Original title: NanoPyx: super-fast bioimage analysis powered by adaptive machine learning
Authors: Bruno M Saraiva,Inês M Cunha,António D Brito,Gautier Follain,Raquel Portela,Robert Haase,Pedro M Pereira,Guillaume Jacquemet,Ricardo Henriques
In this article, the authors introduce NanoPyx, a new bioimage analysis framework designed to handle large and complex microscopy datasets. The framework includes a unique component called the Liquid Engine, which is an agent-based machine-learning system that predicts strategies to accelerate image analysis tasks. Unlike traditional methods, the Liquid Engine generates multiple variations of CPU and GPU code using a meta-programming system. These variations are then benchmarked against each other to determine the most optimal performance under the user’s computational environment. The authors conducted experiments focusing on super-resolution analysis methods and found that the Liquid Engine improved computational speed by over 10 times. NanoPyx is made accessible to users through a Python library, code-free Jupyter notebooks, and a napari plugin, making it suitable for individuals with varying levels of coding proficiency. The principles of optimization used in the Liquid Engine also have broader implications for high-performance computing fields.
Original article: https://www.biorxiv.org/content/10.1101/2023.08.13.553080v2