Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 20 de 92
Filter
Add filters

Year range
1.
Mol Med Rep ; 23(5)2021 May.
Article in English | MEDLINE | ID: covidwho-1167801

ABSTRACT

Although the COVID­19 epidemic has lasted for months, it has not yet been successfully controlled, and little is known about neonatal COVID­19. Therefore, literature search was conducted for references in PubMed, Science Direct, ProQuest, Web of Science and China National Knowledge Infrastructure for detailed case reports on neonatal COVID­19 published as of July 15, 2020, to facilitate the clinical treatment, epidemic prevention and control of neonatal COVID­19. Forty nonoverlapping case reports focusing mainly on the demographic characteristics, transmission modes, clinical features, treatments and prognosis of neonatal COVID­19, including 3 in Chinese and 37 in English, were available.


Subject(s)
/pathology , /physiology , Antibodies, Viral/analysis , Antiviral Agents/therapeutic use , Asymptomatic Diseases , /transmission , Humans , Infant, Newborn , Milk, Human/virology , RNA, Viral/metabolism , /immunology , Thorax/diagnostic imaging
2.
PLoS One ; 16(4): e0249450, 2021.
Article in English | MEDLINE | ID: covidwho-1167118

ABSTRACT

Coronavirus disease 2019 (COVID-19) has been spread out all over the world. Although a real-time reverse-transcription polymerase chain reaction (RT-PCR) test has been used as a primary diagnostic tool for COVID-19, the utility of CT based diagnostic tools have been suggested to improve the diagnostic accuracy and reliability. Herein we propose a semi-supervised deep neural network for an improved detection of COVID-19. The proposed method utilizes CT images in a supervised and unsupervised manner to improve the accuracy and robustness of COVID-19 diagnosis. Both labeled and unlabeled CT images are employed. Labeled CT images are used for supervised leaning. Unlabeled CT images are utilized for unsupervised learning in a way that the feature representations are invariant to perturbations in CT images. To systematically evaluate the proposed method, two COVID-19 CT datasets and three public CT datasets with no COVID-19 CT images are employed. In distinguishing COVID-19 from non-COVID-19 CT images, the proposed method achieves an overall accuracy of 99.83%, sensitivity of 0.9286, specificity of 0.9832, and positive predictive value (PPV) of 0.9192. The results are consistent between the COVID-19 challenge dataset and the public CT datasets. For discriminating between COVID-19 and common pneumonia CT images, the proposed method obtains 97.32% accuracy, 0.9971 sensitivity, 0.9598 specificity, and 0.9326 PPV. Moreover, the comparative experiments with respect to supervised learning and training strategies demonstrate that the proposed method is able to improve the diagnostic accuracy and robustness without exhaustive labeling. The proposed semi-supervised method, exploiting both supervised and unsupervised learning, facilitates an accurate and reliable diagnosis for COVID-19, leading to an improved patient care and management.


Subject(s)
/diagnostic imaging , Neural Networks, Computer , Supervised Machine Learning , Thorax , Tomography, X-Ray Computed , Algorithms , Datasets as Topic , Humans , Thorax/diagnostic imaging , Thorax/pathology
3.
Sci Rep ; 11(1): 6940, 2021 03 25.
Article in English | MEDLINE | ID: covidwho-1152875

ABSTRACT

A better understanding of temporal relationships between chest CT and labs may provide a reference for disease severity over the disease course. Generalized curves of lung opacity volume and density over time can be used as standardized references from well before symptoms develop to over a month after recovery, when residual lung opacities remain. 739 patients with COVID-19 underwent CT and RT-PCR in an outbreak setting between January 21st and April 12th, 2020. 29 of 739 patients had serial exams (121 CTs and 279 laboratory measurements) over 50 ± 16 days, with an average of 4.2 sequential CTs each. Sequential volumes of total lung, overall opacity and opacity subtypes (ground glass opacity [GGO] and consolidation) were extracted using deep learning and manual segmentation. Generalized temporal curves of CT and laboratory measurements were correlated. Lung opacities appeared 3.4 ± 2.2 days prior to symptom onset. Opacity peaked 1 day after symptom onset. GGO onset was earlier and resolved later than consolidation. Lactate dehydrogenase, and C-reactive protein peaked earlier than procalcitonin and leukopenia. The temporal relationships of quantitative CT features and clinical labs have distinctive patterns and peaks in relation to symptom onset, which may inform early clinical course in patients with mild COVID-19 pneumonia, or may shed light upon chronic lung effects or mechanisms of medical countermeasures in clinical trials.


