Two-step machine learning to diagnose and predict involvement of lungs in COVID-19 and pneumonia using CT radiomics.

OBJECTIVE: To develop a two-step machine learning (ML) based model to diagnose and predict involvement of lungs in COVID-19 and non COVID-19 pneumonia patients using CT chest radiomic features. METHODS: Three hundred CT scans (3-classes: 100 COVID-19, 100 pneumonia, and 100 healthy subjects) were enrolled in this study. Diagnostic task included 3-class classification. Severity prediction score for COVID-19 and pneumonia was considered as mild (0-25%), moderate (26-50%), and severe (>50%). Whole lungs were segmented utilizing deep learning-based segmentation. Altogether, 107 features including shape, first-order histogram, second and high order texture features were extracted. Pearson correlation coefficient (PCC≥90%) followed by different features selection algorithms were employed. ML-based supervised algorithms (Naïve Bays, Support Vector Machine, Bagging, Random Forest, K-nearest neighbors, Decision Tree and Ensemble Meta voting) were utilized. The optimal model was selected based on precision, recall and area-under-curve (AUC) by randomizing the training/validation, followed by testing using the test set. RESULTS: Nine pertinent features (2 shape, 1 first-order, and 6 second-order) were obtained after features selection for both phases. In diagnostic task, the performance of 3-class classification using Random Forest was 0.909±0.026, 0.907±0.056, 0.902±0.044, 0.939±0.031, and 0.982±0.010 for precision, recall, F1-score, accuracy, and AUC, respectively. The severity prediction task using Random Forest achieved 0.868±0.123 precision, 0.865±0.121 recall, 0.853±0.139 F1-score, 0.934±0.024 accuracy, and 0.969±0.022 AUC. CONCLUSION: The two-phase ML-based model accurately classified COVID-19 and pneumonia patients using CT radiomics, and adequately predicted severity of lungs involvement. This 2-steps model showed great potential in assessing COVID-19 CT images towards improved management of patients.

COVID-2019 Pneumonia: Severity and distribution of lung changes observed on the initial chest X-ray as an indicator of final outcomes.

Objectives: This study aimed to assess the correlation between the severity of the initial chest x-ray (CXR) abnormalities in patients with a confirmed diagnosis of COVID-19 and the final outcomes. Methods: This retrospective study was conducted at the Royal Hospital, Oman between mid-March and May 2020 and included patients who had been admitted with a confirmed diagnosis of COVID-19 and had a final outcome. Serial CXRs were identified and examined for presence, extent, distribution and progression pattern of radiological abnormalities. Each lung field was divided into three zones on each CXR and a score was allocated for each zone (0 is normal and 1-4 is mild-severe). The scores for all six zones per CXR examination were summed to provide a cumulative chest radiographic score (range: 0-24). Results: A total of 64 patients were included; the majority were male (89.1%) and the mean age was 50.22 ± 14.86 years. The initial CXR was abnormal in 60 patients (93.8%). The most common finding was ground glass opacity (n = 58, 96.7%) followed by consolidation (n = 50, 83.3%). Most patients had bilateral (n = 51, 85.0%), multifocal (n = 57, 95.0%) and mixed central and peripheral (n = 36, 60.0%) lung abnormalities. The median score of initial CXR for deceased patients was significantly higher than recovered patients (17 versus 11; P = 0.009). Five CXR evolution patterns were identified: type I (initial radiograph deteriorates then improves), type II (fluctuate), type III (static), type IV (progressive deterioration) and type V (progressive improvement). Conclusion: A higher baseline CXR score is associated with higher mortality rate and poor prognosis in those with COVID-19 pneumonia.

Chest Radiography Findings of COVID-19 Patients: Differences between the Intubated and Non-intubated Groups.

