Pterygium-Net: a deep learning approach to pterygium detection and localization

Zulkifley, Mohd Asyraf; Abdani, Siti Raihanah; Zulkifley, Nuraisyah Hani

doi:10.1007/s11042-019-08130-x

Cited by 40 publications

(32 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, a deep neural network named Pterygium-Net [71] has been proposed to perform pterygium detection. Like our VggNet16-wbn, the architecture of the proposed Pterygium-Net implements Vgg network, but it only utilizes three initial VggNet layers.…”

Section: Resultsmentioning

confidence: 99%

“…To the best of our knowledge, the implementation of DL approaches for pterygium ocular disease detection or screening is very limited. Recently, pterygium detection has been conducted using TL in a DNN approach by Zulkifley et al [71] in which a Vgg network was adopted to detect and localise pterygium tissues. While the proposed method successfully achieved detection with 95% sensitivity, 98.3% specificity and 81.1% localisation, the size of the database used for validation was rather limited with only 120 images.…”

Section: ) Towards Automated Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Pterygium Detection Using Deep Neural Network

et al. 2020

View full text Add to dashboard Cite

Ocular imaging has developed rapidly and plays a critical role in clinical care and ocular disease management. Development of image processing technologies pertinent to ocular diseases has paved the way for automated diagnostic systems including detection techniques using deep learning (DL) approaches. The prevalence of an abnormal tissue layer in the conjunctiva, known as pterygium eye disease, is increasing due to lack of awareness. Despite the non-cancerous/benign nature of pterygium, a clinical diagnosis from an ophthalmologist is still required to prevent the pterygium tissues from extending into the pupil, which would result in blurred vision. However, current diagnostic methods are mostly dependent on human expertise. Automated detection can potentially serve as an assistive method to reduce diagnosis time by applying a DL approach. Considering the lack of comprehensive research work on pterygium detection using DL, we propose a new architecture consisting of an improved CNN-based trained network named VggNet16-wbn that is derived from VggNet16, a pre-trained CNN algorithm. This paper presents an overview of the DL as a core approach to the transfer learning (TL) concept, as well as current efforts towards automated ocular detection approaches. A new architecture of a CNN-based trained network was proposed based on a network assessment from six CNN pre-trained networks to detect pterygium. This work consists of two main modules, namely, data acquisition and DCNN classification. The proposed trained network, VggNet16-wbn, shows the best performance with 99.22% accuracy, 98.45% sensitivity, and a perfect score on specificity and area under the curve metrics. This work has high potential for creating a pterygium screening system that can be used as a baseline for fully automated detection using a DL approach.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: ) Towards Automated Detectionmentioning

confidence: 99%

Automated Pterygium Detection Using Deep Neural Network

et al. 2020

View full text Add to dashboard Cite

show abstract

“…SqueezeNet [19] reduces the memory requirement by introducing fire modules that employ a 1 × 1 convolution kernel, which reduces the total number of parameters to just 736,963. A compact network with transfer learning was also introduced in [20] with just three convolutional and three dense layers. Then, a deeper model was introduced by Simonyan and Zisserman [21] that comprises of 16 convolutional and three dense layers.…”

Section: Convolutional Neural Network Classifiermentioning

confidence: 99%

COVID-19 Screening Using a Lightweight Convolutional Neural Network with Generative Adversarial Network Data Augmentation

2020

Self Cite

View full text Add to dashboard Cite

COVID-19 is a disease that can be spread easily with minimal physical contact. Currently, the World Health Organization (WHO) has endorsed the reverse transcription-polymerase chain reaction swab test as a diagnostic tool to confirm COVID-19 cases. This test requires at least a day for the results to come out depending on the available facilities. Many countries have adopted a targeted approach in screening potential patients due to the cost. However, there is a need for a fast and accurate screening test to complement this targeted approach, so that the potential virus carriers can be quarantined as early as possible. The X-ray is a good screening modality; it is quick at capturing, cheap, and widely available, even in third world countries. Therefore, a deep learning approach has been proposed to automate the screening process by introducing LightCovidNet, a lightweight deep learning model that is suitable for the mobile platform. It is important to have a lightweight model so that it can be used all over the world even on a standard mobile phone. The model has been trained with additional synthetic data that were generated from the conditional deep convolutional generative adversarial network. LightCovidNet consists of three components, which are entry, middle, and exit flows. The middle flow comprises five units of feed-forward convolutional neural networks that are built using separable convolution operators. The exit flow is designed to improve the multi-scale capability of the network through a simplified spatial pyramid pooling module. It is a symmetrical architecture with three parallel pooling branches that enable the network to learn multi-scale features, which is suitable for cases wherein the X-ray images were captured from all over the world independently. Besides, the usage of separable convolution has managed to reduce the memory usage without affecting the classification accuracy. The proposed method managed to get the best mean accuracy of 0.9697 with a low memory requirement of just 841,771 parameters. Moreover, the symmetrical spatial pyramid pooling module is the most crucial component; the absence of this module will reduce the screening accuracy to just 0.9237. Hence, the developed model is suitable to be implemented for mass COVID-19 screening.

show abstract

“…Therefore, home monitoring using a smartphone will help in the diagnosis and treatment of patients with upper respiratory symptoms to improve convenience and to reduce transmission. There have been several approaches adopting deep learning for automated diagnosis of several diseases using images captured by smartphones [ 4 , 5 ].…”

Section: Introductionmentioning

confidence: 99%

Toward automated severe pharyngitis detection with smartphone camera using deep learning networks

Yoo

Choi

Jang³

et al. 2020

Computers in Biology and Medicine

View full text Add to dashboard Cite

Purpose Severe pharyngitis is frequently associated with inflammations caused by streptococcal pharyngitis, which can cause immune-mediated and post-infectious complications. The recent global pandemic of coronavirus disease (COVID-19) encourages the use of telemedicine for patients with respiratory symptoms. This study, therefore, purposes automated detection of severe pharyngitis using a deep learning framework with self-taken throat images. Methods A dataset composed of two classes of 131 throat images with pharyngitis and 208 normal throat images was collected. Before the training classifier, we constructed a cycle consistency generative adversarial network (CycleGAN) to augment the training dataset. The ResNet50, Inception-v3, and MobileNet-v2 architectures were trained with transfer learning and validated using a randomly selected test dataset. The performance of the models was evaluated based on the accuracy and area under the receiver operating characteristic curve (ROC-AUC). Results The CycleGAN-based synthetic images reflected the pragmatic characteristic features of pharyngitis. Using the synthetic throat images, the deep learning model demonstrated a significant improvement in the accuracy of the pharyngitis diagnosis. ResNet50 with GAN-based augmentation showed the best ROC-AUC of 0.988 for pharyngitis detection in the test dataset. In the 4-fold cross-validation using the ResNet50, the highest detection accuracy and ROC-AUC achieved were 95.3% and 0.992, respectively. Conclusion The deep learning model for smartphone-based pharyngitis screening allows fast identification of severe pharyngitis with a potential of the timely diagnosis of pharyngitis. In the recent pandemic of COVID-19, this framework will help patients with upper respiratory symptoms to improve convenience in diagnosis and reduce transmission.

show abstract

Pterygium-Net: a deep learning approach to pterygium detection and localization

Cited by 40 publications

References 29 publications

Automated Pterygium Detection Using Deep Neural Network

Automated Pterygium Detection Using Deep Neural Network

COVID-19 Screening Using a Lightweight Convolutional Neural Network with Generative Adversarial Network Data Augmentation

Toward automated severe pharyngitis detection with smartphone camera using deep learning networks

Contact Info

Product

Resources

About