1
|
Celik G. CovidCoughNet: A new method based on convolutional neural networks and deep feature extraction using pitch-shifting data augmentation for covid-19 detection from cough, breath, and voice signals. Comput Biol Med 2023; 163:107153. [PMID: 37321101 PMCID: PMC10249348 DOI: 10.1016/j.compbiomed.2023.107153] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/09/2023] [Revised: 05/25/2023] [Accepted: 06/07/2023] [Indexed: 06/17/2023]
Abstract
This study proposes a new deep learning-based method that demonstrates high performance in detecting Covid-19 disease from cough, breath, and voice signals. This impressive method, named CovidCoughNet, consists of a deep feature extraction network (InceptionFireNet) and a prediction network (DeepConvNet). The InceptionFireNet architecture, based on Inception and Fire modules, was designed to extract important feature maps. The DeepConvNet architecture, which is made up of convolutional neural network blocks, was developed to predict the feature vectors obtained from the InceptionFireNet architecture. The COUGHVID dataset containing cough data and the Coswara dataset containing cough, breath, and voice signals were used as the data sets. The pitch-shifting technique was used to data augmentation the signal data, which significantly contributed to improving performance. Additionally, Chroma features (CF), Root mean square energy (RMSE), Spectral centroid (SC), Spectral bandwidth (SB), Spectral rolloff (SR), Zero crossing rate (ZCR), and Mel frequency cepstral coefficients (MFCC) feature extraction techniques were used to extract important features from voice signals. Experimental studies have shown that using the pitch-shifting technique improved performance by around 3% compared to raw signals. When the proposed model was used with the COUGHVID dataset (Healthy, Covid-19, and Symptomatic), a high performance of 99.19% accuracy, 0.99 precision, 0.98 recall, 0.98 F1-Score, 97.77% specificity, and 98.44% AUC was achieved. Similarly, when the voice data in the Coswara dataset was used, higher performance was achieved compared to the cough and breath studies, with 99.63% accuracy, 100% precision, 0.99 recall, 0.99 F1-Score, 99.24% specificity, and 99.24% AUC. Moreover, when compared with current studies in the literature, the proposed model was observed to exhibit highly successful performance. The codes and details of the experimental studies can be accessed from the relevant Github page: (https://github.com/GaffariCelik/CovidCoughNet).
Collapse
Affiliation(s)
- Gaffari Celik
- Agri Ibrahim Cecen University, Department of Computer Technology, Agri, Turkey.
| |
Collapse
|
2
|
Kathamuthu ND, Subramaniam S, Le QH, Muthusamy S, Panchal H, Sundararajan SCM, Alrubaie AJ, Zahra MMA. A deep transfer learning-based convolution neural network model for COVID-19 detection using computed tomography scan images for medical applications. ADVANCES IN ENGINEERING SOFTWARE (BARKING, LONDON, ENGLAND : 1992) 2023; 175:103317. [PMID: 36311489 PMCID: PMC9595382 DOI: 10.1016/j.advengsoft.2022.103317] [Citation(s) in RCA: 9] [Impact Index Per Article: 9.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 08/26/2022] [Revised: 10/09/2022] [Accepted: 10/19/2022] [Indexed: 05/26/2023]
Abstract
The Coronavirus (COVID-19) has become a critical and extreme epidemic because of its international dissemination. COVID-19 is the world's most serious health, economic, and survival danger. This disease affects not only a single country but the entire planet due to this infectious disease. Illnesses of Covid-19 spread at a much faster rate than usual influenza cases. Because of its high transmissibility and early diagnosis, it isn't easy to manage COVID-19. The popularly used RT-PCR method for COVID-19 disease diagnosis may provide false negatives. COVID-19 can be detected non-invasively using medical imaging procedures such as chest CT and chest x-ray. Deep learning is the most effective machine learning approach for examining a considerable quantity of chest computed tomography (CT) pictures that can significantly affect Covid-19 screening. Convolutional neural network (CNN) is one of the most popular deep learning techniques right now, and its gaining traction due to its potential to transform several spheres of human life. This research aims to develop conceptual transfer learning enhanced CNN framework models for detecting COVID-19 with CT scan images. Though with minimal datasets, these techniques were demonstrated to be effective in detecting the presence of COVID-19. This proposed research looks into several deep transfer learning-based CNN approaches for detecting the presence of COVID-19 in chest CT images.VGG16, VGG19, Densenet121, InceptionV3, Xception, and Resnet50 are the foundation models used in this work. Each model's performance was evaluated using a confusion matrix and various performance measures such as accuracy, recall, precision, f1-score, loss, and ROC. The VGG16 model performed much better than the other models in this study (98.00 % accuracy). Promising outcomes from experiments have revealed the merits of the proposed model for detecting and monitoring COVID-19 patients. This could help practitioners and academics create a tool to help minimal health professionals decide on the best course of therapy.
