Caltech Unveils CellSAM AI for Biological Imaging

Researchers at Caltech have introduced CellSAM, a foundation AI model designed for universal biological imaging ^[1]. Published in Nature Methods, the tool automates the identification and segmentation of cells within complex datasets ^[1][2]. This technology, known as the Cell Segment Anything Model, allows for the tracking of millions of cells across various species ^[1]. The development addresses a significant bottleneck in biological research by reducing manual data processing ^[1]. Previously, scientists often spent extensive time manually defining cell boundaries for different experiments ^[2]. CellSAM adapts to various imaging modalities, enabling faster analysis without the need for constant retraining ^[1][2].