Radiomics-Based Assessment of OCT Angiography Images for Identifying Vascular Leakage through Quantitative Vessel Characteristics in Diabetic Retinopathy Patients
Nasser Shoeibi1 *, Esmat Ramezanzadeh2 , Hoda Zare2 , Mohammad Hossein Bahreyni-Toosi2 , Mehrdad Motamed Shariati1 , Hamed Tabesh3 , Hossein Rabbani4 , Mohammad Hossein Moattar5 , Maryam Dourandish6
- Eye Research Center, Mashhad University of Medical Sciences, Mashhad, Iran
- Department of Medical Physics, Faculty of Medicine, Mashhad University of Medical Sciences, Mashhad, Iran
- Department of Health Information Technology, Ferdows Faculty of Medical Sciences, Birjand, Iran
- Medical Image and Signal Processing Research Center, Isfahan University of Medical Sciences, Isfahan, Iran
- Department of Computer Engineering, Mashhad Branch, Islamic Azad University, Mashhad, Iran
- Gilan University of Medical Sciences, Rasht, Iran
Abstract: To utilize radiomics analysis to extract quantitative vessel characteristics from OCTA images to facilitate the detection of leakage sources that are otherwise undetectable.
Methods: A total of 87 DR patients were prospectively evaluated between 2022 and 2023. Radiomics features—including vessel density, branch number, tortuosity, texture analysis, vessel pixel count, box fractal dimension, wavelet fractal dimension, vessel length, branch points, perfusion density, vessel intensity, vessel diameter, perfused capillary density, and signal voids—were extracted using advanced image processing techniques. These quantitative characteristics were then analyzed to identify patterns associated with vascular leakage.
Results: Preliminary findings indicate that specific radiomic features correlate with the presence of vascular leakage. Notably, certain extracted features effectively highlight leakage-prone areas not readily visible in standard OCTA images. Vessel connectivity, perfusion density, signal voids and width variation are among the most impressive features.
Conclusion: This study underscores the potential of radiomics analysis to enhance the diagnostic capabilities of OCTA imaging. Quantitative assessment of vascular characteristics may improve the identification of vascular leakage. By integrating radiomics analysis with OCTA imaging, this study presents a novel approach to detecting vascular leakage source in DR patients. The quantitative assessment of vessel characteristics not only enhances understanding of vascular pathology but also holds promise for improving diagnostic accuracy in clinical practice.
