sporIDer

sporIDer is a software platform designed to assist researchers in identifying spores from microscopy images. It integrates image processing, machine learning, and annotation tools to support work across microbiology, mycology, and environmental biology. The platform aims to standardize spore identification, reduce manual workload, and enable reproducible analyses in both academic research and teaching laboratories.

Key components include a convolutional neural network–based classifier, support for multiple imaging modalities (brightfield, fluorescence, and polarized light), batch processing workflows, and an interactive labeling interface. Users can upload image datasets, annotate spores, and iteratively improve model performance through active learning. Outputs include spore class labels, per-object confidence scores, segmentation masks, and export options for CSV, JSON, or image-based results.

sporIDer emphasizes transparency and interoperability. Models and annotated datasets can be exported, and the software provides explainable AI tools to illuminate decisions. The project supports reproducibility through configurable preprocessing pipelines, versioned datasets, and integration with common bioimage-analysis workflows via scripts and APIs. It runs on desktop workstations and in HPC environments and is designed to be cross-platform.

sporIDer was developed by a collaborative team of researchers from universities and a nonprofit bioinformatics organization and released as an open-source project. The development prioritizes open data formats, modular architecture, and community contributions. Documentation, tutorials, and a user forum accompany the project to support adoption and collaboration.

It is intended for research use and educational purposes; clinical diagnostics require formal validation and regulatory clearance. Users should critically assess outputs, especially for novel spore morphologies or atypical imaging conditions.