An intelligent deep network for dental medical image processing system

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 84, P. 104708 - 104708

Published: March 6, 2023

Language: Английский

---

Unlocking the Power of 3D Convolutional Neural Networks for COVID-19 Detection: A Comprehensive Review

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 23, 2025

The advent of three-dimensional convolutional neural networks (3D CNNs) has revolutionized the detection and analysis COVID-19 cases. As imaging technologies have advanced, 3D CNNs emerged as a powerful tool for segmenting classifying in medical images. These demonstrated both high accuracy rapid capabilities, making them crucial effective diagnostics. This study offers thorough review various CNN algorithms, evaluating their efficacy across range modalities. systematically examines recent advancements methodologies. process involved comprehensive screening abstracts titles to ensure relevance, followed by meticulous selection research papers from academic repositories. evaluates these based on specific criteria provides detailed insights into network architectures algorithms used detection. reveals significant trends use segmentation classification. It highlights key findings, including diverse employed compared other diseases, which predominantly utilize encoder/decoder frameworks. an in-depth methods, discussing strengths, limitations, potential areas future research. reviewed total 60 published repositories, Springer Elsevier. this implications clinical diagnosis treatment strategies. Despite some efficiency underscore advancing image findings suggest that could significantly enhance management COVID-19, contributing improved healthcare outcomes.

Language: Английский

---

EVAE-Net: An Ensemble Variational Autoencoder Deep Learning Network for COVID-19 Classification Based on Chest X-ray Images

Diagnostics, Journal Year: 2022, Volume and Issue: 12(11), P. 2569 - 2569

Published: Oct. 22, 2022

The COVID-19 pandemic has had a significant impact on many lives and the economies of countries since late December 2019. Early detection with high accuracy is essential to help break chain transmission. Several radiological methodologies, such as CT scan chest X-ray, have been employed in diagnosing monitoring disease. Still, these methodologies are time-consuming require trial error. Machine learning techniques currently being applied by several studies deal COVID-19. This study exploits latent embeddings variational autoencoders combined ensemble propose three effective EVAE-Net models detect Two encoders trained X-ray images generate two feature maps. maps concatenated passed either or individual reparameterization phase sampling from distribution. classification head for classification. Radiography Dataset Kaggle source images. performances evaluated. proposed model shows satisfactory performance, best achieving 99.19% 98.66% four classes classes, respectively.

Language: Английский

---

Ensemble of Deep Neural Networks based on Condorcet’s Jury Theorem for screening Covid-19 and Pneumonia from radiograph images

Computers in Biology and Medicine, Journal Year: 2022, Volume and Issue: 149, P. 105979 - 105979

Published: Aug. 25, 2022

Language: Английский

---

MTMC-AUR2CNet: Multi-textural multi-class attention recurrent residual convolutional neural network for COVID-19 classification using chest X-ray images

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 85, P. 104857 - 104857

Published: March 21, 2023

Language: Английский

---

A Novel Deep Learning-Based Classification Framework for COVID-19 Assisted with Weighted Average Ensemble Modeling

Diagnostics, Journal Year: 2023, Volume and Issue: 13(10), P. 1806 - 1806

Published: May 19, 2023

COVID-19 is an infectious disease caused by the deadly virus SARS-CoV-2 that affects lung of patient. Different symptoms, including fever, muscle pain and respiratory syndrome, can be identified in COVID-19-affected patients. The needs to diagnosed a timely manner, otherwise infection turn into severe form patient's life may danger. In this work, ensemble deep learning-based technique proposed for detection classify with high accuracy, efficiency, reliability. A weighted average (WAE) prediction was performed combining three CNN models, namely Xception, VGG19 ResNet50V2, where 97.25% 94.10% accuracy achieved binary multiclass classification, respectively. To accurately detect disease, different test methods have been developed, some which are even being used real-time situations. RT-PCR one most successful methods, worldwide sensitivity. However, complexity time-consuming manual processes limitations method. make process automated, researchers across world started use learning applied on medical imaging. Although existing systems offer limitations, variance, overfitting generalization errors, found degrade system performance. Some reasons behind those lack reliable data resources, missing preprocessing techniques, proper model selection, etc., eventually create reliability issues. Reliability important factor any healthcare system. Here, transfer better techniques two benchmark datasets makes work more reliable. hyperparameter tuning ensures than using randomly selected single model.

Language: Английский

---

Handling imbalanced class in melanoma: Kemeny–Young rule based optimal rank aggregation and Self-Adaptive Differential Evolution Optimization

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 125, P. 106738 - 106738

Published: July 11, 2023

Language: Английский

---

A Real Time Method for Distinguishing COVID-19 Utilizing 2D-CNN and Transfer Learning

Sensors, Journal Year: 2023, Volume and Issue: 23(9), P. 4458 - 4458

Published: May 3, 2023

Rapid identification of COVID-19 can assist in making decisions for effective treatment and epidemic prevention. The PCR-based test is expert-dependent, time-consuming, has limited sensitivity. By inspecting Chest R-ray (CXR) images, COVID-19, pneumonia, other lung infections be detected real time. current, state-of-the-art literature suggests that deep learning (DL) highly advantageous automatic disease classification utilizing the CXR images. goal this study to develop models by employing DL identifying disorders more efficiently. For study, a dataset 18,564 images with seven categories was created from multiple publicly available sources. Four architectures including proposed CNN model pretrained VGG-16, VGG-19, Inception-v3 were applied identify healthy six diseases (fibrosis, opacity, viral bacterial tuberculosis). Accuracy, precision, recall, f1 score, area under curve (AUC), testing time used evaluate performance these four models. results demonstrated outperformed all employed seven-class an accuracy 93.15% average values f1-score, AUC 0.9343, 0.9443, 0.9386, 0.9939. equally performed well when multiclass classifications normal as common classes considered, yielding 98%, 97.49%, 97.81%, 96%, 96.75% two, three, four, five, classes, respectively. also shorter training times compared transfer

Language: Английский

---

Contactless blood oxygen estimation from face videos: A multi-model fusion method based on deep learning

Biomedical Signal Processing and Control, Journal Year: 2022, Volume and Issue: 81, P. 104487 - 104487

Published: Dec. 10, 2022

Language: Английский

---

Heterogeneous fusion of biometric and deep physiological features for accurate porcine cough recognition

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(2), P. e0297655 - e0297655

Published: Feb. 1, 2024

Accurate identification of porcine cough plays a vital role in comprehensive respiratory health monitoring and diagnosis pigs. It serves as fundamental prerequisite for stress-free animal management, reducing pig mortality rates, improving the economic efficiency farming industry. Creating representative multi-source signal signature is crucial step toward automating its identification. To this end, feature fusion method that combines biological features extracted from acoustic source segment with deep physiological derived thermal images proposed paper. First, various domains are sound signals. determine most effective combination features, an SVM-based recursive elimination cross-validation algorithm (SVM-RFECV) employed. Second, shallow convolutional neural network (named ThermographicNet) constructed to extract images. Finally, two heterogeneous integrated at early stage input into support vector machine (SVM) recognition. Through rigorous experimentation, performance approach evaluated, achieving impressive accuracy 98.79% recognizing cough. These results further underscore effectiveness combining thereby establishing robust representation

Language: Английский

---