Optimizing Coronary Artery Disease Detection Using a New Triple Concatenated Convolution Neural Network DOI Creative Commons
Slamet Riyadi,

Febriyanti Azahra Abidin,

Cahya Damarjati

et al.

Ingénierie des systèmes d information, Journal Year: 2024, Volume and Issue: 29(4), P. 1581 - 1589

Published: Aug. 21, 2024

Coronary artery disease (CAD) is a pathological condition that often fatal and the main cause of death throughout world.Early detection this very important to avoid severe complications such as heart attacks sudden death.This study employs artificial intelligence, specifically deep learning via Convolutional Neural Networks (CNNs), enhance CAD detection.While CNN architectures like ResNet50V2 MobileNetV2 exhibit satisfactory performance individually, they possess distinct strengths weaknesses.ResNet50V2 requires significant computing resources, hindering its scalability, while struggles with extracting complex features from medical images.Therefore, research aims combine EfficientNetV2B0, ResNet50V2, using transfer techniques detection.The methodology involves leveraging pre-trained models fine-tuning them on coronary dataset.Modified models, particularly EfficientNetV2B0 MobileNetV2, achieve high accuracies 94% 86%, respectively, yields 72%.However, combining boosts accuracy 95%, addressing individual model limitations.The concatenated demonstrates superior predictive capabilities, more accurate predictions fewer errors than models.

Language: Английский

Automated detection and forecasting of COVID-19 using deep learning techniques: A review DOI
Afshin Shoeibi, Marjane Khodatars, Mahboobeh Jafari

et al.

Neurocomputing, Journal Year: 2024, Volume and Issue: 577, P. 127317 - 127317

Published: Jan. 26, 2024

Language: Английский

Citations

56
TeutongNet: A Fine-Tuned Deep Learning Model for Improved Forest Fire Detection DOI Creative Commons
Ghazi Mauer Idroes, Aga Maulana,

Rivansyah Suhendra

et al.

Leuser Journal of Environmental Studies, Journal Year: 2023, Volume and Issue: 1(1), P. 1 - 8

Published: June 22, 2023

Forest fires have emerged as a significant threat to the environment, wildlife, and human lives, necessitating development of effective early detection systems for firefighting mitigation efforts. In this study, we introduce TeutongNet, modified ResNet50V2 model designed detect forest accurately. The is trained on curated dataset evaluated using various metrics. Results show that TeutongNet achieves high accuracy (98.68%) with low false positive negative rates. model's performance further supported by ROC curve analysis, which indicates degree in classifying fire non-fire images. demonstrates its effectiveness reliable detection, providing valuable insights improved management strategies.

Language: Английский

Citations

31
Enhanced PRIM recognition using PRI sound and deep learning techniques DOI Creative Commons
Seyed Majid Hasani Azhdari, Azar Mahmoodzadeh, Mohammad Khishe

et al.

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(5), P. e0298373 - e0298373

Published: May 1, 2024

Pulse repetition interval modulation (PRIM) is integral to radar identification in modern electronic support measure (ESM) and intelligence (ELINT) systems. Various distortions, including missing pulses, spurious unintended jitters, noise from antenna scans, often hinder the accurate recognition of PRIM. This research introduces a novel three-stage approach for PRIM recognition, emphasizing innovative use PRI sound. A transfer learning-aided deep convolutional neural network (DCNN) initially used feature extraction. followed by an extreme learning machine (ELM) real-time classification. Finally, gray wolf optimizer (GWO) refines network's robustness. To evaluate proposed method, we develop real experimental dataset consisting sound six common patterns. We utilized eight pre-trained DCNN architectures evaluation, with VGG16 ResNet50V2 notably achieving accuracies 97.53% 96.92%. Integrating ELM GWO further optimized accuracy rates 98.80% 97.58. advances offering enhanced method potential address real-world distortions ESM ELINT

Language: Английский

Citations

1
Advancing Virtual Interviews: AI-Driven Facial Emotion Recognition for Better Recruitment DOI Open Access

Rohini Mehta,

Pulicharla Sai Pravalika,

Bellamkonda Venkata Naga Durga Sai

et al.

International Journal of Innovative Science and Research Technology (IJISRT), Journal Year: 2024, Volume and Issue: unknown, P. 2288 - 2296

Published: Aug. 8, 2024

Behavior analysis involves the detailed process of identifying, modeling, and comprehending various nuances patterns emotional expressions exhibited by individuals. It poses a significant challenge to accurately detect predict facial emotions, especially in contexts like remote interviews, which have become increasingly prevalent. Notably, many participants struggle convey their thoughts interviewers with happy expression good posture, may unfairly diminish chances employment, despite qualifications. To address this challenge, artificial intelligence techniques such as image classification offer promising solutions. By leveraging AI models, behavior can be applied perceive interpret reactions, thereby paving way anticipate future behaviors based on learned participants. Despite existing works emotion recognition (FER) using classification, there is limited research focused platforms interviews online courses. In paper, our primary focus lies emotions happiness, sadness, anger, surprise, eye contact, neutrality, smile, confusion, stooped posture. We curated dataset, comprising diverse range sample captured through participants' video recordings other images documenting speech during interviews. Additionally, we integrated datasets FER 2013 Celebrity Emotions dataset. Through investigation, explore variety deep learning methodologies, including VGG19, ResNet50V2, ResNet152V2, Inception-ResNetV2, Xception, EfficientNet B0, YOLO V8 analyze emotions. Our results demonstrate an accuracy 73% v8 model. However, discovered that categories well surprised confused, are not disjoint, leading potential inaccuracies classification. Furthermore, considered posture non-essential class since conducted via webcam, does allow for observation removing these overlapping categories, achieved remarkable increase around 76.88%

Language: Английский

Citations

1
Optimizing Coronary Artery Disease Detection Using a New Triple Concatenated Convolution Neural Network DOI Creative Commons
Slamet Riyadi,

Febriyanti Azahra Abidin,

Cahya Damarjati

et al.

Ingénierie des systèmes d information, Journal Year: 2024, Volume and Issue: 29(4), P. 1581 - 1589

Published: Aug. 21, 2024

Coronary artery disease (CAD) is a pathological condition that often fatal and the main cause of death throughout world.Early detection this very important to avoid severe complications such as heart attacks sudden death.This study employs artificial intelligence, specifically deep learning via Convolutional Neural Networks (CNNs), enhance CAD detection.While CNN architectures like ResNet50V2 MobileNetV2 exhibit satisfactory performance individually, they possess distinct strengths weaknesses.ResNet50V2 requires significant computing resources, hindering its scalability, while struggles with extracting complex features from medical images.Therefore, research aims combine EfficientNetV2B0, ResNet50V2, using transfer techniques detection.The methodology involves leveraging pre-trained models fine-tuning them on coronary dataset.Modified models, particularly EfficientNetV2B0 MobileNetV2, achieve high accuracies 94% 86%, respectively, yields 72%.However, combining boosts accuracy 95%, addressing individual model limitations.The concatenated demonstrates superior predictive capabilities, more accurate predictions fewer errors than models.

Language: Английский

Citations

0