OBJECTIVES: We sought to identify the chest radiography differences at presentation between two groups of hospitalized confirmed COVID-19 patients; intubated group compared to non-intubated group. METHODS: We retrospectively collected the data of confirmed hospitalized COVID-19 patients at the Royal Hospital, Muscat, Oman, from March to April 2020. Radiographic and clinical data were collected from the hospital and radiology information systems and compared between two groups based on intubation status. RESULTS: Twenty-six patients confirmed to have COVID-19 by reverse-transcriptase polymerase chain reaction test were included in the study; 15/26 were non-intubated, and 11/26 were intubated. Overall, 88.5% were males in the intubated group. Respiratory symptoms were the most common presentation (84.6%) followed by fever (76.9%), with no statistical difference between the two groups. There was a statistically significant difference in having diabetes mellitus (p = 0.020) in which 8/11 and 4/15 were recorded to have diabetes mellitus in the intubated and non-intubated groups, respectively. Other comorbidities showed no statistically significant difference. The radiographic analysis redemonstrates the peripheral lower zone distribution but no statistically significant difference among the two groups. There were no differences between the intubated and non-intubated chest radiography in laterality involvement, central and peripheral distribution, and lesions type. However, upper zones involvement was more noted in the intubated group with 10/11 (90.9%) compared to 7/15 (46.7%) in non-intubated cases (p = 0.036). There were higher numbers of zone involvement in intubated cases than non-intubated cases: 9/11 (81.8%) of intubated patients had 10-12 areas of involvement on chest radiographs compared to 3/15 (20.0%) in the non-intubated group. Half of the cases were discharged home; 3/11 from the intubated group and 10/15 from the non-intubated group. Five patients died from the intubated group (5/11) versus 3/15 from the non-intubated group. Five patients are still hospitalized (three from the intubated group and two from the non-intubated group). CONCLUSIONS: The radiographic findings among intubated and non-intubated hospitalized COVID-19 patients demonstrate differences in the number of zones involved. More upper zone involvement was noted in the intubated group. Male sex and diabetes mellitus carried a poorer prognosis and were more associated with the intubated group.

Artificial intelligence-driven assessment of radiological images for COVID-19.

Artificial Intelligence (AI) methods have significant potential for diagnosis and prognosis of COVID-19 infections. Rapid identification of COVID-19 and its severity in individual patients is expected to enable better control of the disease individually and at-large. There has been remarkable interest by the scientific community in using imaging biomarkers to improve detection and management of COVID-19. Exploratory tools such as AI-based models may help explain the complex biological mechanisms and provide better understanding of the underlying pathophysiological processes. The present review focuses on AI-based COVID-19 studies as applies to chest x-ray (CXR) and computed tomography (CT) imaging modalities, and the associated challenges. Explicit radiomics, deep learning methods, and hybrid methods that combine both deep learning and explicit radiomics have the potential to enhance the ability and usefulness of radiological images to assist clinicians in the current COVID-19 pandemic. The aims of this review are: first, to outline COVID-19 AI-analysis workflows, including acquisition of data, feature selection, segmentation methods, feature extraction, and multi-variate model development and validation as appropriate for AI-based COVID-19 studies. Secondly, existing limitations of AI-based COVID-19 analyses are discussed, highlighting potential improvements that can be made. Finally, the impact of AI and radiomics methods and the associated clinical outcomes are summarized. In this review, pipelines that include the key steps for AI-based COVID-19 signatures identification are elaborated. Sample size, non-standard imaging protocols, segmentation, availability of public COVID-19 databases, combination of imaging and clinical information and full clinical validation remain major limitations and challenges. We conclude that AI-based assessment of CXR and CT images has significant potential as a viable pathway for the diagnosis, follow-up and prognosis of COVID-19.

COVID-19 Associated Pneumonia: A review of chest radiograph and computed tomography findings.

Medical imaging, including chest radiography and computed tomography, plays a major role in the diagnosis and follow-up of patients with COVID-19 associated pneumonia. This review aims to summarise current information on this topic based on the existing literature. A search of the Google Scholar (Google LLC, Mountain View, California, USA) and MEDLINE® (National Library of Medicine, Bethesda, Maryland, USA) databases was conducted for articles published until April 2020. A total of 30 articles involving 4,002 patients were identified. The most frequently reported imaging findings were bilateral ground glass and consolidative pulmonary opacities with a predominant lower lobe and peripheral subpleural distribution.