VGG19 Network Assisted Joint Segmentation and Classification of Lung Nodules in CT Images

Diagnostics, Journal Year: 2021, Volume and Issue: 11(12), P. 2208 - 2208

Published: Nov. 26, 2021

Pulmonary nodule is one of the lung diseases and its early diagnosis treatment are essential to cure patient. This paper introduces a deep learning framework support automated detection nodules in computed tomography (CT) images. The proposed employs VGG-SegNet supported mining pre-trained DL-based classification detection. CT images implemented using attained features, then these features serially concatenated with handcrafted such as Grey Level Co-Occurrence Matrix (GLCM), Local-Binary-Pattern (LBP) Pyramid Histogram Oriented Gradients (PHOG) enhance disease accuracy. used for experiments collected from LIDC-IDRI Lung-PET-CT-Dx datasets. experimental results show that VGG19 architecture can achieve an accuracy 97.83% SVM-RBF classifier.

Language: Английский

---

COVID-19 Case Recognition from Chest CT Images by Deep Learning, Entropy-Controlled Firefly Optimization, and Parallel Feature Fusion

Sensors, Journal Year: 2021, Volume and Issue: 21(21), P. 7286 - 7286

Published: Nov. 2, 2021

In healthcare, a multitude of data is collected from medical sensors and devices, such as X-ray machines, magnetic resonance imaging, computed tomography (CT), so on, that can be analyzed by artificial intelligence methods for early diagnosis diseases. Recently, the outbreak COVID-19 disease caused many deaths. Computer vision researchers support doctors employing deep learning techniques on images to diagnose patients. Various were proposed case classification. A new automated technique using parallel fusion optimization models. The starts with contrast enhancement combination top-hat Wiener filters. Two pre-trained models (AlexNet VGG16) are employed fine-tuned according target classes (COVID-19 healthy). Features extracted fused approach—parallel positive correlation. Optimal features selected entropy-controlled firefly method. classified machine classifiers multiclass vector (MC-SVM). Experiments carried out Radiopaedia database achieved an accuracy 98%. Moreover, detailed analysis conducted shows improved performance scheme.

Language: Английский

---

A Multi-Agent Deep Reinforcement Learning Approach for Enhancement of COVID-19 CT Image Segmentation

Journal of Personalized Medicine, Journal Year: 2022, Volume and Issue: 12(2), P. 309 - 309

Published: Feb. 18, 2022

Currently, most mask extraction techniques are based on convolutional neural networks (CNNs). However, there still numerous problems that need to solve. Thus, the advanced methods deploy artificial intelligence (AI) necessary. The use of cooperative agents in increases efficiency automatic image segmentation. Hence, we introduce a new method is multi-agent deep reinforcement learning (DRL) minimize long-term manual and enhance medical segmentation frameworks. A DRL-based introduced deal with issues. This utilizes modified version Deep Q-Network enable detector select masks from studied. Based COVID-19 computed tomography (CT) images, used DRL extraction-based extract visual features infected areas provide an accurate clinical diagnosis while optimizing pathogenic diagnostic test saving time. We collected CT images different cases (normal chest CT, pneumonia, typical viral cases, COVID-19). Experimental validation achieved precision 97.12% Dice 80.81%, sensitivity 79.97%, specificity 99.48%, 85.21%, F1 score 83.01%, structural metric 84.38%, mean absolute error 0.86%. Additionally, results clearly reflected ground truth. reveal proof principle for using effective COVID-19.

Language: Английский

---

A privacy-aware method for COVID-19 detection in chest CT images using lightweight deep conventional neural network and blockchain

Computers in Biology and Medicine, Journal Year: 2022, Volume and Issue: 145, P. 105461 - 105461

Published: March 28, 2022

Language: Английский

---

RADIC:A tool for diagnosing COVID-19 from chest CT and X-ray scans using deep learning and quad-radiomics

Chemometrics and Intelligent Laboratory Systems, Journal Year: 2023, Volume and Issue: 233, P. 104750 - 104750

Published: Jan. 2, 2023

Language: Английский

---

A computer-aided diagnostic framework for coronavirus diagnosis using texture-based radiomics images

Digital Health, Journal Year: 2022, Volume and Issue: 8, P. 205520762210925 - 205520762210925

Published: Jan. 1, 2022

The accurate and rapid detection of the novel coronavirus infection, is very important to prevent fast spread such disease. Thus, reducing negative effects that influenced many industrial sectors, especially healthcare. Artificial intelligence techniques in particular deep learning could help precise diagnosis from computed tomography images. Most artificial intelligence-based studies used original images build their models; however, integration texture-based radiomics improve diagnostic accuracy diseases. This study proposes a computer-assisted framework based on multiple approaches. It first trains three Residual Networks (ResNets) with two including discrete wavelet transform gray-level covariance matrix instead Then, it fuses features sets extracted each using cosine transform. Thereafter, further combines fused obtained convolutional neural networks. Finally, support vector machine classifiers are utilized for classification procedure. proposed method validated experimentally benchmark severe respiratory syndrome 2 image dataset. accuracies attained indicate (gray-level matrix, transform) training ResNet-18 (83.22%, 74.9%), ResNet-50 (80.94%, 78.39%), ResNet-101 (80.54%, 77.99%) better than (70.34%, 76.51%, 73.42%) ResNet-18, ResNet-50, ResNet-101, respectively. Furthermore, sensitivity, specificity, accuracy, precision, F1-score achieved after fusion steps 99.47%, 99.72%, 99.60%, 99.60% which proves combining ResNets has boosted its performance. fusing mined several networks only one type approach single network. performance allows be by radiologists attaining diagnosis.

