Customized VGG19 Architecture for Pneumonia Detection in Chest X-Rays

Pattern Recognition Letters, Journal Year: 2021, Volume and Issue: 143, P. 67 - 74

Published: Jan. 8, 2021

Language: Английский

---

A machine learning method for defect detection and visualization in selective laser sintering based on convolutional neural networks

Additive manufacturing, Journal Year: 2021, Volume and Issue: 41, P. 101965 - 101965

Published: March 23, 2021

Part defects and irregularities that influence the part quality is an especially large problem in additive manufacturing (AM) processes such as selective laser sintering (SLS). Destructive non-destructive testing procedures are currently mostly used for control defect detection of AM parts after production. In this context, machine learning (ML) algorithms increasingly being to enable computer-aided through automatic classification data. Convolutional neural networks (CNN) based on ML methods widely task. paper, complex transfer (TL) presented, which powder bed SLS process using very small datasets. The proposed use VGG16 Xception CNN model with pretrained weights from ImageNet dataset initialization adapted classifier classify good defective image data recorded during manufacturing. Known performance metrics were determined evaluate compare models. architecture achieved best results Accuracy (0.958), Precision (0.939), Recall (0.980), F1-Score (0.959) AUC value (0.982). These show effectiveness can offer alternative method assurance documentation additively manufactured parts.

Language: Английский

---

Transfer learning for medical images analyses: A survey

Neurocomputing, Journal Year: 2022, Volume and Issue: 489, P. 230 - 254

Published: March 17, 2022

Language: Английский

---

Detection and classification of lung diseases for pneumonia and Covid-19 using machine and deep learning techniques

Journal of Ambient Intelligence and Humanized Computing, Journal Year: 2021, Volume and Issue: 14(4), P. 3239 - 3259

Published: Sept. 18, 2021

Since the arrival of novel Covid-19, several types researches have been initiated for its accurate prediction across world. The earlier lung disease pneumonia is closely related to as patients died due high chest congestion (pneumonic condition). It challenging differentiate Covid-19 and diseases medical experts. X-ray imaging most reliable method prediction. In this paper, we propose a framework predictions like from images patients. consists dataset acquisition, image quality enhancement, adaptive region interest (ROI) estimation, features extraction, anticipation. used two publically available datasets. As degraded while taking X-ray, applied enhancement using median filtering followed by histogram equalization. For ROI extraction regions, designed modified growing technique that dynamic selection based on pixel intensity values morphological operations. detection diseases, robust set plays vital role. We extracted visual, shape, texture, each normalization. normalization, formulated enhance classification results. Soft computing methods such artificial neural network (ANN), support vector machine (SVM), K-nearest neighbour (KNN), ensemble classifier, deep learning classifier are classification. disease, architecture has proposed recurrent (RNN) with long short-term memory (LSTM). Experimental results show robustness efficiency model in comparison existing state-of-the-art methods.

Language: Английский

---

LungNet22: A Fine-Tuned Model for Multiclass Classification and Prediction of Lung Disease Using X-ray Images

Journal of Personalized Medicine, Journal Year: 2022, Volume and Issue: 12(5), P. 680 - 680

Published: April 24, 2022

In recent years, lung disease has increased manyfold, causing millions of casualties annually. To combat the crisis, an efficient, reliable, and affordable diagnosis technique become indispensable. this study, a multiclass classification from frontal chest X-ray imaging using fine-tuned CNN model is proposed. The conducted on 10 classes lungs, namely COVID-19, Effusion, Tuberculosis, Pneumonia, Lung Opacity, Mass, Nodule, Pneumothorax, Pulmonary Fibrosis, along with Normal class. dataset collective gathered multiple sources. After pre-processing balancing eight augmentation techniques, total 80,000 images were fed to for purposes. Initially, pre-trained models, AlexNet, GoogLeNet, InceptionV3, MobileNetV2, VGG16, ResNet 50, DenseNet121, EfficientNetB7, employed dataset. Among these, VGG16 achieved highest accuracy at 92.95%. further improve accuracy, LungNet22 was constructed upon primary structure model. An ablation study used in work determine different hyper-parameters. Using Adam Optimizer, proposed commendable 98.89%. verify performance model, several matrices, including ROC curve AUC values, computed as well.

