CLASSIFICATION OF X-RAY AND CT IMAGES IN DIFFERENT COLOR SPACES USING ROBUST CNN

Mühendislik Bilimleri ve Tasarım Dergisi, Journal Year: 2024, Volume and Issue: 12(3), P. 505 - 516

Published: Sept. 26, 2024

Since deep learning models have been successfully used in many fields, they to identify sick and healthy people X-ray or Computed Tomography (CT) chest radiology images. In this study, Covid-19 pneumonia classification is performed on both CT images using the robust Convolutional Neural Network (CNN). BGR, HSV, CIE LAB color space transformations are applied show that model performs a successful independent of dataset characteristics. The binary accuracy rates for 98.7% 98.4%, 97.6% 99.4%, respectively. Precision, Recall, Specificity, F1 score, Mean Squared Error metrics calculated each dataset. addition, 5-fold cross-validation proved model. Although transformed into different spaces, proposed classification. Thus, even if image characteristics device brands change, computer-based system will be able make disease diagnoses at low cost where expert personnel insufficient.

Language: Английский

---

A Hybrid Model Based on Deep Features and Ensemble Learning for the Diagnosis of COVID-19: DeepFeat-E

Turkish Journal of Science and Technology, Journal Year: 2023, Volume and Issue: 18(1), P. 183 - 198

Published: March 11, 2023

COVID-19, which has been declared a pandemic disease, affected the lives of millions people and caused major epidemic. Despite development vaccines vaccination to prevent transmission COVID-19 case rates fluctuate worldwide. Therefore, rapid reliable diagnosis disease is critical importance. For this purpose, hybrid model based on transfer learning methods ensemble classifiers proposed in study. In approach, called DeepFeat-E, process performed by using deep features obtained from models consisting classical machine methods. To test dataset 21,165 X-ray images including 10,192 Normal, 6012 Lung Opacity, 1345 Viral Pneumonia 3616 were used. With highest accuracy was achieved with DenseNet201 Stacking method. Accordingly, 90.17%, 94.99% 94.93% for four, three two class applications, respectively. According results study, it seen that system can be used quickly reliably lower respiratory tract infections.

Language: Английский

---

A Robust Ensemble Convolutional Neural Networks for Diagnosing Chest Diseases

Journal of Computer Science, Journal Year: 2023, Volume and Issue: 19(12), P. 1520 - 1540

Published: Nov. 20, 2023

Radiologists employ X-ray images to differentiate various chest diseases. Given the intricate and meticulous nature of this diagnostic procedure, assistance automated models becomes imperative in detecting diagnosing diseases from images. This research paper proposed a novel approach called Ensemble Convolutional Neural Network for Diagnosing Chest Diseases (ECDCNet), aimed at accurately efficiently fifteen different through analysis lungs. The ECDCNet model comprised stack five CNNs: ResNet152V2, DenseNet121, Inceptionv3, Vogg19, Wavelet transform-CNN with architectures hyper-parameters enhance overall prediction performance. applied image segmentation lung's region using U-Net localize focus on relevant space facilitate identification specific radiological signs such as nodules, opacities, cavities, consolidation. Furthermore, study exploited three ensemble CNN strategies: Average voting, majority CNN-ensemble strategy Weighted Performance Metrics Strategy (WPME) set weights stage. WPME used four evaluation measures assessing importance each base model, including precision, recall, F1-score, accuracy, model. achieved an accuracy 95.3, 95.8 96.1% average collected dataset 110804 Further, it 97.9, 98.2 98.9% another public 13150

Language: Английский

---

A Comprehensive Review on COVID-19 Detection Based on Cough Sounds, Symptoms, CXR, and CT Images

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 75412 - 75425

Published: Jan. 1, 2024

The worldwide spread of the coronavirus illness has led to requirement creating machine-based technologies identify diseases. pandemic caused by new coronaviruses resulted in a significant loss life and necessitates development several affordable diagnostic methods detect presence COVID-19 infection. Thankfully, current era advanced technology, including transfer learning (TL) approaches, improved areas human health enabled identification chronic communicable There is need for thorough investigation order combat transmission this alarming virus via use evidence-based intelligence models implementation preventive measures. present systematic review focuses on examination TL fuzzy ensemble techniques that have been described literature pertaining strategies detecting COVID-19. Multiple studies used cough sounds, CT scans, X-ray images, symptoms information cases application DL/ML, TL, ensemble, inference approaches discussed paper.

Language: Английский

---

DANN: A Deep Attention Neural Network for Automatic Fruit Image Classification

Studies in computational intelligence, Journal Year: 2024, Volume and Issue: unknown, P. 533 - 561

Published: Jan. 1, 2024

Language: Английский

---