Comparative Analysis of Improved of FCM Algorithms in the Segmentation of Retinal Blood Vessels

The main purpose of identifying and locating the vessels of the retina is to specify the various tissues from the vascular structure of the retina (which could be differencied between wide or tight) of the background of the fundus image. There exist several segmentation techniques that are spreading to divide the retinal vessels, depending on the issues and complexity of the retinal images. Fuzzy c-means is one of the most often used algorithms for retinal image segmentation due to its effectiveness and speed. This paper analyzes the performance of improved FCM algorithms for retinal image segmentation in terms of their ability and capability in segmenting and isolating blood vessels. The process we followed in our paper consists of two phases. Firstly, the pre-processing phase, where the green channel is taken for the color image of the retina. Contrast enhancement is performed through CLAHE , proceeded by applying bottom-hat filtering to bottom-hat filtering is applied with the purpose to define the region of interest. Secondly, in the segmentation phase the obtained image is segmented using FCM algorithms. The algorithms chosen for this study are: FCM, EnFCM, SFCM, FGFCM, FRFCM, DSFCM_N, FCM_SICM and, SSFCM performed on DRIVE and STARE databases. Experiments accomplished on DRIVE and STARE databases demonstrate that the DSFCM_N algorithm achieves better results on the DRIVE database, whereas the FGFCM algorithm provides better results on the STARE database in term of accuracy. Concerning time consumption. The FRFCM algorithm requires less time than other algorithms in the segmentation of retinal images.

Deep Learning in Retinal Image Segmentation and Feature Extraction: A Review

Image recognition and understanding is considered as a remarkable subfield of Artificial Intelligence (AI). In practice, retinal image data have high dimensionality leading to enormous size data. As the morphological retinal image datasets can be analyzed in an expansive and non-invasive way, AI more precisely Deep Learning (DL) methods are facilitating in developing intelligent retinal image analysis tools. The most recently developed DL technique, Convolutional Neural Network (CNN) showed remarkable efficiency in identifying, localizing, and quantifying the complex and hierarchical image features that are responsible for severe cardiovascular diseases. Different deep layered CNN architectures such as LeeNet, AlexNet, and ResNet have been developed exploiting CNN morphology. This wide variety of CNN structures can iteratively learn complex data structures of different datasets through supervised or unsupervised learning and perform exquisite analysis for feature recognition independently to diagnose threatening cardiovascular diseases. In modern ophthalmic practice, DL based automated methods are being used in retinopathy screening, grading, identifying, and quantifying the pathological features to employ further therapeutic approaches and offering a wide potentiality to get rid of ophthalmic system complexity. In this review, the recent advances of DL technologies in retinal image segmentation and feature extraction are extensively discussed. To accomplish this study the pertinent materials were extracted from different publicly available databases and online sources deploying the relevant keywords that includes retinal imaging, artificial intelligence, deep learning and retinal database. For the associated publications the reference lists of selected articles were further investigated.

ATTEMPTING AN INDEPTH ANALYSIS OF RETINAL IMAGE SEGMENTATION

These days eye weaknesses are a typical issue in all age group individuals begins from a newborn child to mature age. The discovery and extraction of these infections is a troublesome and tedious assignment. Computerized retinal pictures are considered; the first important strategy is to separate vessel in fundus pictures. Thus, three methods are utilized regulated techniques; here, the training set applies to remove vessel data by the pre-trained algorithm. This strategy is physically dealt with using gold std; vessel extraction is done before pathology calculations are involved in unaided recognition and extraction programs. The preparation set and ground truth marking will not be straightforwardly appropriate to the analysis. Retinal vessels extraction is improving as a result of noninvasive imaging of the retinal pictures likewise the information got from the design of the vasculature, and this data is essential for the identification and analysis of a fundus picture retinal sicknesses and pathologies, which incorporates glaucoma, hypertension, Diabetics Retina chart, and Age-based Macula De-age. Quick division calculations can recognize these.

Assessing the validity of a cross-platform retinal image segmentation tool in normal and diseased retina

Purpose- To compare automated retinal image segmentation using cross-platform and proprietary software on images captured using Heidelberg HRA + OCT in normal and diseased eyes. Methods- Study of retinal layer segmentations of normal, intermediate dry Age-related Macular Degeneration (iAMD) and Diabetic Macular Edema (DME) eyes performed using Heidelberg Spectralis HRA + OCT and automated OCT segmentation software Orion. Results- Orion was significantly better than Heidelberg in the segmentation of NFL and INL layers in normal eyes. Orion generated significantly better segmentation only for NFL in iAMD and for INL and OPL layers in DME eyes when compared to the ‘gold standard’ of manual segmentation. To understand where differences lay, we directly compared layer volumes between Orion and Heidelberg software. In normal eyes, all retinal layer volumes calculated by the two softwares were moderate-strongly correlated except OUTLY. In iAMD eyes, GCIPL, INL, ONL, INLY, TRV layer volumes were moderate-strongly correlated between softwares. In eyes with DME, all layer volume values were moderate-strongly correlated between softwares. Conclusion - Findings suggest that cross-platform Orion retinal layer segmentation software can be used reliably to study the retinal layers and compares well against manual segmentation and the commonly used proprietary software for normal eyes and in particular for diseased eyes.

Multi-Scale and Multi-Branch Convolutional Neural Network for Retinal Image Segmentation

The accurate segmentation of retinal images is a basic step in screening for retinopathy and glaucoma. Most existing retinal image segmentation methods have insufficient feature information extraction. They are susceptible to the impact of the lesion area and poor image quality, resulting in the poor recovery of contextual information. This also causes the segmentation results of the model to be noisy and low in accuracy. Therefore, this paper proposes a multi-scale and multi-branch convolutional neural network model (multi-scale and multi-branch network (MSMB-Net)) for retinal image segmentation. The model uses atrous convolution with different expansion rates and skip connections to reduce the loss of feature information. Receiving domains of different sizes captures global context information. The model fully integrates shallow and deep semantic information and retains rich spatial information. The network embeds an improved attention mechanism to obtain more detailed information, which can improve the accuracy of segmentation. Finally, the method of this paper was validated on the fundus vascular datasets, DRIVE, STARE and CHASE datasets, with accuracies/F1 of 0.9708/0.8320, 0.9753/0.8469 and 0.9767/0.8190, respectively. The effectiveness of the method in this paper was further validated on the optic disc visual cup DRISHTI-GS1 dataset with an accuracy/F1 of 0.9985/0.9770. Experimental results show that, compared with existing retinal image segmentation methods, our proposed method has good segmentation performance in all four benchmark tests.