Collapse
Affiliation(s)
- Nirmala Devi Kathamuthu
- Department of Computer Science and Engineering, Kongu Engineering College (Autonomous), Perundurai, Erode, Tamil Nadu, India
| | - Shanthi Subramaniam
- Department of Computer Science and Engineering, Kongu Engineering College (Autonomous), Perundurai, Erode, Tamil Nadu, India
| | - Quynh Hoang Le
- Institute of Research and Development, Duy Tan University, Da Nang, Vietnam
- School of Medicine and Pharmacy, Duy Tan University, Da Nang, Vietnam
| | - Suresh Muthusamy
- Department of Electronics and Communication Engineering, Kongu Engineering College (Autonomous), Perundurai, Erode, Tamil Nadu, India
| | - Hitesh Panchal
- Department of Mechanical Engineering, Government Engineering College, Patan, Gujarat, India
| | | | - Ali Jawad Alrubaie
- Department of Medical Instrumentation Techniques Engineering, Al- Mustaqbal University College, 51001, Hilla, Iraq
| | - Musaddak Maher Abdul Zahra
- Computer Techniques Engineering Department, Al-Mustaqbal University College, Hillah 51001, Iraq
- Electrical Engineering Department, College of Engineering, University of Babylon, Hilla, Babil, Iraq
| |
Collapse
|
3
|
Bhatele KR, Jha A, Tiwari D, Bhatele M, Sharma S, Mithora MR, Singhal S. COVID-19 Detection: A Systematic Review of Machine and Deep Learning-Based Approaches Utilizing Chest X-Rays and CT Scans. Cognit Comput 2022:1-38. [PMID: 36593991 PMCID: PMC9797382 DOI: 10.1007/s12559-022-10076-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/24/2022] [Accepted: 11/15/2022] [Indexed: 12/30/2022]
Abstract
This review study presents the state-of-the-art machine and deep learning-based COVID-19 detection approaches utilizing the chest X-rays or computed tomography (CT) scans. This study aims to systematically scrutinize as well as to discourse challenges and limitations of the existing state-of-the-art research published in this domain from March 2020 to August 2021. This study also presents a comparative analysis of the performance of four majorly used deep transfer learning (DTL) models like VGG16, VGG19, ResNet50, and DenseNet over the COVID-19 local CT scans dataset and global chest X-ray dataset. A brief illustration of the majorly used chest X-ray and CT scan datasets of COVID-19 patients utilized in state-of-the-art COVID-19 detection approaches are also presented for future research. The research databases like IEEE Xplore, PubMed, and Web of Science are searched exhaustively for carrying out this survey. For the comparison analysis, four deep transfer learning models like VGG16, VGG19, ResNet50, and DenseNet are initially fine-tuned and trained using the augmented local CT scans and global chest X-ray dataset in order to observe their performance. This review study summarizes major findings like AI technique employed, type of classification performed, used datasets, results in terms of accuracy, specificity, sensitivity, F1 score, etc., along with the limitations, and future work for COVID-19 detection in tabular manner for conciseness. The performance analysis of the four majorly used deep transfer learning models affirms that Visual Geometry Group 19 (VGG19) model delivered the best performance over both COVID-19 local CT scans dataset and global chest X-ray dataset.
