Advanced OCTA imaging segmentation: Unsupervised, non-linear retinal vessel detection using modified self-organizing maps and joint MGRF modeling.
Ahmed AlksasAhmed SharafeldeenHossam Magdy BalahaMohammad Z. HaqAli MahmoudMohammed GhazalNorah Saleh AlghamdiMarah Talal AlhalabiJawad YousafHarpal SandhuAyman El-BazPublished in: Comput. Methods Programs Biomed. (2024)
Keyphrases
- vessel detection
- self organizing maps
- retinal images
- blood vessels
- unsupervised learning
- competitive learning
- optical coherence tomography
- neural network
- neural gas
- image analysis
- input data
- semi supervised
- k means
- supervised learning
- oct images
- medical imaging
- image processing
- fully unsupervised
- som neural network
- feature maps
- markov gibbs random field
- pairwise
- pattern recognition
- region of interest
- segmentation method
- segmentation algorithm
- medical images
- edge detection
- high dimensional
- image segmentation