Orbit Image Analysis: An open-source whole slide image analysis tool.

Area of Science:

• Digital Pathology
• Computational Biology
• Medical Image Analysis

Background:

• Whole slide imaging generates large datasets requiring advanced analysis tools.
• Existing tools may lack flexibility, scalability, or machine learning capabilities for complex tasks.

Purpose of the Study:

• To introduce Orbit Image Analysis, an open-source tool for sophisticated whole slide imaging.
• To demonstrate Orbit's capabilities in handling complex tissue analyses and large datasets.

Main Methods:

• Orbit Image Analysis utilizes a tile-based map-reduce framework for generic algorithm execution.
• It incorporates machine learning for deep learning segmentation and complex object detection.
• The tool supports local deployment or integration with OMERO and leverages Apache Spark for distributed computing.

Main Results:

• Orbit Image Analysis successfully performed complex analyses in three real-world applications.
• Quantification of idiopathic lung fibrosis was achieved.
• Nerve fibre density quantification was performed.
• Glomeruli detection in kidney tissue was successfully implemented.

Conclusions:

• Orbit Image Analysis provides a flexible, scalable, and powerful open-source solution for whole slide imaging.
• Its machine learning and distributed computing features enable advanced analysis of intricate biological samples.
• The tool is applicable to diverse fields within digital pathology and biomedical research.