Fundus-DeepNet: Multi-label deep learning classification system for enhanced detection of multiple ocular diseases through data fusion of fundus images

Al-Fahdawi, Shumoos; Al-Waisy, Alaa S.; Zeebaree, Diyar Qader; Qahwaji, Rami; Natiq, Hayder; Mohammed, Mazin Abed; Nedoma, Jan; Martinek, Radek; Deveci, Muhammet

doi:10.1016/j.inffus.2023.102059

Cited by 21 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we provide a comprehensive review of the recent work carried out in different areas of eye disease recognition (Al-Fahdawi et al. , 2024).…”

Section: Related Workmentioning

confidence: 99%

“…In this section, we provide a comprehensive review of the recent work carried out in different areas of eye disease recognition (Al-Fahdawi et al, 2024). Proposed Fundus-DeepNet for the detection of multiple ocular diseases in fundus images.…”

Section: Related Workmentioning

confidence: 99%

“…To validate the proposed model, we performed tests on the eye disease datasets used in the related work. As an effort to perform tests on the datasets used in the related work, we selected the OIA-ODIR (Al-Fahdawi et al, 2024) and ACRIMA (Sonti and Dhuli, 2023) datasets and computed the metric mAP by utilizing the weights obtained by training the proposed EYE-YOLO model on the created eye disease detection dataset as used in this work. From the OIA-ODIR dataset, we selected image samples for cataract, glaucoma and normal eye and from the ACRIMA dataset, the image samples for glaucoma and normal eye were selected.…”

Section: Ijicc 173mentioning

confidence: 99%

See 2 more Smart Citations

EYE-YOLO: a multi-spatial pyramid pooling and Focal-EIOU loss inspired tiny YOLOv7 for fundus eye disease detection

Kumar,

Dhanalakshmi

2024

IJICC

View full text Add to dashboard Cite

PurposeThe purpose of this work is to present an approach for autonomous detection of eye disease in fundus images. Furthermore, this work presents an improved variant of the Tiny YOLOv7 model developed specifically for eye disease detection. The model proposed in this work is a highly useful tool for the development of applications for autonomous detection of eye diseases in fundus images that can help and assist ophthalmologists.Design/methodology/approachThe approach adopted to carry out this work is twofold. Firstly, a richly annotated dataset consisting of eye disease classes, namely, cataract, glaucoma, retinal disease and normal eye, was created. Secondly, an improved variant of the Tiny YOLOv7 model was developed and proposed as EYE-YOLO. The proposed EYE-YOLO model has been developed by integrating multi-spatial pyramid pooling in the feature extraction network and Focal-EIOU loss in the detection network of the Tiny YOLOv7 model. Moreover, at run time, the mosaic augmentation strategy has been utilized with the proposed model to achieve benchmark results. Further, evaluations have been carried out for performance metrics, namely, precision, recall, F1 Score, average precision (AP) and mean average precision (mAP).FindingsThe proposed EYE-YOLO achieved 28% higher precision, 18% higher recall, 24% higher F1 Score and 30.81% higher mAP than the Tiny YOLOv7 model. Moreover, in terms of AP for each class of the employed dataset, it achieved 9.74% higher AP for cataract, 27.73% higher AP for glaucoma, 72.50% higher AP for retina disease and 13.26% higher AP for normal eye. In comparison to the state-of-the-art Tiny YOLOv5, Tiny YOLOv6 and Tiny YOLOv8 models, the proposed EYE-YOLO achieved 6–23.32% higher mAP.Originality/valueThis work addresses the problem of eye disease recognition as a bounding box regression and detection problem. Whereas, the work in the related research is largely based on eye disease classification. The other highlight of this work is to propose a richly annotated dataset for different eye diseases useful for training deep learning-based object detectors. The major highlight of this work lies in the proposal of an improved variant of the Tiny YOLOv7 model focusing on eye disease detection. The proposed modifications in the Tiny YOLOv7 aided the proposed model in achieving better results as compared to the state-of-the-art Tiny YOLOv8 and YOLOv8 Nano.

show abstract

“…In this section, we provide a comprehensive review of the recent work carried out in different areas of eye disease recognition (Al-Fahdawi et al. , 2024).…”

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Ijicc 173mentioning

confidence: 99%

See 1 more Smart Citation

EYE-YOLO: a multi-spatial pyramid pooling and Focal-EIOU loss inspired tiny YOLOv7 for fundus eye disease detection

Kumar,

Dhanalakshmi

2024

IJICC

View full text Add to dashboard Cite

show abstract

“…Our method uses non-structural feature selection methods like histogram of oriented gradients (HOGs), local binary pattern (LBP), and deep features from MobileNet and VGG19 to make it easier to show glaucoma fundus images completely. We use a k-nearest neighbors (k-NN) classifier to evaluate the effectiveness of CFO-CS feature selection (FS) on glaucoma fundus images, which is a commonly used technique in such tasks [ 32 , 33 , 34 ]. We evaluate the proposed framework’s effectiveness using various metrics, including accuracy, F1 score, precision, specificity, sensitivity, and the Matthew correlation coefficient.…”