Subject(s)
/diagnostic imaging , Clinical Chemistry Tests , Hematologic Tests , Thorax/diagnostic imaging , Adult , /virology , Female , Humans , Male , Middle Aged , Retrospective Studies , Severity of Illness Index , Thorax/pathology , Tomography, X-Ray Computed
4.
PLoS One ; 16(3): e0248957, 2021.
Article in English | MEDLINE | ID: covidwho-1146198

ABSTRACT

The characteristics and evolution of pulmonary fibrosis in patients with coronavirus disease 2019 (COVID-19) have not been adequately studied. AI-assisted chest high-resolution computed tomography (HRCT) was used to investigate the proportion of COVID-19 patients with pulmonary fibrosis, the relationship between the degree of fibrosis and the clinical classification of COVID-19, the characteristics of and risk factors for pulmonary fibrosis, and the evolution of pulmonary fibrosis after discharge. The incidence of pulmonary fibrosis in patients with severe or critical COVID-19 was significantly higher than that in patients with moderate COVID-19. There were significant differences in the degree of pulmonary inflammation and the extent of the affected area among patients with mild, moderate and severe pulmonary fibrosis. The IL-6 level in the acute stage and albumin level were independent risk factors for pulmonary fibrosis. Ground-glass opacities, linear opacities, interlobular septal thickening, reticulation, honeycombing, bronchiectasis and the extent of the affected area were significantly improved 30, 60 and 90 days after discharge compared with at discharge. The more severe the clinical classification of COVID-19, the more severe the residual pulmonary fibrosis was; however, in most patients, pulmonary fibrosis was improved or even resolved within 90 days after discharge.


Subject(s)
Artificial Intelligence , Pulmonary Fibrosis/diagnosis , Thorax/diagnostic imaging , /complications , Female , Humans , Image Processing, Computer-Assisted , Interleukin-6/metabolism , Male , Middle Aged , Patient Discharge , Pulmonary Fibrosis/etiology , Risk Factors , Severity of Illness Index , Tomography, X-Ray Computed
5.
Sci Rep ; 11(1): 6422, 2021 03 19.
Article in English | MEDLINE | ID: covidwho-1142463

ABSTRACT

Coronavirus disease 2019 (COVID-19) has spread in more than 100 countries and regions around the world, raising grave global concerns. COVID-19 has a similar pattern of infection, clinical symptoms, and chest imaging findings to influenza pneumonia. In this retrospective study, we analysed clinical and chest CT data of 24 patients with COVID-19 and 79 patients with influenza pneumonia. Univariate analysis demonstrated that the temperature, systolic pressure, cough and sputum production could distinguish COVID-19 from influenza pneumonia. The diagnostic sensitivity and specificity for the clinical features are 0.783 and 0.747, and the AUC value is 0.819. Univariate analysis demonstrates that nine CT features, central-peripheral distribution, superior-inferior distribution, anterior-posterior distribution, patches of GGO, GGO nodule, vascular enlargement in GGO, air bronchogram, bronchiectasis within focus, interlobular septal thickening, could distinguish COVID-19 from influenza pneumonia. The diagnostic sensitivity and specificity for the CT features are 0.750 and 0.962, and the AUC value is 0.927. Finally, a multivariate logistic regression model combined the variables from the clinical variables and CT features models was made. The combined model contained six features: systolic blood pressure, sputum production, vascular enlargement in the GGO, GGO nodule, central-peripheral distribution and bronchiectasis within focus. The diagnostic sensitivity and specificity for the combined features are 0.87 and 0.96, and the AUC value is 0.961. In conclusion, some CT features or clinical variables can differentiate COVID-19 from influenza pneumonia. Moreover, CT features combined with clinical variables had higher diagnostic performance.


Subject(s)
/diagnosis , Influenza, Human/diagnosis , Pneumonia, Viral/diagnosis , Adult , Diagnosis, Differential , Female , Humans , Influenza, Human/diagnostic imaging , Male , Middle Aged , Pneumonia, Viral/diagnostic imaging , Retrospective Studies , Thorax/diagnostic imaging , Tomography, X-Ray Computed , Young Adult
6.
Sci Rep ; 11(1): 5975, 2021 03 16.
Article in English | MEDLINE | ID: covidwho-1137818

ABSTRACT

Since the emergence of SARS-CoV-2, numerous studies have been attempting to determine biomarkers, which could rapidly and efficiently predict COVID-19 severity, however there is lack of consensus on a specific one. This retrospective cohort study is a comprehensive analysis of the initial symptoms, comorbidities and laboratory evaluation of patients, diagnosed with COVID-19 in Huoshenshan Hospital, Wuhan, from 4th February to 12th March, 2020. Based on the data collected from 63 severely ill patients from the onset of symptoms till the full recovery or demise, we found not only age (average 70) but also blood indicators as significant risk factors associated with multiple organ failure. The blood indices of all patients showed hepatic, renal, cardiac and hematopoietic dysfunction with imbalanced coagulatory biomarkers. We noticed that the levels of LDH (85%, P < .001), HBDH (76%, P < .001) and CRP (65%, P < .001) were significantly elevated in deceased patients, indicating hepatic impairment. Similarly, increased CK (15%, P = .002), Cre (37%, P = 0.102) and CysC (74%, P = 0.384) indicated renal damage. Cardiac injury was obvious from the significantly elevated level of Myoglobin (52%, P < .01), Troponin-I (65%, P = 0.273) and BNP (50%, P = .787). SARS-CoV-2 disturbs the hemolymphatic system as WBC# (73%, P = .002) and NEUT# (78%, P < .001) were significantly elevated in deceased patients. Likewise, the level of D-dimer (80%, P < .171), PT (87%, P = .031) and TT (57%, P = .053) was elevated, indicating coagulatory imbalances. We identified myoglobin and CRP as specific risk factors related to mortality and highly correlated to organ failure in COVID-19 disease.


