STAMP is a powerful AI-driven image analysis tool designed for pathology applications. It functions as a weakly-supervised deep-learning pipeline, allowing researchers to derive image-based biomarkers from whole-slide histopathology images without the need for pixel-level annotations. This tool is particularly useful in multi-center studies, providing a structured workflow that enhances collaboration between clinical researchers and machine-learning engineers. STAMP supports various tumor types and is backed by a peer-reviewed protocol, ensuring reliability and reproducibility in computational pathology projects.

The software is designed to be both beginner-friendly and suitable for advanced users, featuring a zero-code CLI for routine tasks alongside customizable options for in-depth research. With support for over 20 foundation models, STAMP enables multi-task learning, including classification, regression, and survival analysis. It also generates heatmaps for model transparency and offers built-in metrics for analysis. Overall, STAMP stands out as a comprehensive solution for those looking to leverage AI in pathology.