Section: Introductionmentioning

confidence: 99%

Enhancing Medical Image Classification with an Advanced Feature Selection Algorithm: A Novel Approach to Improving the Cuckoo Search Algorithm by Incorporating Caputo Fractional Order

Habeb,

Taresh,

et al. 2024

Diagnostics

View full text Add to dashboard Cite

Glaucoma is a chronic eye condition that seriously impairs vision and requires early diagnosis and treatment. Automated detection techniques are essential for obtaining a timely diagnosis. In this paper, we propose a novel method for feature selection that integrates the cuckoo search algorithm with Caputo fractional order (CFO-CS) to enhance the performance of glaucoma classification. However, when using the infinite series, the Caputo definition has memory length truncation issues. Therefore, we suggest a fixed memory step and an adjustable term count for optimization. We conducted experiments integrating various feature extraction techniques, including histograms of oriented gradients (HOGs), local binary patterns (LBPs), and deep features from MobileNet and VGG19, to create a unified vector. We evaluate the informative features selected from the proposed method using the k-nearest neighbor. Furthermore, we use data augmentation to enhance the diversity and quantity of the training set. The proposed method enhances convergence speed and the attainment of optimal solutions during training. The results demonstrate superior performance on the test set, achieving 92.62% accuracy, 94.70% precision, 93.52% F1-Score, 92.98% specificity, 92.36% sensitivity, and 85.00% Matthew’s correlation coefficient. The results confirm the efficiency of the proposed method, rendering it a generalizable and applicable technique in ophthalmology.

show abstract

“…Shumoos Al-Fahdawiet al (15) have introduced the Fundus-Deep Net system, which includes image pre-processing incorporated with the SoftMax layer and Discriminative Restricted Boltzmann Machine (DRBM) that specifically identifies eight different ocular diseases.…”

Section: Introductionmentioning

confidence: 99%

Brain Tumor Prediction and Segmentation with Morphological Region-based Active Contour Model and Refinement using Boltzmann Monte Carlo Method in MRI Images

Srivaishnavi,

Perumal,

Anishiya

2024

IJST

View full text Add to dashboard Cite

Objectives: The primary goal of the research work is to accurately detect the precise location of the brain tumor in the radiological Magnetic Resonance Imaging (MRI) images of human brain using segmentation method. Methods: In this research work, we introduce mainly the Morphological Region-based Active Contour model and Boltzmann Monte Carlo method (MACB model), involving a comprehensive three-step methodology for the segmentation of the brain, MRI images in order to detect brain tumor. The initial step involves pre-processing which includes Gaussian filtering for noise reduction and Contrast Limited Adaptive Histogram Equalization (CLAHE) technique to enhance image features. In the second step, we identify tumor-related clusters using morphological operations and delineate the tumor regions using Active Contour (Snake) model to get a segmented image. In the final step, the Boltzmann Monte Carlo method is used to refine the edges of the segmented image. To evaluate the effectiveness of this approach, the 2D brain tumor datasets, available in the public domain, are used. The first dataset is taken from Kaggle website and has 3064 MRI human brain images and its respective ground truth images which is used for segmentation. The second dataset is used for visualization of segmented tumor, available in the same Kaggle website. Findings: The Performance metrics for finding similarity between the segmented images generated using the proposed MACB model and the ground truth images, available in the first dataset, exhibit higher values. That is, the proposed method has achieved higher values of Dice Similarity Coefficient (DSC): 93.26%, Jaccard Co-efficient: 86.44%, Sensitivity: 97.27%, Specificity: 99.43% and Pixel accuracy: 98.95%. Novelty: In this research work, MACB model is proposed for the detection, segmentation, and refinement process of brain tumor by incorporating Boltzmann Monte Carlo method with Morphological Region-Based Active Contour model. This novel approach has resulted in enhanced precision and efficiency in the brain tumor segmentation process. Keywords: Brain Tumor Segmentation, Morphological Operation, Active Contour, Boltzmann Monte Carlo Method, Magnetic Resonance Imaging

show abstract

Fundus-DeepNet: Multi-label deep learning classification system for enhanced detection of multiple ocular diseases through data fusion of fundus images

Cited by 21 publications

References 17 publications

EYE-YOLO: a multi-spatial pyramid pooling and Focal-EIOU loss inspired tiny YOLOv7 for fundus eye disease detection

EYE-YOLO: a multi-spatial pyramid pooling and Focal-EIOU loss inspired tiny YOLOv7 for fundus eye disease detection

Enhancing Medical Image Classification with an Advanced Feature Selection Algorithm: A Novel Approach to Improving the Cuckoo Search Algorithm by Incorporating Caputo Fractional Order

Brain Tumor Prediction and Segmentation with Morphological Region-based Active Contour Model and Refinement using Boltzmann Monte Carlo Method in MRI Images

Contact Info

Product

Resources

About