Subject(s)
C-Reactive Protein/analysis , Myoglobin/analysis , Aged , Aged, 80 and over , Biomarkers/blood , /mortality , Comorbidity , Female , Humans , Male , Middle Aged , Multiple Organ Failure/etiology , Retrospective Studies , Risk Factors , Severity of Illness Index , Survival Analysis , Thorax/diagnostic imaging , Tomography, X-Ray Computed , Troponin I/blood
7.
PLoS One ; 16(3): e0247686, 2021.
Article in English | MEDLINE | ID: covidwho-1115303

ABSTRACT

OBJECTIVES: The aim of this study was to investigate possible patterns of demand for chest imaging during the first wave of the SARS-CoV-2 pandemic and derive a decision aid for the allocation of resources in future pandemic challenges. MATERIALS AND METHODS: Time data of requests for patients with suspected or confirmed coronavirus disease 2019 (COVID-19) lung disease were analyzed between February 27th and May 27th 2020. A multinomial logistic regression model was used to evaluate differences in the number of requests between 3 time intervals (I1: 6am - 2pm, I2: 2pm - 10pm, I3: 10pm - 6am). A cosinor model was applied to investigate the demand per hour. Requests per day were compared to the number of regional COVID-19 cases. RESULTS: 551 COVID-19 related chest imagings (32.8% outpatients, 67.2% in-patients) of 243 patients were conducted (33.3% female, 66.7% male, mean age 60 ± 17 years). Most exams for outpatients were required during I2 (I1 vs. I2: odds ratio (OR) = 0.73, 95% confidence interval (CI) 0.62-0.86, p = 0.01; I2 vs. I3: OR = 1.24, 95% CI 1.04-1.48, p = 0.03) with an acrophase at 7:29 pm. Requests for in-patients decreased from I1 to I3 (I1 vs. I2: OR = 1.24, 95% CI 1.09-1.41, p = 0.01; I2 vs. I3: OR = 1.16, 95% CI 1.05-1.28, p = 0.01) with an acrophase at 12:51 pm. The number of requests per day for outpatients developed similarly to regional cases while demand for in-patients increased later and persisted longer. CONCLUSIONS: The demand for COVID-19 related chest imaging displayed distinct distribution patterns depending on the sector of patient care and point of time during the SARS-CoV-2 pandemic. These patterns should be considered in the allocation of resources in future pandemic challenges with similar disease characteristics.


Subject(s)
/diagnostic imaging , Diagnostic Imaging/trends , Thorax/diagnostic imaging , Adult , Aged , Diagnostic Tests, Routine/trends , Female , Humans , Male , Middle Aged , Models, Theoretical , Pandemics , Pilot Projects , Thorax/virology
8.
Biomark Med ; 15(4): 285-293, 2021 03.
Article in English | MEDLINE | ID: covidwho-1105969

ABSTRACT

Background: Troponin levels may be elevated in COVID-19 infection. The aim of this study was to the explore relation between troponin levels and COVID-19 severity. Materials, methods & Results: One hundred and forty consecutive patients with COVID-19 pneumonia were included. Diagnosis of COVID-19 pneumonia was based on positive chest computed tomography (CT) findings. Quantitative PCR test was performed in all patients. Only 74 patients were quantitative PCR-positive. Twenty four patients had severe CT findings and 27 patients had progressive disease. These patients had significantly lower albumin and higher ferritin, D-dimer, lactate dehydrogenase, C-reactive protein, and high-sensitivity cardiac troponin I (hs-cTnI). Conclusion: COVID-19 patients with severe CT findings and progressive disease had higher hs-cTnI levels suggesting the use of hs-cTnI in risk stratification.


