpretability of ML models, limited availability of annotated datasets, improving the accuracy and quality of training data, and navigating ethical considerations.