Language: Английский

---

An Ensemble Learning Model for COVID-19 Detection from Blood Test Samples

Sensors, Journal Year: 2022, Volume and Issue: 22(6), P. 2224 - 2224

Published: March 13, 2022

Current research endeavors in the application of artificial intelligence (AI) methods diagnosis COVID-19 disease has proven indispensable with very promising results. Despite these results, there are still limitations real-time detection using reverse transcription polymerase chain reaction (RT-PCR) test data, such as limited datasets, imbalance classes, a high misclassification rate models, and need for specialized identifying best features thus improving prediction rates. This study aims to investigate apply ensemble learning approach develop models effective routine laboratory blood Hence, an machine learning-based system is presented, aiming aid clinicians diagnose this virus effectively. The experiment was conducted custom convolutional neural network (CNN) first-stage classifier 15 supervised algorithms second-stage classifier: K-Nearest Neighbors, Support Vector Machine (Linear RBF), Naive Bayes, Decision Tree, Random Forest, MultiLayer Perceptron, AdaBoost, ExtraTrees, Logistic Regression, Linear Quadratic Discriminant Analysis (LDA/QDA), Passive, Ridge, Stochastic Gradient Descent Classifier. Our findings show that model based on DNN ExtraTrees achieved mean accuracy 99.28% area under curve (AUC) 99.4%, while AdaBoost gave AUC 98.8% San Raffaele Hospital dataset, respectively. comparison proposed other state-of-the-art approaches same dataset shows method outperforms several diagnostics methods.

Language: Английский

---

RESCOVIDTCNnet: A residual neural network-based framework for COVID-19 detection using TCN and EWT with chest X-ray images

Expert Systems with Applications, Journal Year: 2022, Volume and Issue: 204, P. 117410 - 117410

Published: April 27, 2022

Language: Английский

---

A quantization assisted U-Net study with ICA and deep features fusion for breast cancer identification using ultrasonic data

PeerJ Computer Science, Journal Year: 2021, Volume and Issue: 7, P. e805 - e805

Published: Dec. 16, 2021

Breast cancer is one of the leading causes death in women worldwide-the rapid increase breast has brought about more accessible diagnosis resources. The ultrasonic modality for relatively cost-effective and valuable. Lesion isolation images a challenging task due to its robustness intensity similarity. Accurate detection lesions using can reduce rates. In this research, quantization-assisted U-Net approach segmentation proposed. It contains two step segmentation: (1) (2) quantization. quantization assists U-Net-based order isolate exact lesion areas from sonography images. Independent Component Analysis (ICA) method then uses isolated extract features are fused with deep automatic features. Public ultrasonic-modality-based datasets such as Ultrasound Images Dataset (BUSI) Open Access Database Raw Ultrasonic Signals (OASBUD) used evaluation comparison. OASBUD data extracted same However, classification was done after feature regularization lasso method. obtained results allow us propose computer-aided design (CAD) system identification modalities.

Language: Английский

---

An Optimized Decision Support Model for COVID-19 Diagnostics Based on Complex Fuzzy Hypersoft Mapping

Mathematics, Journal Year: 2022, Volume and Issue: 10(14), P. 2472 - 2472

Published: July 15, 2022

COVID-19 has shaken the entire world economy and affected millions of people in a brief period. numerous overlapping symptoms with other upper respiratory conditions, making it hard for diagnosticians to diagnose correctly. Several mathematical models have been presented its diagnosis treatment. This article delivers framework based on novel agile fuzzy-like arrangement, namely, complex fuzzy hypersoft (CFHS) set, which is formation (CF) set (an extension soft set). First, elementary theory CFHS developed, considers amplitude term (A-term) phase (P-term) numbers simultaneously tackle uncertainty, ambivalence, mediocrity data. In two components, this new hybrid versatile. provides access broad spectrum membership function values by broadening them unit circle an Argand plane incorporating additional term, P-term, accommodate data’s periodic nature. Second, categorizes distinct attribute into corresponding sub-valued sets better understanding. The CFHS-mapping inverse mapping (INM) can manage such issues. Our proposed validated study establishing link between medicines. For types, table constructed relying interval [0,1]. computation CFHS-mapping, identifies disease selects optimum medication Furthermore, generalized provided, help specialist extract patient’s health record predict how long will take overcome infection.

Language: Английский

---