Subject(s)
Real-Time Polymerase Chain Reaction , Tomography, X-Ray Computed , Adult , Aged , Biomarkers/blood , C-Reactive Protein/metabolism , /diagnosis , Female , Ferritins/metabolism , Fibrin Fibrinogen Degradation Products/metabolism , Heart Diseases , Humans , L-Lactate Dehydrogenase/blood , Male , Middle Aged , Serum Albumin, Human/metabolism , Thorax/diagnostic imaging , Troponin I/blood
9.
PLoS One ; 16(2): e0247176, 2021.
Article in English | MEDLINE | ID: covidwho-1099926

ABSTRACT

The outbreak of coronavirus disease 2019 (COVID-19) has had an immense impact on world health and daily life in many countries. Sturdy observing of the initial site of infection in patients is crucial to gain control in the struggle with COVID-19. The early automated detection of the recent coronavirus disease (COVID-19) will help to limit its dissemination worldwide. Many initial studies have focused on the identification of the genetic material of coronavirus and have a poor detection rate for long-term surgery. The first imaging procedure that played an important role in COVID-19 treatment was the chest X-ray. Radiological imaging is often used as a method that emphasizes the performance of chest X-rays. Recent findings indicate the presence of COVID-19 in patients with irregular findings on chest X-rays. There are many reports on this topic that include machine learning strategies for the identification of COVID-19 using chest X-rays. Other current studies have used non-public datasets and complex artificial intelligence (AI) systems. In our research, we suggested a new COVID-19 identification technique based on the locality-weighted learning and self-organization map (LWL-SOM) strategy for detecting and capturing COVID-19 cases. We first grouped images from chest X-ray datasets based on their similar features in different clusters using the SOM strategy in order to discriminate between the COVID-19 and non-COVID-19 cases. Then, we built our intelligent learning model based on the LWL algorithm to diagnose and detect COVID-19 cases. The proposed SOM-LWL model improved the correlation coefficient performance results between the Covid19, no-finding, and pneumonia cases; pneumonia and no-finding cases; Covid19 and pneumonia cases; and Covid19 and no-finding cases from 0.9613 to 0.9788, 0.6113 to 1 0.8783 to 0.9999, and 0.8894 to 1, respectively. The proposed LWL-SOM had better results for discriminating COVID-19 and non-COVID-19 patients than the current machine learning-based solutions using AI evaluation measures.


Subject(s)
/diagnostic imaging , Radiographic Image Interpretation, Computer-Assisted/methods , Radiography/methods , Algorithms , Artificial Intelligence , Databases, Factual , Deep Learning , Humans , Image Processing, Computer-Assisted/methods , Lung/diagnostic imaging , Lung/pathology , Machine Learning , Neural Networks, Computer , Thorax/diagnostic imaging , Tomography, X-Ray Computed/methods
10.
J Infect Dev Ctries ; 15(1): 69-72, 2021 Jan 31.
Article in English | MEDLINE | ID: covidwho-1079737

ABSTRACT

There is rising concern that patients who recover from COVID-19 may be at risk of recurrence. Increased rates of infection and recurrence in healthcare workers could cause the healthcare system collapse and a further worsening of the COVID-19 pandemic. Herein, we reported the clinically symptomatic recurrent COVID-19 cases in the two healthcare workers who treated and recovered from symptomatic and laboratory confirmed COVID-19. We discuss important questions in the COVID-19 pandemic waiting to be answered, such as the protection period of the acquired immunity, the severity of recurrence and how long after the first infection occurs. We aimed to emphasize that healthcare workers should continue to pay maximum attention to the measures without compromising.


Subject(s)
/diagnosis , Health Personnel , Female , Humans , Male , Middle Aged , Recurrence , Thorax/diagnostic imaging , Tomography, X-Ray Computed
11.
Interdiscip Sci ; 13(1): 73-82, 2021 Mar.
Article in English | MEDLINE | ID: covidwho-1074514

ABSTRACT

Corona Virus Disease (COVID-19) has spread globally quickly, and has resulted in a large number of causalities and medical resources insufficiency in many countries. Reverse-transcriptase polymerase chain reaction (RT-PCR) testing is adopted as biopsy tool for confirmation of virus infection. However, its accuracy is as low as 60-70%, which is inefficient to uncover the infected. In comparison, the chest CT has been considered as the prior choice in diagnosis and monitoring progress of COVID-19 infection. Although the COVID-19 diagnostic systems based on artificial intelligence have been developed for assisting doctors in diagnosis, the small sample size and the excessive time consumption limit their applications. To this end, this paper proposed a diagnosis prototype system for COVID-19 infection testing. The proposed deep learning model is trained and is tested on 2267 CT sequences from 1357 patients clinically confirmed with COVID-19 and 1235 CT sequences from non-infected people. The main highlights of the prototype system are: (1) no data augmentation is needed to accurately discriminate the COVID-19 from normal controls with the specificity of 0.92 and sensitivity of 0.93; (2) the raw DICOM image is not necessary in testing. Highly compressed image like Jpeg can be used to allow a quick diagnosis; and (3) it discriminates the virus infection within 6 seconds and thus allows an online test with light cost. We also applied our model on 48 asymptomatic patients diagnosed with COVID-19. We found that: (1) the positive rate of RT-PCR assay is 63.5% (687/1082). (2) 45.8% (22/48) of the RT-PCR assay is negative for asymptomatic patients, yet the accuracy of CT scans is 95.8%. The online detection system is available: http://212.64.70.65/covid .


