A Hybrid Deep Learning CNN model for COVID-19 detection from chest X-rays

Heliyon, Journal Year: 2024, Volume and Issue: 10(5), P. e26938 - e26938

Published: Feb. 29, 2024

Coronavirus disease (COVID-2019) is emerging in Wuhan, China 2019. It has spread throughout the world since year 2020. Millions of people were affected and caused death to them till now. To avoid spreading COVID-2019, various precautions restrictions have been taken by all nations. At same time, infected persons are needed identify isolate, medical treatment should be provided them. Due a deficient number Reverse Transcription Polymerase Chain Reaction (RT-PCR) tests, Chest X-ray image becoming an effective technique for diagnosing COVID-19. In this work, Hybrid Deep Learning CNN model proposed diagnosis COVID-19 using chest X-rays. The consists heading base model. utilizes two pre-trained deep learning structures such as VGG16 VGG19. feature dimensions from these models reduced incorporating different pooling layers, max average. part, dense layers size three with activation functions also added. A dropout layer supplemented overfitting. experimental analyses conducted efficacy hybrid existing transfer architectures VGG16, VGG19, EfficientNetB0 ResNet50 radiology database. Various classification techniques, K-Nearest Neighbor (KNN), Naive Bayes, Random Forest, Support Vector Machine (SVM), Neural Network, used performance comparison average along SVM-linear neural networks, both achieved accuracy 92%.These can employed assist radiologists physicians avoiding misdiagnosis rates validate positive cases.

Language: Английский

---

An Alzheimer’s disease classification model using transfer learning Densenet with embedded healthcare decision support system

Decision Analytics Journal, Journal Year: 2023, Volume and Issue: 9, P. 100348 - 100348

Published: Oct. 29, 2023

Training a Convolutional Neural Network (CNN) from scratch is time-consuming and expensive. In this study, we propose implementing the DenseNet architecture for classification of AD in three classes. Our approach leverages transfer learning architectures as base model showcases superior performance on MRI dataset compared to other techniques. We use variety methodologies provide thorough study our model. first create baseline without data augmentation, addressing difficulties classifying Alzheimer's disease (AD) caused by high-dimensional brain scans. The improved obtained through augmentation then highlighted, demonstrating its effectiveness handling sparse assisting generalization. also investigate impact omitting particular transformations modifying split ratios, providing more insights into behavior Through comprehensive evaluation, demonstrate that proposed system achieves an accuracy 96.5% impressive AUC 99%, surpassing previous methods. This mainly highlights architecture, current limitations future recommendation. Moreover, incorporating healthcare decision support further aid valuable diagnosis decision-making clinical settings.

Language: Английский

---

Achieving a New Artificial Intelligence System for Serum Protein Electrophoresis to Recognize M-Spikes

ACS Omega, Journal Year: 2025, Volume and Issue: 10(6), P. 5770 - 5777

Published: Feb. 3, 2025

In order to accurately identify the low-concentration M-spikes in serum protein electrophoresis (SPE) patterns, a new artificial intelligence (AI) system is explored. 166,003 SPE data sets, which were equally divided into 4 training sets and 1 optimal set, utilized establish evaluate AI system, namely, "AIRSPE". 10,014 internal test 1861 external with immunofixation (IFE) results as gold standard used assess performance of AIRSPE including sensitivity, negative predictive value, concordance. group different concentrations M-spikes, consistencies manual interpretation IF-positive compared. selected MobileNetv2, performed F1-score 84.60%, precision 76.20%, recall 95.20%, loss 26.80%, accuracy 89.48%, time 14 ms. sensitivity values 95.21% 97.65%, respectively, no significant difference compared set (P > 0.05). IFE showed concordance (k = 0.832) that implies an almost perfect agreement, was higher than between 0.699). The identified by positive detected but not mainly concentrated γ-fraction, M-spike lower 0.5 g/L. AIRSPE, established through deep learning validated results, significantly outperforms detecting demonstrating its potential assist clinical screening for M-spikes.

Language: Английский

---

Evaluating pre-processing and deep learning methods in medical imaging: Combined effectiveness across multiple modalities

Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 119, P. 558 - 586

Published: Feb. 10, 2025

Language: Английский

---

X-COVNet: Externally Validated Model for Computer-Aided Diagnosis of Pneumonia-Like Lung Diseases in Chest X-Rays

Lecture notes in computer science, Journal Year: 2025, Volume and Issue: unknown, P. 161 - 172

Published: Jan. 1, 2025

Language: Английский

---

Deep learning models for early and accurate diagnosis of ventilator-associated pneumonia in mechanically ventilated neonates

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 189, P. 109942 - 109942

Published: March 3, 2025

Language: Английский

---

Optimized fine-tuned ensemble classifier using Bayesian optimization for the detection of ear diseases

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 191, P. 110092 - 110092

Published: April 10, 2025

Language: Английский

---

Exploratory Architectures Analysis of Various Pre-trained Image Classification Models for Deep Learning

Journal of Advances in Information Technology, Journal Year: 2024, Volume and Issue: 15(1), P. 66 - 78

Published: Jan. 1, 2024

The image classification is one of the significant applications in area Deep Learning (DL) with respective to various sectors.Different types neural network architectures are available perform and each which produces different accuracy.The dataset features used influence outcome model.The research community working towards generalized model at least domain specific.On this gesture contemporary survey models identified using knowledge information management methods move further provide optimal architecture also classify images narrow down sector specific.The study systematically presents architecture, its variants, layers parameters for version model.Domain specific limitations type detailed.It helps researchers select appropriate sector.

Language: Английский

---

COVID-19 detection from Chest X-ray images using a novel lightweight hybrid CNN architecture

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: May 21, 2024

Language: Английский

---

MonuNet: a high performance deep learning network for Kolkata heritage image classification

Heritage Science, Journal Year: 2024, Volume and Issue: 12(1)

Published: July 16, 2024

Abstract Kolkata, renowned as the City of Joy, boasts a rich tapestry cultural heritage spanning centuries. Despite significance its architectural marvels, accessing comprehensive visual documentation Kolkata's sites remains challenge. In online searches, limited imagery often fails to provide detailed understanding these historical landmarks. To address this gap, paper introduces MonuNet, high-performance deep-learning network specifically designed for classification images from Kolkata. The development MonuNet addresses critical need efficient and accurate identification which are significant tangible heritages. dataset used train is organized by sites, each category within represents distinct sites. It includes 13 prominent in For there 50 images, making it structured collection where (heritage site) equally represented. proposed utilizes unique architecture incorporating Dense channel attention module Parallel-spatial capture intricate details spatial relationships images. Experimental evaluations demonstrate superior performance classifying Kolkata with an accuracy 89%, Precision 87.77%, Recall 86.61%. successful deployment holds implications preservation, tourism enhancement, urban planning aligning United Nations Sustainable Development Goals (SDGs) sustainable city development. By providing robust tool automatic promises enrich repositories documentation, thereby enhancing accessibility researchers, tourists, planners alike. Graphical

Language: Английский

---