Language: Английский

---

Deep features to detect pulmonary abnormalities in chest X-rays due to infectious diseaseX: Covid-19, pneumonia, and tuberculosis

Information Sciences, Journal Year: 2022, Volume and Issue: 592, P. 389 - 401

Published: Feb. 4, 2022

Language: Английский

---

A systematic literature review on deep learning approaches for pneumonia detection using chest X-ray images

Multimedia Tools and Applications, Journal Year: 2023, Volume and Issue: 83(8), P. 24101 - 24151

Published: Aug. 9, 2023

Language: Английский

---

Hybrid Inception Architecture with Residual Connection: Fine-tuned Inception-ResNet Deep Learning Model for Lung Inflammation Diagnosis from Chest Radiographs

Procedia Computer Science, Journal Year: 2024, Volume and Issue: 235, P. 1841 - 1850

Published: Jan. 1, 2024

Diagnosing lung inflammation, particularly pneumonia, is of paramount importance for effectively treating and managing the disease. Pneumonia a common respiratory infection caused by bacteria, viruses, or fungi can indiscriminately affect people all ages. As highlighted World Health Organization (WHO), this prevalent disease tragically accounts substantial 15% global mortality in children under five years age. This article presents comparative study Inception-ResNet deep learning model's performance diagnosing pneumonia from chest radiographs. The leverages Mendeley's X-ray images dataset, which contains 5856 2D images, including both Viral Bacterial images. model compared with seven other state-of-the-art convolutional neural networks (CNNs), experimental results demonstrate superiority extracting essential features saving computation runtime. Furthermore, we examine impact transfer fine-tuning improving models. provides valuable insights into using models diagnosis highlights potential field. In classification accuracy, Inception-ResNet-V2 showed superior to models, ResNet152V2, MobileNet-V3 (Large Small), EfficientNetV2 InceptionV3, NASNet-Mobile, margins. It outperformed them 2.6%, 6.5%, 7.1%, 13%, 16.1%, 3.9%, 1.6%, respectively, demonstrating its significant advantage accurate classification.

Language: Английский

---

A Survey of Deep Learning for Lung Disease Detection on Medical Images: State-of-the-Art, Taxonomy, Issues and Future Directions

Journal of Imaging, Journal Year: 2020, Volume and Issue: 6(12), P. 131 - 131

Published: Dec. 1, 2020

The recent developments of deep learning support the identification and classification lung diseases in medical images. Hence, numerous work on detection disease using can be found literature. This paper presents a survey for There has only been one published last five years regarding directed at detection. However, their is lacking presentation taxonomy analysis trend work. objectives this are to present state-of-the-art based systems, visualise trends domain identify remaining issues potential future directions domain. Ninety-eight articles from 2016 2020 were considered survey. consists seven attributes that common surveyed articles: image types, features, data augmentation, types algorithms, transfer learning, ensemble classifiers diseases. presented could used by other researchers plan research contributions activities. direction suggested further improve efficiency increase number aided applications.

Language: Английский

---

IoT and Deep Learning Based Approach for Rapid Screening and Face Mask Detection for Infection Spread Control of COVID-19

Applied Sciences, Journal Year: 2021, Volume and Issue: 11(8), P. 3495 - 3495

Published: April 13, 2021

The spread of COVID-19 has been taken on pandemic magnitudes and already over 200 countries in a few months. In this time emergency COVID-19, especially when there is still need to follow the precautions developed vaccines are not available all developing first phase vaccine distribution, virus spreading rapidly through direct indirect contacts. World Health Organization (WHO) provides standard recommendations preventing importance face masks for protection from virus. excessive use manual disinfection systems also become source infection. That why research aims design develop low-cost, rapid, scalable, effective control screening system minimize chances risk COVID-19. We proposed an IoT-based Smart Screening Disinfection Walkthrough Gate (SSDWG) public places entrance. SSDWG designed do rapid screening, including temperature measuring using contact-free sensor storing record suspected individual further monitoring. Our implemented real-time deep learning models mask detection classification. This module classified individuals who wear properly, improperly, without VGG-16, MobileNetV2, Inception v3, ResNet-50, CNN transfer approach. achieved highest accuracy 99.81% while VGG-16 second 99.6% MobileNetV2 classification module. classify types worn by individuals, either N-95 or surgical masks. compared results our with state-of-the-art methods, we highly suggested that could be used prevent local transmission reduce human carriers

Language: Английский

---