Machine Learning Based Automated Segmentation and Hybrid Feature Analysis for Diabetic Retinopathy Classification Using Fundus Image

Ali, Aqib; Qadri, Salman; Khan, Wali Ullah; Kumam, Wiyada; Göktaş, Atilla; Jamal, Farrukh; Chesneau, Christophe; Anam, Sania; Sulaiman, Muhammad

doi:10.3390/e22050567

Cited by 55 publications

(41 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Results obtained by our technique are presented for homogeneous and heterogeneous porous media. The mathematical model for heterogeneous porous media is presented in Eq (29)(30)(31), and for homogeneous porous media, the ordinary differential equation together with its conditions is given in Eq (40)(41)(42). It is obvious from both models (for homogeneous and heterogenous media) that saturation depends on time T and distance X.…”

Section: Numerical Resultsmentioning

confidence: 99%

“…where 1 and 2 are the fitness-based errors in solutions for problem in Eq (29) and the boundary conditions Eq (30)(31), mathematically it is given as,…”

Section: Fitness Function Formulationmentioning

confidence: 99%

“…Plant propagation algorithm is modified and applied to constrained, unconstrained prob-lems, and theoretical analysis are studied in [24]- [26]. A state-of-the-art survey is published in [27], where evolutionary algorithms are investigated in terms of decomposition and indicator functions [28]- [30]. In electrical engineering, several metaheuristics are used to solve complex optimization problems [31]- [34].…”

mentioning

confidence: 99%

See 2 more Smart Citations

Analysis of Multi-Phase Flow Through Porous Media for Imbibition Phenomena by Using the LeNN-WOA-NM Algorithm

et al. 2020

Self Cite

View full text Add to dashboard Cite

The flow of fluids in multi-phase porous media results due to many interesting natural phenomena. The counter-current water imbibition phenomena, that occur during oil extraction through a cylindrical well is an interesting problem in petroleum engineering. During the secondary oil recovery process, water is injected into a porous media having heterogenous and homogenous characteristics. Due to the difference in viscosities of fluids in oil wells, the counter-current imbibition phenomenon occurs. At that moment, the imbibition equation V i = −V n is satisfied by the viscosities of oil and water. In this paper, we have analyzed the governing mathematical model of the imbibition phenomenon occurring during the secondary oil recovery process. A new soft computing algorithm is designed and adapted to analyze the mathematical model of dual-phase flow in detail. Weighted Legendre polynomials based artificial neural networks are hybridized with an efficient global optimizer the Whale Optimization Algorithm (WOA) and a local optimizer the Nelder-Mead algorithm. It is established, that our algorithm LeNN-WOA-NM is efficient and reliable in calculating high-quality solutions in less time. We have compared our experimental outcome with state-of-the-art results. The quality of our solutions is judged based on values of absolute errors, MAD, TIC, and ENSE. It is obvious that LeNN-WOA-NM algorithm can solve real application problems efficiently and accurately. INDEX TERMS Counter-current imbibition phenomena, homogeneous porous media, heterogeneous porous media, capillary pressure, soft computing algorithm, weighted Legendre neural networks, Nelder-Mead algorithm, whale optimization algorithm. I. INTRODUCTION T HIS paper models the phenomena of counter-current imbibition in the multi-phase flow of two immiscible fluids through heterogeneous and homogenous porous mediums that occurs during the secondary oil recovery process or water flooding [1]. In applied sciences, various changes occur in porous mediums having multi-phase flows of different fluids. In the past few decades, researchers are focused on studying the flows in underground petroleum reservoirs [2]. The primary recovery process includes the production of oil by simple natural decompression with no effect of external forces at wells [3]. In secondary oil recovery or water flooding process, fluids like water, polymer, steam are injected into a reservoir through injection wells located in rocks that have fluid communication with production wells. This process is called the immiscible displacement process [3].

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

“…where 1 and 2 are the fitness-based errors in solutions for problem in Eq (29) and the boundary conditions Eq (30)(31), mathematically it is given as,…”

Section: Fitness Function Formulationmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Analysis of Multi-Phase Flow Through Porous Media for Imbibition Phenomena by Using the LeNN-WOA-NM Algorithm

et al. 2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Five algorithms: multi-layer perceptron (MLP) [ 30 , 62 ], K-nearest neighbors (KNN) [ 63 , 64 ], quadratic discriminant analysis (QDA) [ 33 ], logistic regression (LR) [ 64 , 65 ], and support vector machine (SVM) [ 65 , 66 ] are evaluated in terms of accuracy, precision, and recall performance metrics as the main criteria for finalizing the hyper-parameters and model selection. The selection of classifiers is based on the hypothesis that one of these algorithms or improvised version after hyperparameter optimization will be the most suitable for classifying the microbes in general with high accuracy and lowest false alarm rate.…”