Subject(s)
/diagnostic imaging , Data Compression , Deep Learning , Thorax/diagnostic imaging , Tomography, X-Ray Computed , Adolescent , Adult , Aged , Aged, 80 and over , Child , Child, Preschool , Cohort Studies , Female , Humans , Infant , Infant, Newborn , Male , Middle Aged , ROC Curve , Reproducibility of Results , Young Adult
12.
PLoS One ; 16(1): e0245547, 2021.
Article in English | MEDLINE | ID: covidwho-1067419

ABSTRACT

Endemic human coronaviruses (HCoVs) and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are members of the family Coronaviridae. Comparing the findings of the infections caused by these viruses would help reveal the novel characteristics of SARS-CoV-2 and provide insight into the unique pathogenesis of SARS-CoV-2 infection. This study aimed to compare the clinical and radiological characteristics of SARS-CoV-2 and endemic HCoVs infection in adult hospitalized patients with community-acquired pneumonia (CAP). This study was performed at a university-affiliated tertiary hospital in the Republic of Korea, between January 1, 2015, and July 31, 2020. A total of 109 consecutive patients who were over 18 years of age with confirmed SARS-CoV-2 and endemic HCoVs were enrolled. Finally, 19 patients with SARS-CoV-2 CAP were compared to 40 patients with endemic HCoV CAP. Flu-like symptoms such as cough, sore throat, headache, myalgia, and prolonged fever were more common in SARS-CoV-2 CAP, whereas clinical findings suggestive of bacterial pneumonia such as dyspnea, leukocytosis with left shift, and increased C-reactive protein were more common in endemic HCoV CAP. Bilateral peripherally distributed ground-glass opacities (GGOs) were typical radiologic findings in SARS-CoV-2 CAP, whereas mixed patterns of GGOs, consolidations, micronodules, and pleural effusion were observed in endemic HCoV CAP. Coinfection was not observed in patients with SARS-CoV-2 CAP, but was observed in more than half of the patients with endemic HCoV CAP. There were distinctive differences in the clinical and radiologic findings between SARS-CoV-2 and endemic HCoV CAP. Further investigations are required to elucidate the mechanism underlying this difference. Follow-up observations are needed to determine if the presentation of SARS-CoV-2 CAP changes with repeated infection.


Subject(s)
/diagnostic imaging , Pneumonia, Viral/diagnostic imaging , Aged , /pathology , Cohort Studies , Coinfection/diagnostic imaging , Coinfection/epidemiology , Coinfection/pathology , Coinfection/virology , Community-Acquired Infections , Coronavirus/isolation & purification , Endemic Diseases , Female , Humans , Lung/diagnostic imaging , Male , Middle Aged , Middle East Respiratory Syndrome Coronavirus/isolation & purification , Pandemics , Pneumonia, Viral/epidemiology , Pneumonia, Viral/pathology , Pneumonia, Viral/virology , Radiography, Thoracic/methods , Republic of Korea/epidemiology , Retrospective Studies , Risk Factors , Thorax/diagnostic imaging
13.
PLoS One ; 16(1): e0245518, 2021.
Article in English | MEDLINE | ID: covidwho-1067418

ABSTRACT

OBJECTIVES: High-risk CXR features in COVID-19 are not clearly defined. We aimed to identify CXR features that correlate with severe COVID-19. METHODS: All confirmed COVID-19 patients admitted within the study period were screened. Those with suboptimal baseline CXR were excluded. CXRs were reviewed by three independent radiologists and opacities recorded according to zones and laterality. The primary endpoint was defined as hypoxia requiring supplemental oxygen, and CXR features were assessed for association with this endpoint to identify high-risk features. These features were then used to define criteria for a high-risk CXR, and clinical features and outcomes of patients with and without baseline high-risk CXR were compared using logistic regression analysis. RESULTS: 109 patients were included. In the initial analysis of 40 patients (36.7%) with abnormal baseline CXR, presence of bilateral opacities, multifocal opacities, or any upper or middle zone opacity were associated with supplemental oxygen requirement. Of the entire cohort, 29 patients (26.6%) had a baseline CXR with at least one of these features. Having a high-risk baseline CXR was significantly associated with requiring supplemental oxygen in univariate (odds ratio 14.0, 95% confidence interval 3.90-55.60) and multivariate (adjusted odds ratio 8.38, 95% CI 2.43-28.97, P = 0.001) analyses. CONCLUSION: We identified several high-risk CXR features that are significantly associated with severe illness. The association of upper or middle zone opacities with severe illness has not been previously emphasized. Recognition of these specific high-risk CXR features is important to prioritize limited healthcare resources for sicker patients.


