Automated Classification of VMT Pathology from Optical Coherence Tomography B-scans
[ 1 ] Instytut Automatyki i Robotyki, Wydział Automatyki, Robotyki i Elektrotechniki, Politechnika Poznańska | [ P ] employee
[2.2] Automation, electronics, electrical engineering and space technologies
2022
chapter in monograph / paper
english
- optical coherence tomography (OCT)
- classification
- macular pathology
- machine learning
- VGG
EN The research presented in this article aims to determine the possibility of identifying macular pathologies from optical coherence tomography (OCT) images. Early detection of a pathology in the human eye retina is beneficial for patient recovery. Disorders commonly evaluated in clinical practice and research are drusen in age-related macular degeneration (AMD), diabetic macular edema (DME), and choroidal neovascularization (CNV). Our work extends this set of pathologies to include the pathology associated with improper posterior vitreous detachment, namely vitreomacular traction (VMT). Experimental research was based on single line OCT B-scans of healthy eyes and eyes with the investigated pathologies. AMD, DME and CNV images were taken from a public OCT dataset for retina classification, while healthy subjects and VMT images were acquired using an Avanti Optovue device. The accuracy results of more than 97% show the possibility of automatic preliminary diagnosis and differentiation between macular pathologies.
07.11.2022
104 - 109
20