Section: Methodsmentioning

confidence: 99%

An Improvised Machine Learning Model Based on Mutual Information Feature Selection Approach for Microbes Classification

Dhindsa

Bhatia

Agrawal

et al. 2021

Entropy

View full text Add to dashboard Cite

The accurate classification of microbes is critical in today’s context for monitoring the ecological balance of a habitat. Hence, in this research work, a novel method to automate the process of identifying microorganisms has been implemented. To extract the bodies of microorganisms accurately, a generalized segmentation mechanism which consists of a combination of convolution filter (Kirsch) and a variance-based pixel clustering algorithm (Otsu) is proposed. With exhaustive corroboration, a set of twenty-five features were identified to map the characteristics and morphology for all kinds of microbes. Multiple techniques for feature selection were tested and it was found that mutual information (MI)-based models gave the best performance. Exhaustive hyperparameter tuning of multilayer layer perceptron (MLP), k-nearest neighbors (KNN), quadratic discriminant analysis (QDA), logistic regression (LR), and support vector machine (SVM) was done. It was found that SVM radial required further improvisation to attain a maximum possible level of accuracy. Comparative analysis between SVM and improvised SVM (ISVM) through a 10-fold cross validation method ultimately showed that ISVM resulted in a 2% higher performance in terms of accuracy (98.2%), precision (98.2%), recall (98.1%), and F1 score (98.1%).

show abstract

“…The sensitivity, specificity, and AUC of the system were 92.18% , 94.5% , and 92.4% , respectively. Other studies that use FP for automated diagnosis are also provided for reference [10][11][12][13][14] .…”

mentioning

confidence: 99%

Precise higher-order reflectivity and morphology models for early diagnosis of diabetic retinopathy using OCT images

Sharafeldeen

Elsharkawy

Khalifa

et al. 2021

Sci Rep

View full text Add to dashboard Cite

This study proposes a novel computer assisted diagnostic (CAD) system for early diagnosis of diabetic retinopathy (DR) using optical coherence tomography (OCT) B-scans. The CAD system is based on fusing novel OCT markers that describe both the morphology/anatomy and the reflectivity of retinal layers to improve DR diagnosis. This system separates retinal layers automatically using a segmentation approach based on an adaptive appearance and their prior shape information. High-order morphological and novel reflectivity markers are extracted from individual segmented layers. Namely, the morphological markers are layer thickness and tortuosity while the reflectivity markers are the 1st-order reflectivity of the layer in addition to local and global high-order reflectivity based on Markov-Gibbs random field (MGRF) and gray-level co-occurrence matrix (GLCM), respectively. The extracted image-derived markers are represented using cumulative distribution function (CDF) descriptors. The constructed CDFs are then described using their statistical measures, i.e., the 10th through 90th percentiles with a 10% increment. For individual layer classification, each extracted descriptor of a given layer is fed to a support vector machine (SVM) classifier with a linear kernel. The results of the four classifiers are then fused using a backpropagation neural network (BNN) to diagnose each retinal layer. For global subject diagnosis, classification outputs (probabilities) of the twelve layers are fused using another BNN to make the final diagnosis of the B-scan. This system is validated and tested on 130 patients, with two scans for both eyes (i.e. 260 OCT images), with a balanced number of normal and DR subjects using different validation metrics: 2-folds, 4-folds, 10-folds, and leave-one-subject-out (LOSO) cross-validation approaches. The performance of the proposed system was evaluated using sensitivity, specificity, F1-score, and accuracy metrics. The system’s performance after the fusion of these different markers showed better performance compared with individual markers and other machine learning fusion methods. Namely, it achieved $$96.15\%$$ 96.15 % , $$99.23\%$$ 99.23 % , $$97.66\%$$ 97.66 % , and $$97.69\%$$ 97.69 % , respectively, using the LOSO cross-validation technique. The reported results, based on the integration of morphology and reflectivity markers and by using state-of-the-art machine learning classifications, demonstrate the ability of the proposed system to diagnose the DR early.

show abstract

Machine Learning Based Automated Segmentation and Hybrid Feature Analysis for Diabetic Retinopathy Classification Using Fundus Image

Cited by 55 publications

References 30 publications

Analysis of Multi-Phase Flow Through Porous Media for Imbibition Phenomena by Using the LeNN-WOA-NM Algorithm

Analysis of Multi-Phase Flow Through Porous Media for Imbibition Phenomena by Using the LeNN-WOA-NM Algorithm

An Improvised Machine Learning Model Based on Mutual Information Feature Selection Approach for Microbes Classification

Precise higher-order reflectivity and morphology models for early diagnosis of diabetic retinopathy using OCT images

Contact Info

Product

Resources

About