Fake or real news about COVID-19? Pretrained transformer model to detect potential misleading news

The European Physical Journal Special Topics, Journal Year: 2022, Volume and Issue: 231(18-20), P. 3347 - 3356

Published: Jan. 13, 2022

Language: Английский

---

ETECADx: Ensemble Self-Attention Transformer Encoder for Breast Cancer Diagnosis Using Full-Field Digital X-ray Breast Images

Diagnostics, Journal Year: 2022, Volume and Issue: 13(1), P. 89 - 89

Published: Dec. 28, 2022

Early detection of breast cancer is an essential procedure to reduce the mortality rate among women. In this paper, a new AI-based computer-aided diagnosis (CAD) framework called ETECADx proposed by fusing benefits both ensemble transfer learning convolutional neural networks as well self-attention mechanism vision transformer encoder (ViT). The accurate and precious high-level deep features are generated via backbone network, while used diagnose probabilities in two approaches: Approach A (i.e., binary classification) B multi-classification). To build CAD system, benchmark public multi-class INbreast dataset used. Meanwhile, private real images collected annotated expert radiologists validate prediction performance framework. promising evaluation results achieved using mammograms with overall accuracies 98.58% 97.87% for approaches, respectively. Compared individual networks, model improves 6.6% 4.6% approaches. hybrid shows further improvement when ViT-based network 8.1% 6.2% diagnosis, For validation purposes images, system provides encouraging 97.16% 89.40% has capability predict lesions single mammogram average 0.048 s. Such could be useful helpful assist practical applications providing second supporting opinion distinguishing various malignancies.

Language: Английский

---

Cancer Prognosis and Diagnosis Methods Based on Ensemble Learning

ACM Computing Surveys, Journal Year: 2023, Volume and Issue: 55(12), P. 1 - 34

Published: Jan. 17, 2023

Ensemble methods try to improve performance via integrating different kinds of input data, features, or learning algorithms. In addition other areas, they are finding their applications in cancer prognosis and diagnosis. However, this area, the research community is lagging behind technology. A systematic review along with a taxonomy on ensemble used diagnosis can pave way for keep pace technology even lead trend. article, we first present an overview existing relevant surveys highlight shortcomings, which raise need new survey focusing Classifiers (ECs) types. Then, exhaustively methods, including traditional ones as well those based deep learning. The leads identification best-studied types, best related purposes, prevailing data most common decision-making strategies, evaluating methodologies. Moreover, establish future directions researchers interested following trends working less-studied aspects area.

Language: Английский

---

Artificial intelligence on COVID-19 pneumonia detection using chest xray images

PLoS ONE, Journal Year: 2021, Volume and Issue: 16(10), P. e0257884 - e0257884

Published: Oct. 14, 2021

Recent studies show the potential of artificial intelligence (AI) as a screening tool to detect COVID-19 pneumonia based on chest x-ray (CXR) images. However, issues datasets and study designs from medical technical perspectives, well questions vulnerability robustness AI algorithms have emerged. In this study, we address these with more realistic development AI-driven detection models by generating our own data through retrospective clinical augment dataset aggregated external sources. We optimized five deep learning architectures, implemented strategies manipulating distribution quantitatively compare designs, introduced several scenarios evaluate diagnostic performance models. At current level availability, model depends hyperparameter tuning has less dependency quantity data. InceptionV3 attained highest in distinguishing normal CXR two-class scenario sensitivity (Sn), specificity (Sp), positive predictive value (PPV) 96%. The higher general 91-96% Sn, 94-98% Sp, 90-96% PPV three-class compared four-class scenario. accuracy, F1-score, g-mean 96% For detection, 86% 99% 91% an AUC 0.99 CXR. Its capability differentiating non-COVID-19 0.98 micro-average for other classes.

Language: Английский

---

Arabic Document Classification: Performance Investigation of Preprocessing and Representation Techniques

Mathematical Problems in Engineering, Journal Year: 2022, Volume and Issue: 2022, P. 1 - 16

Published: April 30, 2022

With the increasing number of online social posts, review comments, and digital documentations, Arabic text classification (ATC) task has been hugely required for many spontaneous natural language processing (NLP) applications, especially within coronavirus pandemics. The variations in meaning same words could directly affect performance any AI-based framework. This work aims to identify effectiveness machine learning (ML) algorithms through preprocessing representation techniques. is measured via different Basically, ATC process influenced by several factors such as stemming preprocessing, method feature extraction selection, nature datasets, algorithm. To improve overall performance, techniques are mainly used convert each word into its root decrease dimension among datasets. Feature selection always play crucial roles represent a meaningful way accuracy rate. selected classifiers this study performed based on various algorithms. evaluation results compared using multinomial Naive Bayes (MNB), Bernoulli (BNB), Stochastic Gradient Descent (SGD), Support Vector Classifier (SVC), Logistic Regression (LR), Linear SVC. All these AI evaluated five balanced unbalanced benchmark datasets: BBC corpus, CNN Open-Source corpus (OSAc), ArCovidVac, AlKhaleej. show that strongly depends technique, methods datasets used. For considered linear SVC outperformed other when prominent features selected.

Language: Английский

---