Automatic Detection of Diabetic Retinopathy Using Custom CNN and Grad-CAM

Daanouni, Othmane; Cherradi, Bouchaib; Tmiri, Amal

doi:10.1007/978-981-15-6048-4_2

Cited by 30 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ey stated that efficient treatment and potential cures for a number of eye illnesses that may cause vision loss should be available in India by the year 2020. Organizing certification programs for technicians and physicians will help them get the knowledge and abilities necessary for DR identification at an early stage [1][2][3]. Numerous techniques have been put out to find DR. is section focuses on deep learning and neural network techniques for multiclass classification.…”

Section: Review Of Literaturementioning

confidence: 99%

See 1 more Smart Citation

Minimized Computations of Deep Learning Technique for Early Diagnosis of Diabetic Retinopathy Using IoT-Based Medical Devices

Ayoub

Khan

Jadhav³

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Diabetes mellitus is the main cause of diabetic retinopathy, the most common cause of blindness worldwide. In order to slow down or prevent vision loss and degeneration, early detection and treatment are essential. For the purpose of detecting and classifying diabetic retinopathy on fundus retina images, numerous artificial intelligence-based algorithms have been put forth by the scientific community. Due to its real-time relevance to everyone’s lives, smart healthcare is attracting a lot of interest. With the convergence of IoT, this attention has increased. The leading cause of blindness among persons in their working years is diabetic eye disease. Millions of people live in the most populous Asian nations, including China and India, and the number of diabetics among them is on the rise. To provide medical screening and diagnosis for this rising population of diabetes patients, skilled clinicians faced significant challenges. Our objective is to use deep learning techniques to automatically detect blind spots in eyes and determine how serious they may be. We suggest an enhanced convolutional neural network (ECNN) utilizing a genetic algorithm in this paper. The ECNN technique’s accuracy results are compared to those of existing approaches like the K-nearest neighbor approach, convolutional neural network, and support vector machine with the genetic algorithm.

show abstract

Section: Review Of Literaturementioning

confidence: 99%

“…With 463 million cases worldwide and a projected 700 million cases by 2045 [ 1 ], diabetes mellitus is a severe public health issue. The most prevalent eye condition associated with diabetes, diabetic retinopathy (DR), affects at least one-third of diabetics [ 2 ].…”

Section: Introductionmentioning

confidence: 99%

Minimized Computations of Deep Learning Technique for Early Diagnosis of Diabetic Retinopathy Using IoT-Based Medical Devices

Ayoub

Khan

Jadhav³

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…Pneumonia is caused by many germs, including bacteria and viruses like the novel coronavirus (COVID- 19). Certain symptoms are specific to COVID-19, such as chest pain, breathing difficulties, coughing with mucus, high fever, intense diarrhea, abdominal pain, and extreme fatigue.…”

Section: Introductionmentioning

confidence: 99%

“…These smart tools constitute an important aid for medical professionals to gain time, effort, and accuracy [14,15]. The diagnosis of most diseases can be aided by ML and DL algorithms, such as brain tumor detection using the CNN technique [16,17], diabetes mellitus prediction [18,19], patients with atherosclerosis disease classification [20][21][22], and detecting pneumonia on lung images [23,24].…”

Section: Introductionmentioning

confidence: 99%

Combining CNN and Grad-Cam for COVID-19 Disease Prediction and Visual Explanation

Moujahid¹,

Cherradi²,

Al-Sarem³

et al. 2022

Intelligent Automation &Amp; Soft Computing

Self Cite

View full text Add to dashboard Cite

“…Based on the reverse gradient direction and applying that perturbation to the original sample; a structure connecting the sample with its adversarial neighbors is obtained. On the other hand, the attention model has become an integral part of DL, leading to impressive performance in image classification of DR [31], [32] and captioning. Furthermore, attention performance has been improved by MHA [33] which appeals for the ability to receive knowledge from various representation subspaces at multiple locations simultaneously.…”

mentioning

confidence: 99%

NSL-MHA-CNN: A Novel CNN Architecture for Robust Diabetic Retinopathy Prediction Against Adversarial Attacks

2022

Self Cite

View full text Add to dashboard Cite

Convolution Neural Network (CNN) models have gained ground in research activities particularly in medical images used for Diabetes Retinopathy (DR) detection. X-ray, MRI, and CT scans have all been used to validate CNN models, with classification accuracy generally reaching that of trained doctors. It is mandatory to evaluate the strength of CNN models used in medical tasks against adversarial attacks especially in healthcare; that is to say, the security of such models is becoming extremely relevant to the diagnosis as this latter will guide highstakes decision-making. However, little study has been conducted to better comprehend this issue. This paper focuses on MobileNet CNN architecture in order to investigate its vulnerability against fast gradient sign methods (FGSM) adversarial attacks. For this end, a Neural Structure Learning (NSL) and a Multi-Head Attention (MHA) have been used to effectively reduce the vulnerability against attack by end-to-end CNN training with adversarial neighbors that produce adversarial perturbations on optical coherence tomography (OCT) images. With suggested model NSL-MHA-CNN, there has been an ability to maintain model performance on adversarial attack without increasing cost of training. Through theoretical assistance and empirical validation, it was possible to examine the stability of MobileNet architecture and demonstrate its susceptibility, particularly to adversarial attack. The experiments in this paper show that indiscernible degrees of perturbation ε < 0.01 were sufficient to cause a task failure resulting to misclassification in majority of the time. Moreover, empirical simulation shows that the proposed approach advanced in this paper can be an effective method to defense against adversarial attack at level of CNN model testing.

show abstract

Automatic Detection of Diabetic Retinopathy Using Custom CNN and Grad-CAM

Cited by 30 publications

References 14 publications

Minimized Computations of Deep Learning Technique for Early Diagnosis of Diabetic Retinopathy Using IoT-Based Medical Devices

Minimized Computations of Deep Learning Technique for Early Diagnosis of Diabetic Retinopathy Using IoT-Based Medical Devices

Combining CNN and Grad-Cam for COVID-19 Disease Prediction and Visual Explanation

NSL-MHA-CNN: A Novel CNN Architecture for Robust Diabetic Retinopathy Prediction Against Adversarial Attacks

Contact Info

Product

Resources

About