Collapse
Affiliation(s)
| | - Anand Jha
- RJIT BSF Academy, Tekanpur, Gwalior India
| | | | | | | | | | | |
Collapse
|
4
|
Khalifa NEM, Manogaran G, Taha MHN, Loey M. A deep learning semantic segmentation architecture for COVID-19 lesions discovery in limited chest CT datasets. EXPERT SYSTEMS 2022; 39:e12742. [PMID: 34177038 PMCID: PMC8209878 DOI: 10.1111/exsy.12742] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 03/04/2021] [Revised: 04/28/2021] [Accepted: 04/30/2021] [Indexed: 05/10/2023]
Abstract
During the epidemic of COVID-19, Computed Tomography (CT) is used to help in the diagnosis of patients. Most current studies on this subject appear to be focused on broad and private annotated data which are impractical to access from an organization, particularly while radiologists are fighting the coronavirus disease. It is challenging to equate these techniques since they were built on separate datasets, educated on various training sets, and tested using different metrics. In this research, a deep learning semantic segmentation architecture for COVID-19 lesions detection in limited chest CT datasets will be presented. The proposed model architecture consists of the encoder and the decoder components. The encoder component contains three layers of convolution and pooling, while the decoder contains three layers of deconvolutional and upsampling. The dataset consists of 20 CT scans of lungs belongs to 20 patients from two sources of data. The total number of images in the dataset is 3520 CT scans with its labelled images. The dataset is split into 70% for the training phase and 30% for the testing phase. Images of the dataset are passed through the pre-processing phase to be resized and normalized. Five experimental trials are conducted through the research with different images selected for the training and the testing phases for every trial. The proposed model achieves 0.993 in the global accuracy, and 0.987, 0.799, 0.874 for weighted IoU, mean IoU and mean BF score accordingly. The performance metrics such as precision, sensitivity, specificity and F1 score strengthens the obtained results. The proposed model outperforms the related works which use the same dataset in terms of performance and IoU metrics.
Collapse
Affiliation(s)
- Nour Eldeen M. Khalifa
- Department of Information TechnologyFaculty of Computers & Artificial Intelligence, Cairo UniversityCairoEgypt
| | - Gunasekaran Manogaran
- University of CaliforniaDavisCaliforniaUSA
- College of Information and Electrical EngineeringAsia UniversityTaichungTaiwan
| | - Mohamed Hamed N. Taha
- Department of Information TechnologyFaculty of Computers & Artificial Intelligence, Cairo UniversityCairoEgypt
| | - Mohamed Loey
- Department of Computer Science, Faculty of Computers and Artificial IntelligenceBenha UniversityBenhaEgypt
| |
Collapse
|
5
|
Rashno E, Akbari A, Nasersharif B. Uncertainty handling in convolutional neural networks. Neural Comput Appl 2022; 34:16753-16769. [PMID: 35756151 PMCID: PMC9206226 DOI: 10.1007/s00521-022-07313-2] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/09/2019] [Accepted: 04/18/2022] [Indexed: 11/26/2022]
Abstract
The performance of convolutional neural networks is degraded by noisy data, especially in the test phase. To address this challenge, a new convolutional neural network structure with data indeterminacy handling in the neutrosophic (NS) domain, named as Neutrosophic Convolutional Neural Networks, is proposed for image classification. For this task, images are firstly mapped from the pixel domain to three sets true (T), indeterminacy (I) and false (F) in NS domain by the proposed method. Then, NCNN with two parallel paths, one with the input of T and another with I, is constructed followed by an appropriate combination of paths to generate the final output. Here, two paths are trained simultaneously, and neural network weights are updated using back propagation algorithm. The effectiveness of NCNN to handle noisy data is analyzed mathematically in terms of the weights update rule. Proposed two paths NS idea is applied to two basic models: CNN and VGG-Net to construct NCNN and NVGG-Net, respectively. The proposed method has been evaluated on MNIST, CIFAR-10 and CIFAR-100 datasets contaminated with 20 levels of Gaussian noise. Results show that two-path NCNN outperforms CNN by 5.11% and 2.21% in 5 pairs (training, test) with different levels of noise on MNIST and CIFAR-10 datasets, respectively. Finally, NVGG-Net increases the accuracy by 3.09% and 2.57% compared to VGG-Net on CIFAR-10 and CIFAR-100 datasets, respectively.
Collapse
Affiliation(s)
- Elyas Rashno
- Department of Computer Engineering, Iran University of Science and Technology, Narmak, Tehran, 1684613114 Iran
| | - Ahmad Akbari
- Department of Computer Engineering, Iran University of Science and Technology, Narmak, Tehran, 1684613114 Iran
| | - Babak Nasersharif
- Department of Computer Engineering, K.N. Toosi University of Technology, Tehran, Iran
| |
Collapse
|
6
|
Quek SG, Selvachandran G, Ajay D, Chellamani P, Taniar D, Fujita H, Duong P, Son LH, Giang NL. New concepts of pentapartitioned neutrosophic graphs and applications for determining safest paths and towns in response to COVID-19. COMPUTATIONAL AND APPLIED MATHEMATICS 2022; 41:151. [PMCID: PMC9022744 DOI: 10.1007/s40314-022-01823-4] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/03/2021] [Revised: 10/03/2021] [Accepted: 03/03/2022] [Indexed: 06/18/2023]
Abstract
Pentapartitioned neutrosophic sets are a generalization of the single-valued and quadri-partitioned single-valued neutrosophic sets, and utilizes five symbol-valued neutrosophic logic. In this paper, we introduce some novel concepts regarding pentapartitioned neutrosophic graphs (PPNGs), and emphasize the effectiveness at interpreting extremely heterogeneous data that are prevalent in our daily life, particularly data gathered from various different sources which are becoming increasingly common place in the current times. The applicability of the proposed PPNG is demonstrated by applying the PPNGs on a potential real-life scenario on responding to the spread of COVID-19, where PPNGs are used to determine the safest path of travel and the safest place to stay to minimize the chances of getting infected. Both of this information have proven to be vital aspects in the efforts to combat the spread of the COVID-19 pandemic while providing the necessary support to the domestic economies, most of which are currently in recession due to the adverse effects brought upon by the pandemic. Hence, the PPNGs are applicable to all countries around the world and can be used under any circumstances such as pandemics or even in regular situations to optimize the travelling time and distance.
