OpenPhi

OpenPhi provides a standardized programmatic interface to access Philips iSyntax whole-slide images (WSIs) from the Philips Ultra Fast Scanner, enabling interoperable computational analysis and the development of deep-learning-based diagnostic and prognostic methods in digital pathology.

Key Features:

• iSyntax access: Provides programmatic read access to iSyntax-formatted whole-slide images produced by the Philips Ultra Fast Scanner.
• Philips scanner compatibility: Compatible with WSIs produced by the Philips Ultra Fast Scanner (iSyntax).
• Python implementation: Implemented in Python to facilitate integration into analysis pipelines.
• API for computational analysis: Exposes an API for integration into computational-image-analysis and deep-learning workflows for diagnostics and prognostics.
• Dependency on Philips SDK: Requires the Philips Software Development Kit (SDK) for full functionality and access to iSyntax content.
• Interoperability: Enables integration of iSyntax WSIs into vendor-neutral computational applications.

Scientific Applications:

• Deep learning for diagnostics: Supports training and deployment of deep-learning models on iSyntax WSIs for diagnostic tasks.
• Prognostic modeling: Enables development of prognostic algorithms using whole-slide image data.
• Computational pathology algorithm development: Facilitates algorithm development and evaluation on clinical-grade WSI data from the Philips Ultra Fast Scanner.
• Quantitative image analysis: Provides standardized access for quantitative feature extraction and image-based biomarker studies.

Methodology:

Implements a Python API that accesses iSyntax-formatted WSIs via the Philips Software Development Kit (SDK).