Subject(s)
/diagnostic imaging , Adult , /virology , Cohort Studies , Emergency Service, Hospital , Female , Hospitalization , Humans , Lung/diagnostic imaging , Male , Middle Aged , Radiography, Thoracic/methods , Retrospective Studies , Risk Factors , Severity of Illness Index , Thorax/diagnostic imaging , Tomography, X-Ray Computed/methods
14.
Comput Intell Neurosci ; 2021: 8890226, 2021.
Article in English | MEDLINE | ID: covidwho-1066962

ABSTRACT

The novel coronavirus, SARS-CoV-2, can be deadly to people, causing COVID-19. The ease of its propagation, coupled with its high capacity for illness and death in infected individuals, makes it a hazard to the community. Chest X-rays are one of the most common but most difficult to interpret radiographic examination for early diagnosis of coronavirus-related infections. They carry a considerable amount of anatomical and physiological information, but it is sometimes difficult even for the expert radiologist to derive the related information they contain. Automatic classification using deep learning models can help in better assessing these infections swiftly. Deep CNN models, namely, MobileNet, ResNet50, and InceptionV3, were applied with different variations, including training the model from the start, fine-tuning along with adjusting learned weights of all layers, and fine-tuning with learned weights along with augmentation. Fine-tuning with augmentation produced the best results in pretrained models. Out of these, two best-performing models (MobileNet and InceptionV3) selected for ensemble learning produced accuracy and FScore of 95.18% and 90.34%, and 95.75% and 91.47%, respectively. The proposed hybrid ensemble model generated with the merger of these deep models produced a classification accuracy and FScore of 96.49% and 92.97%. For test dataset, which was separately kept, the model generated accuracy and FScore of 94.19% and 88.64%. Automatic classification using deep ensemble learning can help radiologists in the correct identification of coronavirus-related infections in chest X-rays. Consequently, this swift and computer-aided diagnosis can help in saving precious human lives and minimizing the social and economic impact on society.


Subject(s)
/classification , Image Processing, Computer-Assisted/methods , Thorax/diagnostic imaging , Algorithms , Computer Simulation , Deep Learning , Diagnosis, Computer-Assisted , Humans , Machine Learning , Neural Networks, Computer , Reproducibility of Results , Software
15.
J Coll Physicians Surg Pak ; 31(1): 14-20, 2021 Jan.
Article in English | MEDLINE | ID: covidwho-1068267

ABSTRACT

OBJECTIVE:   To identify utility of chest computed tomography severity score (CT-SS) as an additional tool to COVID-19 pneumonia imaging classification in assessing severity of COVID-19. STUDY DESIGN: Descriptive analytical study Place and Duration of Study: Armed Forces Institute of Radiology and Imaging, (AFIRI) Rawalpindi, from April 2020 to June 2020. METHODOLOGY: Five hundred suspected COVID-19 cases referred for high resolution computed tomography - chest were included in the study. Cases were categorised by radiological findings using COVID-19 pneumonia imaging classification, proposed in the radiological society of North America expert consensus statement on reporting chest CT findings related to COVID-19. CT-SS was calculated for all scans. Patients were clinically classified according to disease severity as per 'Diagnosis And Treatment Program of Pneumonia of New Coronavirus Infection' recommended by China's National Health Commission. The relationships between radiological findings, CT-SS, and clinical severity were explored. RESULTS: Based on the radiological findings, 298 cases were graded as typical, 34 as indeterminate, 15 as atypical, and 153 as negative for pneumonia. The apical and posterior basal segments of lower lobes were most commonly involved. The CT-SS showed higher values in patients of severe group as compared to those in moderate group (p < 0.05). CT-SS threshold for recognising severe COVID-19 was 18.5 (area under curve, 0.960), with 84.3% sensitivity and 92.5% specificity. CONCLUSION: In coherence with COVID-19 pneumonia imaging classification, CT-SS may provide a comprehensive and objective assessment of COVID-19 severity. Key Words: COVID-19, COVID-19 pneumonia, CT-SS, High resolution computed tomography.


Subject(s)
Lung/diagnostic imaging , Pneumonia, Viral/diagnostic imaging , Thorax/diagnostic imaging , Tomography, X-Ray Computed/methods , Adolescent , Adult , Aged , Aged, 80 and over , Child , Child, Preschool , Diagnostic Tests, Routine , Female , Humans , Infant , Infant, Newborn , Male , Middle Aged , Pakistan , Radiography, Thoracic/methods , Tertiary Healthcare , Young Adult
16.
BMC Infect Dis ; 21(1): 141, 2021 Feb 03.
Article in English | MEDLINE | ID: covidwho-1063188