Collapse
Affiliation(s)
- Shio Gai Quek
- Department of Actuarial Science and Applied Statistics, Faculty of Business and Management, UCSI University, Jalan Menara Gading, Cheras, 56000 Kuala Lumpur, Malaysia
| | - Ganeshsree Selvachandran
- Department of Actuarial Science and Applied Statistics, Faculty of Business and Management, UCSI University, Jalan Menara Gading, Cheras, 56000 Kuala Lumpur, Malaysia
| | - D. Ajay
- Department of Mathematics, Sacred Heart College (Autonomous), Tamil Nadu, Tirupattur, India
| | - P. Chellamani
- Department of Mathematics, Sacred Heart College (Autonomous), Tamil Nadu, Tirupattur, India
| | - David Taniar
- Faculty of Information Technology, Monash University, Wellington Rd, Clayton, VIC 3800 Australia
| | - Hamido Fujita
- Chairman of Intelligent Software Methodologies and Technologies Incorporated Association, (i-SOMET Inc), Morioka, 020-0104 Japan
- Regional Research Center, Iwate Prefectural University, Iwate, Japan
| | - Phet Duong
- Faculty of Information Technology, HUTECH University, Ho Chi Minh City, Vietnam
| | - Le Hoang Son
- VNU Information Technology Institute, Vietnam National University, Hanoi, Vietnam
| | - Nguyen Long Giang
- Institute of Information Technology, Vietnam Academy of Science and Technology, Hanoi, Vietnam
| |
Collapse
|
7
|
Deng H, Li X. AI-Empowered Computational Examination of Chest Imaging for COVID-19 Treatment: A Review. Front Artif Intell 2021; 4:612914. [PMID: 34368756 PMCID: PMC8333868 DOI: 10.3389/frai.2021.612914] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/02/2020] [Accepted: 06/23/2021] [Indexed: 12/21/2022] Open
Abstract
Since the first case of coronavirus disease 2019 (COVID-19) was discovered in December 2019, COVID-19 swiftly spread over the world. By the end of March 2021, more than 136 million patients have been infected. Since the second and third waves of the COVID-19 outbreak are in full swing, investigating effective and timely solutions for patients' check-ups and treatment is important. Although the SARS-CoV-2 virus-specific reverse transcription polymerase chain reaction test is recommended for the diagnosis of COVID-19, the test results are prone to be false negative in the early course of COVID-19 infection. To enhance the screening efficiency and accessibility, chest images captured via X-ray or computed tomography (CT) provide valuable information when evaluating patients with suspected COVID-19 infection. With advanced artificial intelligence (AI) techniques, AI-driven models training with lung scans emerge as quick diagnostic and screening tools for detecting COVID-19 infection in patients. In this article, we provide a comprehensive review of state-of-the-art AI-empowered methods for computational examination of COVID-19 patients with lung scans. In this regard, we searched for papers and preprints on bioRxiv, medRxiv, and arXiv published for the period from January 1, 2020, to March 31, 2021, using the keywords of COVID, lung scans, and AI. After the quality screening, 96 studies are included in this review. The reviewed studies were grouped into three categories based on their target application scenarios: automatic detection of coronavirus disease, infection segmentation, and severity assessment and prognosis prediction. The latest AI solutions to process and analyze chest images for COVID-19 treatment and their advantages and limitations are presented. In addition to reviewing the rapidly developing techniques, we also summarize publicly accessible lung scan image sets. The article ends with discussions of the challenges in current research and potential directions in designing effective computational solutions to fight against the COVID-19 pandemic in the future.
