A review analysis on early glaucoma detection using structural features

Background: Glaucoma is the leading cause of irreversible blindness worldwide. It is very important to diagnose glaucoma in early stages so that timely management can be done. Spectral domain optical coherence tomography (SD-OCT), is a newer device which helps to diagnose glaucoma early. The aim of our study was to evaluate the RNFL, ONH, and mGCA (GCL+IPL) measurements for early glaucoma detection using spectral domain optical coherence tomography (SD-OCT).Methods: Total 30, POAG (primary open angle glaucoma) suspects were compared with 30 normal controls. The Cirrus HD-OCT optic disc cube 200 × 200 protocol was used to measure ONH, RNFL and macular parameters.Results: The average cpRNFL thickness of all quadrants was significantly lower in POAG suspects, (84.13±7.42 μm versus 103.85±8.95 μm, p<0.001). The superior GCL+IPL thickness of POAG suspects and controls was 75.75±2.60 μm and 80.05±1.74 μm, respectively, (p<0.001). The inferior GCL+IPL thickness of POAG suspects and controls was 75.98±2.59 μm and 80.00±1.79 μm, respectively, (p<0.001).Conclusions: The SD-OCT is an important device to diagnose POAG suspects, early. The GCA measurements and average RNFL (especially superior and inferior) measurements, both are equally good to discriminate between glaucoma suspects and normal controls.

Download Full-text

Efficient Machine Learning Techniques to Detect Glaucoma using Structure and Texture based Features

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.b3374.079220 ◽

2020 ◽

Vol 9 (2) ◽

pp. 193-201

Keyword(s):

Machine Learning ◽

Detection System ◽

Texture Features ◽

Structural Features ◽

Machine Learning Techniques ◽

World Health ◽

Fundus Image ◽

Glaucoma Detection ◽

Learning Techniques ◽

Health Organization

Survey of world health organization has revealed that retinal eye disease Glaucoma is the second leading cause for the blindness worldwide. It is the disease which will steal the vision of the patient without any warning or symptoms. About half of the world Glaucoma patients are estimated to be in Asia. Hence, for social and economic reasons, Glaucoma detection is necessary in preventing blindness and reducing the cost of surgical treatment of the disease. The objective of the paper is to predict and detect Glaucoma efficiently using image processing and machine learning based classification techniques. Segmentation techniques such as unique template approach, Gray Level Coherence Matrix based feature extraction approach and wavelet transform based approach are used to extract these structure and texture based features. Combination of structure based and texture based techniques along with machine learning techniques improves the efficiency of the system. Developed efficient Computer aided Glaucoma detection system classifies a fundus image as either Normal or Glaucomatous image based on the structural features of the fundus image such as Cup-to-Disc Ratio (CDR), Rim-to-Disc Ratio (RDR), Superior and Inferior neuro-retinal rim thicknesses, Vessel structure based features and Distribution of texture features in the fundus images.

Download Full-text