ilastik: interactive machine learning for (bio)image analysis

ilastik is an open-source software project focused on interactive machine learning for bioimage analysis, developed and maintained by a research community whose work is documented through the project's publications page. Its flagship and core offering, ilastik, is a simple, user-friendly tool designed for interactive image classification, segmentation, and analysis, enabling researchers to process large and complex image datasets without requiring expertise in programming or machine learning. The software follows a supervised, interactive workflow: users annotate a small number of representative image regions, and the tool applies machine learning to classify pixels or objects across the full dataset in real time, allowing immediate visual feedback and iterative refinement. Typical application areas include pixel classification for identifying structures in microscopy images, object classification for distinguishing cell types or other biological entities, image segmentation for separating touching or overlapping objects, and tracking of objects over time in time-lapse experiments. These capabilities make the product particularly relevant to the life sciences, including fields such as cell biology, neuroscience, and developmental biology, where high-throughput imaging generates volumes of data that would be impractical to analyze manually. The tool can also handle multidimensional data, including multi-channel, three-dimensional, and time-series images, and is designed to scale to large datasets that exceed available memory. Use cases range from counting and characterizing cells in tissue samples to segmenting neurons in electron microscopy volumes and quantifying experimental outcomes in screening studies. By combining an accessible graphical interface with robust machine learning methods, the publisher's single product addresses a well-defined niche: making interactive, learning-based image analysis practical for biologists and other scientists working with image data.