Collapse
Affiliation(s)
- Hanqiu Deng
- Department of Electrical and Computer Engineering, University of Alberta, Edmonton, AB, Canada
- School of Aerospace Engineering, Beijing Institute of Technology, Beijing, China
| | - Xingyu Li
- Department of Electrical and Computer Engineering, University of Alberta, Edmonton, AB, Canada
| |
Collapse
|
8
|
A Pneumonia Diagnosis Scheme Based on Hybrid Features Extracted from Chest Radiographs Using an Ensemble Learning Algorithm. JOURNAL OF HEALTHCARE ENGINEERING 2021; 2021:8862089. [PMID: 33728035 PMCID: PMC7935583 DOI: 10.1155/2021/8862089] [Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 01/04/2021] [Revised: 02/06/2021] [Accepted: 02/10/2021] [Indexed: 12/16/2022]
Abstract
Pneumonia is a fatal disease responsible for almost one in five child deaths worldwide. Many developing countries have high mortality rates due to pneumonia because of the unavailability of proper and timely diagnostic measures. Using machine learning-based diagnosis methods can help to detect the disease early and in less time and cost. In this study, we proposed a novel method to determine the presence of pneumonia and identify its type (bacterial or viral) through analyzing chest radiographs. We performed a three-class classification based on features containing diverse information of the samples. After using an augmentation technique to balance the dataset's sample sizes, we extracted the chest X-ray images' statistical features, as well as global features by employing a deep learning architecture. We then combined both sets of features and performed the final classification using the RandomForest classifier. A feature selection method was also incorporated to identify the features with the highest relevance. We tested the proposed method on a widely used (but relabeled) chest radiograph dataset to evaluate its performance. The proposed model can classify the dataset's samples with an 86.30% classification accuracy and 86.03% F-score, which assert the model's efficacy and reliability. However, results show that the classifier struggles while distinguishing between viral and bacterial pneumonia samples. Implementing this method will provide a fast and automatic way to detect pneumonia in a patient and identify its type.
Collapse
|
9
|
Hezam IM, Nayeem MK, Foul A, Alrasheedi AF. COVID-19 Vaccine: A neutrosophic MCDM approach for determining the priority groups. RESULTS IN PHYSICS 2021; 20:103654. [PMID: 33520620 PMCID: PMC7832528 DOI: 10.1016/j.rinp.2020.103654] [Citation(s) in RCA: 26] [Impact Index Per Article: 8.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 11/06/2020] [Revised: 11/22/2020] [Accepted: 11/24/2020] [Indexed: 05/05/2023]
Abstract
Since the outbreak of COVID-19, most of the countries around the world have been confronting the loss of lives, struggling with several economical parameters, i.e. low GDP growth, increasing unemployment rate, and others. It's been 11 months since we are struggling with COVID-19 and some of the countries already facing the second wave of COVID-19. To get rid of these problems, inventions of a vaccine and its optimum distribution is a key factor. Many companies are trying to find a vaccine, but for nearly 8 billion people it would be impossible to find a vaccine. Thus, the competition arises, and this competition would be too intense to satisfy all the people of a country with the vaccine. Therefore, at first, governments must identify priority groups for allocating COVID-19 vaccine doses. In this work, we identify four main criteria and fifteen sub-criteria based on age, health status, a woman's status, and the kind of job. The main and sub-criteria will be evaluated using a neutrosophic Analytic Hierarchy Process (AHP). Then, the COVID-19 vaccine alternatives will be ranked using a neutrosophic TOPSIS method. All the results obtained indicate that the healthcare personnel, people with high-risk health, elderly people, essential workers, pregnant and lactating mothers are the most prioritized people to take the vaccine dose first. Also, the results indicate that the most appropriate vaccine for patients and health workers have priority over other alternative vaccines.
Collapse
Affiliation(s)
- Ibrahim M Hezam
- Department of Statistics & Operations Research, College of Sciences, King Saud University, Riyadh, Saudi Arabia
- Department of Mathematics, Ibb University, Ibb, Yemen
| | - Moddassir Khan Nayeem
- Department of Industrial and Production Engineering, American International University-Bangladesh, Bangladesh
| | - Abdelaziz Foul
- Department of Statistics & Operations Research, College of Sciences, King Saud University, Riyadh, Saudi Arabia
| | - Adel Fahad Alrasheedi
- Department of Statistics & Operations Research, College of Sciences, King Saud University, Riyadh, Saudi Arabia
| |
Collapse
|