NSCGCN: A novel deep GCN model to diagnosis COVID-19

Tang, Chaosheng; Hu, Chaochao; Sun, Junding; Wang, Shuihua; Zhang, Yudong

doi:10.1016/j.compbiomed.2022.106151

Cited by 7 publications

(10 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite a complex methodology, the study attained a classification accuracy of 80%, with a significant limitation being low accuracy in multi-class classification and limited class classifications. Even though Tang and his co-workers [17] used two substantial datasets, a chest X-ray (CXR) dataset comprising 6939 images and a CT dataset containing a notable 85,725 images, a Node-Self Convolution Graph Convolutional Network (NSCGCN) and a DenseNet201 feature extractor were employed to classify diseases into two classes: infection and normal. Their method obtained excellent accuracy rates of 97.09% for the CXR dataset and 99.22% for the CT dataset.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Our DCNN architecture introduces unique features that distinguish it from existing models. We have improved our architecture compared to other CNNs [9], modified VGG19 [10], EfficientNet [12], CNN-ELM [15], DenseNet121 [16], and DenseNet201 [17]. This is because it uses new connection patterns and optimization strategies, which makes it better at many tasks and datasets.…”

Section: Elastic Deformation Augmentationmentioning

confidence: 99%

“…Addressing these challenges and enhancing clinical diagnosis, we employed an automated system for chest radiographic images, which can efficiently diagnose respiratory issues. Numerous recently developed automated systems were examined in our literature review [5,[9][10][11][12][13][14][15][16][17][18][19][20][21]. Challenges were encountered by those systems in data handling during image processing and optimal feature extraction, complicated quantification, and high runtime complexity issues in classifying chest X-ray images.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

DeepChestGNN: A Comprehensive Framework for Enhanced Lung Disease Identification through Advanced Graphical Deep Features

Rana,

Hosen,

Tonni

et al. 2024

Sensors

View full text Add to dashboard Cite

Lung diseases are the third-leading cause of mortality in the world. Due to compromised lung function, respiratory difficulties, and physiological complications, lung disease brought on by toxic substances, pollution, infections, or smoking results in millions of deaths every year. Chest X-ray images pose a challenge for classification due to their visual similarity, leading to confusion among radiologists. To imitate those issues, we created an automated system with a large data hub that contains 17 datasets of chest X-ray images for a total of 71,096, and we aim to classify ten different disease classes. For combining various resources, our large datasets contain noise and annotations, class imbalances, data redundancy, etc. We conducted several image pre-processing techniques to eliminate noise and artifacts from images, such as resizing, de-annotation, CLAHE, and filtering. The elastic deformation augmentation technique also generates a balanced dataset. Then, we developed DeepChestGNN, a novel medical image classification model utilizing a deep convolutional neural network (DCNN) to extract 100 significant deep features indicative of various lung diseases. This model, incorporating Batch Normalization, MaxPooling, and Dropout layers, achieved a remarkable 99.74% accuracy in extensive trials. By combining graph neural networks (GNNs) with feedforward layers, the architecture is very flexible when it comes to working with graph data for accurate lung disease classification. This study highlights the significant impact of combining advanced research with clinical application potential in diagnosing lung diseases, providing an optimal framework for precise and efficient disease identification and classification.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Section: Elastic Deformation Augmentationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

DeepChestGNN: A Comprehensive Framework for Enhanced Lung Disease Identification through Advanced Graphical Deep Features

Rana,

Hosen,

Tonni

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Too many graph convolution layers may cause the oversmoothing problem, however, there are image classification studies that solve this problem to some extent [50][51][52], allowing the model to extract deep-level features.…”

Section: Graph Convolutional Network Modelsmentioning

confidence: 99%

“…The ablation experiments demonstrate that their proposed MAMF-GCN has strong robustness and high accuracy. The prominent innovation of NSCGCN [52] is to overcome the over-smoothing problem using feature fusion based on the node-self-convolution algorithm and to preserve the spatial structure information of the original feature graph using the feature reconstruction algorithm. The innovative point of the node-self-convolution algorithm is that the input undirected graph 𝐺𝐺 𝑙𝑙 retains only the node-selfconnected degree matrix 𝐼𝐼.…”

Section: Graph Convolutional Network Modelsmentioning

confidence: 99%

Review of Image Classification Algorithms Based on Graph Convolutional Networks

Tang

2023

EAI Endorsed Trans AI Robotics

View full text Add to dashboard Cite

In recent years, graph convolutional networks (GCNs) have gained widespread attention and applications in image classification tasks. While traditional convolutional neural networks (CNNs) usually represent images as a two-dimensional grid of pixels when processing image data, the classical model of graph neural networks (GNNs), GCNs, can effectively handle data with graph structure, such as social networks, recommender systems, and molecular structures. In this paper, we will introduce the problems that graph convolutional networks have had, such as over-smoothing, and the methods to solve them, and suggest some possible future directions.

show abstract