Automated detection of COVID-19 cases using deep neural networks with X-ray images

Öztürk, Tülin; Talo, Muhammed; Yildirim, Eylul Azra; Baloğlu, Ulaş Baran; Yıldırım, Özal; Acharya, U. Rajendra

doi:10.1016/j.compbiomed.2020.103792

Cited by 2,260 publications

(2,075 citation statements)

References 51 publications

Supporting

Mentioning

2,031

Contrasting

Unclassified

Order By: Relevance

“…Risk factors for silent hypoxemia are old age and having diabetes [9]. Therefore, early detection of silent hypoxemia such as using prehospital pulse oximetry [5], or radiology imaging [16,17] might be used as a red ag sign of impending danger of eminent cardiac arrest or sudden respiratory failure.…”

Section: Discussionmentioning

confidence: 99%

Possible silent hypoxemia in a COVID-19 patient: a case report

Siswanto

Gani

Fauzi

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: It has been hypothesized that silent hypoxemia is the cause of the rapid progressive respiratory failure with severe hypoxia that occurs in some patients with COVID-19 without warning. Here, we reported one COVID-19 case with the possibility of silent hypoxemia. Case presentation: A 60-year-old male presented with complaints of cough that he felt starting two weeks before admission without any breathing difficulty. Complaints were accompanied by fever, runny nose and sore throat. Vital signs examination showed blood pressure 130/75 mmHg, pulse 84 times per minute, normal respiratory rate (RR) of 21 times per minute, body temperature 36.5 C, and 99% oxygen saturation with oxygen via nasal cannula 3 liters per minute were recorded. On physical examination, an increase in vesicular sounds and crackles in both lungs were identified. Chest x-ray showed bilateral pneumonia. Nasopharyngeal and oropharyngeal swab real-time polymerase chain reaction tests for COVID-19 were positive. On the third day of treatment, the patient complained of worsening of shortness of breath, but his RR was still normal with 22 times per minute. On the fifth day of treatment, the patient experienced severe shortness of breath with a RR of 38 times per minute. The patient was then intubated and his blood gas analysis showed respiratory alkalosis (pH 7.54, PaO2 58.9 mmHg, PaCO2 31.1 mmHg, HCO3 26.9 mEq/L, SaO2 94.7%). On the eighth day of treatment, his condition deteriorated starting in the morning, with blood pressure 80/40 mmHg with norepinephrine support, pulse 109 times per minute, and 72% SpO2 with ventilator. In the afternoon, the patient experienced cardiac arrest and underwent basic life support, then resumed strained breathing with return of spontaneous circulation. Blood gas analysis showed severe respiratory acidosis (pH 7.07, PaO2 58.1 mmHg, PaCO2 108.9 mmHg, HCO3 32.1 mEq /L, SaO2 78.7%). Three hours later, he suffered cardiac arrest again, but was unable to be resuscitated. The patient eventually died.Conclusions: Silent hypoxemia might be considered as an early clinical sign of deterioration of patients with COVID-19, thus, the physician may be able to intervene early and decrease its morbidity and mortality.

show abstract

Section: Discussionmentioning

confidence: 99%

Possible silent hypoxemia in a COVID-19 patient: a case report

Siswanto

Gani

Fauzi

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The overall accuracy was 89.6% while average accuracy of detecting COVID-19 was 96.6%. To test the stability and robustness, CoroNet was evaluated on the dataset prepared by Ozturk et al 13 with an accuracy of 90%. 14 .…”

Section: Introductionmentioning

confidence: 99%

“…Ozturk et al developed DarkNet model based on the you only look once (YOLO) system to detect and classify COVID-1913 . Their model achieved the accuracy of 98.08% for classifying COVID-19 and noninfections and 87.02% for distinguish COVID-19 from COVID-19, no-ndings and GP.…”

mentioning

confidence: 99%

Differentiating novel coronavirus pneumonia from general pneumonia based on machine learning

