COVLIAS 2.0-cXAI: Cloud-Based Explainable Deep Learning System for COVID-19 Lesion Localization in Computed Tomography Scans

Diagnostics, Journal Year: 2022, Volume and Issue: 12(6), P. 1482 - 1482

Published: June 16, 2022

Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based AI, “COVLIAS 2.0-cXAI” using four kinds class activation maps (CAM) models. Methodology: Our cohort consisted ~6000 CT slices from two sources (Croatia, 80 patients Italy, 15 control patients). COVLIAS 2.0-cXAI design three stages: (i) automated segmentation hybrid deep learning ResNet-UNet model by automatic adjustment Hounsfield units, hyperparameter optimization, parallel distributed training, (ii) classification DenseNet (DN) models (DN-121, DN-169, DN-201), (iii) CAM visualization techniques: gradient-weighted mapping (Grad-CAM), Grad-CAM++, score-weighted (Score-CAM), FasterScore-CAM. was validated trained senior radiologists its stability reliability. Friedman test also performed on scores radiologists. Results: resulted in dice similarity 0.96, Jaccard index 0.93, correlation coefficient 0.99, with figure-of-merit 95.99%, while classifier accuracies DN nets DN-201) were 98%, 99% loss ~0.003, ~0.0025, ~0.002 50 epochs, respectively. mean AUC all 0.99 (p < 0.0001). showed 80% scans alignment (MAI) between heatmaps gold standard, score out five, establishing clinical settings. Conclusions: successfully AI localization scans.

Language: Английский

---

A deep learning-based framework for detecting COVID-19 patients using chest X-rays

Multimedia Systems, Journal Year: 2022, Volume and Issue: 28(4), P. 1495 - 1513

Published: March 22, 2022

Language: Английский

---

Detection of COVID-19 using deep learning techniques and classification methods

Information Processing & Management, Journal Year: 2022, Volume and Issue: 59(5), P. 103025 - 103025

Published: July 8, 2022

Language: Английский

---

A multiple-input deep residual convolutional neural network for reservoir permeability prediction

Geoenergy Science and Engineering, Journal Year: 2023, Volume and Issue: 222, P. 211420 - 211420

Published: Jan. 5, 2023

Language: Английский

---

Attention deep learning‐based large‐scale learning classifier for Cassava leaf disease classification

Expert Systems, Journal Year: 2021, Volume and Issue: 39(2)

Published: Nov. 3, 2021

Abstract Cassava is a rich source of carbohydrates, and it vulnerable to virus diseases. Literature survey shows that the image recognition integrated deep learning approach successfully employed for leaf disease classification. Mostly, transfer based on convolutional neural network (CNN) models were applied However, existing approaches are not effective in identifying tiny portion overall area. Identifying focussing regions affected by vital achieving good classification accuracy. An attention‐based into pretrained CNN‐based EfficientNet locate identify infected leaf. Penultimate layer features such as A_EfficientNetB4, A_EfficientNetB5, A_EfficientNetB6 extracted. Next, dimensionality extracted was reduced using kernel principal component analysis. The fused passed stacked ensemble meta‐classifier A two‐stage which first stage employs random forest support vector machine (SVM) prediction followed logistic regression Detailed investigation analysis proposed method, attention, non‐attention‐based with CNN tested publicly available benchmark dataset images. method achieved better performances all experiments than several methods well various attention models. can be used deployable tool agricultural field.

Language: Английский

---

Application of machine learning in CT images and X-rays of COVID-19 pneumonia

Medicine, Journal Year: 2021, Volume and Issue: 100(36), P. e26855 - e26855

Published: Sept. 10, 2021

Coronavirus disease (COVID-19) has spread worldwide. X-ray and computed tomography (CT) are 2 technologies widely used in image acquisition, segmentation, diagnosis, evaluation. Artificial intelligence can accurately segment infected parts CT images, assist doctors improving diagnosis efficiency, facilitate the subsequent assessment of severity patient infection. The medical assistant platform based on machine learning help radiologists make clinical decisions helper screening, treatment. By providing scientific methods for recognition, evaluation, we summarized latest developments application artificial COVID-19 lung imaging, provided guidance inspiration to researchers who fighting virus.

Language: Английский

---

A novel fusion based convolutional neural network approach for classification of COVID-19 from chest X-ray images

Biomedical Signal Processing and Control, Journal Year: 2022, Volume and Issue: 77, P. 103778 - 103778

Published: May 2, 2022

Language: Английский

---

A complete framework for accurate recognition and prognosis of COVID-19 patients based on deep transfer learning and feature classification approach

Artificial Intelligence Review, Journal Year: 2022, Volume and Issue: 55(6), P. 5063 - 5108

Published: Jan. 29, 2022

Language: Английский

---

MDFNet: an unsupervised lightweight network for ear print recognition

Journal of Ambient Intelligence and Humanized Computing, Journal Year: 2022, Volume and Issue: 14(10), P. 13773 - 13786

Published: June 18, 2022

Language: Английский

---

SC2Net: A Novel Segmentation-Based Classification Network for Detection of COVID-19 in Chest X-Ray Images

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2022, Volume and Issue: 26(8), P. 4032 - 4043

Published: May 25, 2022

The pandemic of COVID-19 has become a global crisis in public health, which led to massive number deaths and severe economic degradation. To suppress the spread COVID-19, accurate diagnosis at an early stage is crucial. As popularly used real-time reverse transcriptase polymerase chain reaction (RT-PCR) swab test can be lengthy inaccurate, chest screening with radiography imaging still preferred. However, due limited image data difficulty early-stage diagnosis, existing models suffer from ineffective feature extraction poor network convergence optimisation. tackle these issues, segmentation-based classification network, namely SC2Net, proposed for effective detection x-ray (CXR) images. SC2Net consists two subnets: lung segmentation (CLSeg), spatial attention (SANet). In order supress interference background, CLSeg first applied segment region CXR. segmented then fed SANet COVID-19. shallow yet classifier, takes ResNet-18 as extractor enhances high-level via module. For performance evaluation, COVIDGR 1.0 dataset used, high-quality various severity levels Experimental results have shown that, our average accuracy 84.23% F1 score 81.31% outperforming several state-of-the-art approaches.

Language: Английский

---