ABSTRACT

BACKGROUND: The impact of COVID-19 has been devastating on a global scale. The negative conversion time (NCT) of SARS-CoV-2 RNA is closely related to clinical manifestation and disease progression in COVID-19 patients. Our study aimed to predict factors associated with prolonged NCT of SARS-CoV-2 RNA in mild/moderate COVID-19 patients. METHODS: The clinical features, laboratory data and treatment outcomes of COVID-19 patients were retrospectively analyzed. Then univariate and multivariate analysis were used to screen out risk factors of influencing prolonged NCT of SARS-CoV-2 RNA. RESULTS: Thirty-two hospitalized mild/moderate COVID-19 patients were enrolled. The general clinical symptoms were cough (78.1%), fever (75%), diarrhea (68.8%), expectoration (56.3%), and nausea (37.5%). More than 40% of the patients had decreased erythrocyte, hemoglobin and leucocyte and 93.8% patients were detected in abnormalities of chest CT. The median NCT of SARS-CoV-2 RNA was 19.5 days (IQR: 14.25-25). Univariate analysis found fever, nausea, diarrhea and abnormalities in chest CTs were positively associated with prolonged NCT of viral RNA (P< 0.05). The multivariate Cox proportional hazard model revealed that fever [Exp (B), 0.284; 95% CI, 0.114-0.707; P<0.05] and nausea [Exp (B), 0.257; 95%CI, 0.096-0.689; P<0.05] were two significant independent factors. CONCLUSIONS: Fever and nausea were two significant independent factors in prolonged NCT of viral RNA in mild/moderate COVID-19 patients, which provided a useful references for disease progression and treatment of COVID-19.


Subject(s)
/diagnosis , RNA, Viral/metabolism , /genetics , Adult , /pathology , Cough/etiology , Diarrhea/etiology , Female , Fever/etiology , Hospitalization , Humans , Male , Middle Aged , Multivariate Analysis , Proportional Hazards Models , Retrospective Studies , Risk Factors , Severity of Illness Index , Thorax/diagnostic imaging , Time Factors
17.
Crit Care ; 25(1): 4, 2021 01 04.
Article in English | MEDLINE | ID: covidwho-1059728

ABSTRACT

BACKGROUND: Patients with COVID-19 (COVID) may develop acute respiratory distress syndrome with or without sepsis, coagulopathy and visceral damage. While chest CT scans are routinely performed in the initial assessment of patients with severe pulmonary forms, thymus involvement and reactivation have not been investigated so far. METHODS: In this observational study, we systematically scored the enlargement of the thymus and the lung involvement, using CT scans, in all adult patients admitted to the ICU for COVID or any other cause (control group) at one centre between March and April 2020. Initial biological investigations included nasal detection of SARS-CoV-2 ribonucleic acid by polymerase chain reaction (PCR). In a subgroup of 24 patients with different degrees of pulmonary involvement and thymus hypertrophy, plasma cytokine concentrations were measured and the export of mature T cells from the thymus was estimated simultaneously by PCR quantification of T cell receptor excision circles (TRECs). RESULTS: Eighty-seven patients were studied: 50 COVID patients and 37 controls. Non-atrophic or enlarged thymus was more commonly observed in COVID patients than in controls (66% vs. 24%, p < 0.0001). Thymus enlargement in COVID patients was associated with more extensive lung injury score on CT scans (4 [3-5] vs. 2 [1.5-4], p = 0.01), but a lower mortality rate (8.6% vs. 41.2%, p < 0.001). Other factors associated with mortality were age, lymphopaenia, high CRP and co-morbidities. COVID patients had higher concentrations of IL-7 (6.00 [3.72-9.25] vs. 2.17 [1.76-4.4] pg/mL; p = 0.04) and higher thymic production of new lymphocytes (sj/ßTREC ratio = 2.88 [1.98-4.51] vs. 0.23 [0.15-0.60]; p = 0.004). Thymic production was also correlated with the CT scan thymic score (r = 0.38, p = 0.03) and inversely correlated with the number of lymphocytes (r = 0.56, p = 0.007). CONCLUSION: In COVID patients, thymus enlargement was frequent and associated with increased T lymphocyte production, which appears to be a beneficial adaptation to virus-induced lymphopaenia. The lack of thymic activity/reactivation in older SARS-CoV-2 infected patients could contribute to a worse prognosis.


Subject(s)
/complications , Thymus Hyperplasia/diagnostic imaging , Aged , Case-Control Studies , Female , Hospitalization , Humans , Intensive Care Units , Male , Middle Aged , Thorax/diagnostic imaging , Thymus Hyperplasia/virology , Tomography, X-Ray Computed
18.
Int J Comput Assist Radiol Surg ; 16(3): 435-445, 2021 Mar.
Article in English | MEDLINE | ID: covidwho-1041909