Liu

Wang

Liu

et al. 2020

Preprint

View full text Add to dashboard Cite

Background: Chest CT screening as supplementary means is crucial in diagnosing novel coronavirus pneumonia (COVID-19) with high sensitivity and popularity. Machine learning was adept in discovering intricate structures from CT images and achieved expert-level performance in medical image analysis. Methods: An integrated machine learning framework on chest CT images for differentiating COVID-19 from general pneumonia (GP) was developed and validated. Seventy-three confirmed COVID-19 cases were consecutively enrolled together with twenty-seven confirmed general pneumonia patients from Ruian People’s Hospital, from January 2020 to March 2020. To accurately classify COVID-19, region of interest (ROI) delineation was implemented based on ground glass opacities (GGOs) before feature extraction. Then, 34 statistical texture features of COVID-19 and GP ROI images were extracted, including 13 gray level co-occurrence matrix (GLCM) features, 15 gray level-gradient co-occurrence matrix (GLGCM) features and 6 histogram features. High dimensional features impact the classification performance. Thus, ReliefF algorithm was leveraged to select features. The relevance of each features was the average weights calculated by ReliefF in n times. Features with relevance lager than the empirically set threshold T were selected. After feature selection, the optimal feature set along with 4 other selected feature combinations for comparison were applied to the ensemble of bagged tree (EBT) and four other machine learning classifiers including support vector machine (SVM), logistic regression (LR), decision tree (DT), and K-nearest neighbor with Minkowski distance equal weight (KNN) using 10-fold cross-validation. Results and Conclusions: The classification accuracy (ACC), sensitivity (SEN), specificity (SPE) of our proposed method yield 94.16%, 88.62% and 100.00%, respectively. The area under the receiver operating characteristic curve (AUC) was 0.99. The experimental results indicate that the EBT algorithm with statistical textural features based on GGOs for differentiating COVID-19 from general pneumonia achieved high transferability, efficiency, specificity, sensitivity, and impressive accuracy, which is beneficial for inexperienced doctors to more accurately diagnose COVID-19 and essential for controlling the spread of the disease.

show abstract

“…Their model achieved good accuracy for classifying two classes and three classes, respectively. Tulin et al [19] suggested in their study, the (DarkNet) Model as a classifier. This model, they used 17 convolutional layers with different filtering.…”

Section: Introductionmentioning

confidence: 99%

CCBlock based on deep learning for diagnosis COVID-19 in chest x-ray image

Al-Bawi

Al-Kaabi

Jeryo

et al. 2020

Preprint

View full text Add to dashboard Cite

Purpose: COVID-19 pandemic continues to hit countries one after the other and has dramatically affected the health and well-being of the world's population. With the daily increase in the number of people with this disease, the impressive speed of spread and the delay in the results of PCR analysis, it may cause the disease to spread more broadly. Therefore it is necessary to consider finding alternative methods of detection and diagnosis COVID-19 to prohibit the spread of the disease among people. Convolutional Neural Network (CNN) automated detection systems have shown auspicious results in detecting patients with COVID-19 through radiography; thus, we suggest them as an alternative option to diagnose COVID-19.Method: In this study, an early screening model based on the enhancement of classical Visual Geometry Group Network (VGG) with Convolutional Covid Block (CCBlock) was proposed to detect and distinguish COVID-19 from Pneumonia, and healthy people using chest X-ray radiographs. The data set used for model testing is the x-ray images available on public platforms, which consist of 1,828 x-ray images, including 310 images for confirmed COVID-19 patients, 864 images for pneumonia patients, and 654 images for healthy people.Results: The experiment result of the dataset showed that the added enhancements to the classical VGG network with X-ray imaging provide the highest detection performance and overall accuracy of 98.52% for two classes and 95.34% accuracy for three classes.Conclusions: Considering the achievement results obtained, it was found that utilizing the enhanced VGG deep neural network helps radiologists automatically diagnose COVID-19 in X-ray images.

show abstract

Automated detection of COVID-19 cases using deep neural networks with X-ray images

Cited by 2,260 publications

References 51 publications

Possible silent hypoxemia in a COVID-19 patient: a case report

Possible silent hypoxemia in a COVID-19 patient: a case report

Differentiating novel coronavirus pneumonia from general pneumonia based on machine learning

CCBlock based on deep learning for diagnosis COVID-19 in chest x-ray image

Contact Info

Product

Resources

About