Fluorescein videoangiography data analysis protocol for mapping retinal vascular permeability in humans

Area of Science:

• Ophthalmology
• Medical Imaging
• Biomedical Engineering

Background:

• Fluorescein video angiographies (FVAs) are crucial for diagnosing eye diseases like diabetic retinopathy (DR).
• Current methods use kinetic tracer models (DTKM) for biomarker extraction, requiring high-quality FVAs.
• Signal drops in FVAs due to patient movement degrade image quality, hindering accurate analysis.

Purpose of the Study:

• To develop a semi-automated program for removing low-quality frames from FVAs.
• To enhance FVA quality for reliable biomarker extraction using DTKM methods.
• To assess the effectiveness of the frame removal process on FVA quality.

Main Methods:

• Analyzed 15 FVAs from healthy, diabetes mellitus without DR (DMnoDR), and mild non-proliferative DR (NPDR) subjects.
• Developed a semi-automated algorithm to identify and remove frames with signal drops based on average signal analysis and prominence.
• Evaluated FVA quality post-processing using pre-DTKM and DTKM biomarker extraction.

Main Results:

• The semi-automated frame removal process successfully identified and flagged low-quality frames.
• Post-processing resulted in sufficient quality FVIs for subsequent analysis.
• The method demonstrated effectiveness in preparing FVAs for biomarker extraction.

Conclusions:

• A semi-automated frame removal technique can significantly improve FVA quality.
• This method facilitates more accurate biomarker extraction for eye diseases like DR.
• The developed program offers a valuable tool for clinical and research applications of FVA.