ABSTRACT

PURPOSE: Severity scoring is a key step in managing patients with COVID-19 pneumonia. However, manual quantitative analysis by radiologists is a time-consuming task, while qualitative evaluation may be fast but highly subjective. This study aims to develop artificial intelligence (AI)-based methods to quantify disease severity and predict COVID-19 patient outcome. METHODS: We develop an AI-based framework that employs deep neural networks to efficiently segment lung lobes and pulmonary opacities. The volume ratio of pulmonary opacities inside each lung lobe gives the severity scores of the lobes, which are then used to predict ICU admission and mortality with three different machine learning methods. The developed methods were evaluated on datasets from two hospitals (site A: Firoozgar Hospital, Iran, 105 patients; site B: Massachusetts General Hospital, USA, 88 patients). RESULTS: AI-based severity scores are strongly associated with those evaluated by radiologists (Spearman's rank correlation 0.837, [Formula: see text]). Using AI-based scores produced significantly higher ([Formula: see text]) area under the ROC curve (AUC) values. The developed AI method achieved the best performance of AUC = 0.813 (95% CI [0.729, 0.886]) in predicting ICU admission and AUC = 0.741 (95% CI [0.640, 0.837]) in mortality estimation on the two datasets. CONCLUSIONS: Accurate severity scores can be obtained using the developed AI methods over chest CT images. The computed severity scores achieved better performance than radiologists in predicting COVID-19 patient outcome by consistently quantifying image features. Such developed techniques of severity assessment may be extended to other lung diseases beyond the current pandemic.


Subject(s)
Artificial Intelligence , Thorax/diagnostic imaging , Adult , Aged , Aged, 80 and over , Databases, Factual , Female , Hospitalization , Humans , Lung/diagnostic imaging , Male , Middle Aged , Neural Networks, Computer , Pandemics , Prognosis , Retrospective Studies , Severity of Illness Index , Tomography, X-Ray Computed/methods , Treatment Outcome
19.
Sensors (Basel) ; 21(2)2021 Jan 11.
Article in English | MEDLINE | ID: covidwho-1022007

ABSTRACT

This paper explores how well deep learning models trained on chest CT images can diagnose COVID-19 infected people in a fast and automated process. To this end, we adopted advanced deep network architectures and proposed a transfer learning strategy using custom-sized input tailored for each deep architecture to achieve the best performance. We conducted extensive sets of experiments on two CT image datasets, namely, the SARS-CoV-2 CT-scan and the COVID19-CT. The results show superior performances for our models compared with previous studies. Our best models achieved average accuracy, precision, sensitivity, specificity, and F1-score values of 99.4%, 99.6%, 99.8%, 99.6%, and 99.4% on the SARS-CoV-2 dataset, and 92.9%, 91.3%, 93.7%, 92.2%, and 92.5% on the COVID19-CT dataset, respectively. For better interpretability of the results, we applied visualization techniques to provide visual explanations for the models' predictions. Feature visualizations of the learned features show well-separated clusters representing CT images of COVID-19 and non-COVID-19 cases. Moreover, the visualizations indicate that our models are not only capable of identifying COVID-19 cases but also provide accurate localization of the COVID-19-associated regions, as indicated by well-trained radiologists.


Subject(s)
/diagnosis , Deep Learning , Thorax/diagnostic imaging , Tomography, X-Ray Computed , Algorithms , /virology , Databases, Factual , Humans , Neural Networks, Computer , Radiographic Image Interpretation, Computer-Assisted , Thorax/pathology , Thorax/virology
20.
Ann Med ; 53(1): 169-180, 2021 12.
Article in English | MEDLINE | ID: covidwho-1020049

ABSTRACT

OBJECTIVES: Coronavirus disease 2019 (COVID-19) has rapidly swept across the world. This study aimed to explore the relationship between the chest CT findings and clinical characteristics of COVID-19 patients. METHODS: Patients with COVID-19 confirmed by next-generation sequencing or RT-PCR who had undergone more than 4 serial chest CT procedures were retrospectively enrolled. RESULTS: This study included 361 patients - 192 men and 169 women. On initial chest CT, more lesions were identified as multiple bilateral lungs lesions and localised in the peripheral lung. The predominant patterns of abnormality were ground-glass opacities (GGO) (28.5%), consolidation (13.0%), nodule (23.0%), fibrous stripes (5.3%) and mixed (30.2%). Severe cases were more common in patients with a mixed pattern (21.1%) and less common in patients with nodules (2.4%). During follow-up CT, the mediumtotal severity score (TSS) in patients with nodules and fibrous strips was significantly lower than that in patients with mixed patterns in all three stages (p < .01). CONCLUSION: Chest CT plays an important role in diagnosing COVID-19. The CT features may vary by age. Different CT features are not only associated with clinical manifestation but also patient prognosis. Key messages The initial chest CT findings of COVID-19 could help us monitor and predict the outcome. Nodules were more common in non severe cases and had a favorable prognosis. The mixed pattern was more common in severe cases and usually had a relatively poor outcome.


Subject(s)
/diagnostic imaging , Thorax/diagnostic imaging , Tomography, X-Ray Computed , Adolescent , Adult , /epidemiology , Female , Humans , Male , Middle Aged , Retrospective Studies , Young Adult
SELECTION OF CITATIONS
SEARCH DETAIL