A New Early Stage Diabetic Retinopathy Diagnosis Model Using Deep Convolutional Neural Networks and Principal Component Analysis

Mohammedhasan, Mali; Uğuz, Harun

doi:10.18280/ts.370503

Cited by 22 publications

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pretrained versions of these networks on comprehensive public datasets like ImageNet [10] can be directly applied to tasks with labels overlapping with the original dataset or indirectly via transfer learning when labels differ. CNNs have been employed for image classification and object detection across various domains, including medical image classification [11,12], plant disease recognition [13][14][15], face recognition [16], and document classification [17,18].…”

Section: Related Workmentioning

confidence: 99%

Optimizing Hyperparameters for Thai Cuisine Recognition via Convolutional Neural Networks

Theera-Ampornpunt¹,

Treepong²

2023

View full text Add to dashboard Cite

Automated food logging is an essential component of modern dietary management, and food recognition plays a crucial role in this process. However, the recognition of dishes and food items unique to specific cultures or regions remains a less explored area. In this study, we focus on the automatic recognition of Thai cuisine, employing transfer learning techniques and comparing the performance of 20 state-of-the-art convolutional neural networks. We investigate the impact of hyperparameters, such as batch size and image resolution, as well as image preprocessing methods on classification accuracy and training time for the topperforming models. Our evaluation, using the THFOOD-50 dataset consisting of 15,688 images across 50 classes, demonstrates that the optimal model achieves top-1 and top-5 classification accuracies of 90.44% and 99.97%, respectively, representing a significant improvement over previous results. We find that increasing image resolution substantially enhances accuracy, while batch size exerts a negligible effect. Moreover, cropping the edges of images can further improve accuracy, but this technique is only effective when employing low image resolution. Our findings contribute to the development of advanced food recognition algorithms, with potential applications in dietary management and nutrition planning.

show abstract

Section: Related Workmentioning

confidence: 99%

Optimizing Hyperparameters for Thai Cuisine Recognition via Convolutional Neural Networks

Theera-Ampornpunt¹,

Treepong²

2023

View full text Add to dashboard Cite

show abstract

“…This model predicted DR with an accuracy of 95% [23]. Mohammed Hasan et al, suggested a combined method of Convolution Neural Network and Principal Component Analysis for the diagnosis of DR with an accuracy of 98.44% [24]. Hemelings et al, suggested a method based on a deep learning approach to identify glaucoma in which the fundus image was cropped with radius as image size percentage, optic nerve head (ONH) centered with spacing of 10-60%.…”

Section: Related Workmentioning

confidence: 99%

Multiple Eye Disease Detection using Hybrid Adaptive Mutation Swarm Optimization and RNN

Subin¹,

Kannan²

2022

IJACSA

View full text Add to dashboard Cite

The major cause of visual impairment in aged people is due to age related eye diseases such as cataract, diabetic retinopathy, and glaucoma. Early detection of eye diseases is necessary for better diagnosis. This paper concentrates on the early identification of various eye disorders such as cataract, diabetic retinopathy, and glaucoma from retinal fundus images. The proposed method focuses on the automated early detection of multiple diseases using hybrid adaptive mutation swarm optimization and regression neural networks (AED-HSR). In the proposed work, the input images are preprocessed and then multiple features such as entropy, mean, color, intensity, standard deviation, and statistics are extracted from the collected data. The extracted features are segmented by using an adaptive mutation swarm optimization (AMSO) algorithm to segment the disease sector from the fundus image. Finally, the features collected are fed to a regression neural network (RNN) classifier to classify each fundus image as normal or abnormal. If the classifier output is abnormal, then it is classified by the corresponding diseases in terms of cataract, glaucoma, and diabetic retinopathy, which improves the accuracy of detection and classification. Ultimately, the results of the classifiers are evaluated by several performance analyses and the viability of structural and functional features is considered. The proposed system predicts the type of the disease with an accuracy of 0.9808, specificity of 0.9934, sensitivity of 0.9803 and F1 score of 0.9861 respectively.

show abstract

“…Salama and Aly [20] obtained an accuracy of 98.87% with the U-Net model and InceptionV3 with the data augmentation for the breast cancer classification. Data augmentation influences the average precision of the class [21][22][23]. In the classification of mixed gases, Han et al [24] used VGG16, VGG19, ResNet18, ResNet34, and ResNet50 and achieved an accuracy of 96.67% with the adjustment of parameters.…”

Section: Related Workmentioning

confidence: 99%

Effect of Data Augmentation in the Classification and Validation of Tomato Plant Disease with Deep Learning Methods

Wagle¹,

Sampe²,

Faseehuddin³

et al. 2021

View full text Add to dashboard Cite

The paper discusses disease identification and classification in tomato plants, as well as the effect of data augmentation in deep learning models. The database used here is Tomato plant leaves (TPL) images from the PlantVillage Database in the healthy and disease classes. The disease categories have been chosen depending on their occurrence in the Indian States. The proposed ResNet50, ResNet18, and ResNet101 deep-learning model with transfer learning combined with the softmax classification are used to identify and categorize the tomato leaf images into the healthy or diseases classes in the dataset. The unique combination of including the noise and blur in the images and position and color data augmentation makes the dataset robust. Two different data augmentation methods are used for the classification problem, and significant improvement is seen in the classification accuracy with the proposed augmented dataset. The model’s success rate makes the model helpful in extending support in validating a model for identifying plant disease. The validation of models is done on PlantVillage and images taken at Krishi Vigyan Kendra Narayangaon, Pune, India. ResNet101 model trained with augmented dataset outperforms the testing accuracy of 99.99% and validation accuracy of 95.83%.

show abstract

A New Early Stage Diabetic Retinopathy Diagnosis Model Using Deep Convolutional Neural Networks and Principal Component Analysis

Cited by 22 publications

References 38 publications

Optimizing Hyperparameters for Thai Cuisine Recognition via Convolutional Neural Networks

Optimizing Hyperparameters for Thai Cuisine Recognition via Convolutional Neural Networks

Multiple Eye Disease Detection using Hybrid Adaptive Mutation Swarm Optimization and RNN

Effect of Data Augmentation in the Classification and Validation of Tomato Plant Disease with Deep Learning Methods

Contact Info

Product

Resources

About