POLCOVID: a multicenter multiclass chest X-ray database (Poland, 2020-2021).

The outbreak of the SARS-CoV-2 pandemic has put healthcare systems worldwide to their limits, resulting in increased waiting time for diagnosis and required medical assistance. With chest radiographs (CXR) being one of the most common COVID-19 diagnosis methods, many artificial intelligence tools for image-based COVID-19 detection have been developed, often trained on a small number of images from COVID-19-positive patients. Thus, the need for high-quality and well-annotated CXR image databases increased. This paper introduces POLCOVID dataset, containing chest X-ray (CXR) images of patients with COVID-19 or other-type pneumonia, and healthy individuals gathered from 15 Polish hospitals. The original radiographs are accompanied by the preprocessed images limited to the lung area and the corresponding lung masks obtained with the segmentation model. Moreover, the manually created lung masks are provided for a part of POLCOVID dataset and the other four publicly available CXR image collections. POLCOVID dataset can help in pneumonia or COVID-19 diagnosis, while the set of matched images and lung masks may serve for the development of lung segmentation solutions.

Enhancing thoracic disease detection using chest X-rays from PubMed Central Open Access.

Large chest X-rays (CXR) datasets have been collected to train deep learning models to detect thorax pathology on CXR. However, most CXR datasets are from single-center studies and the collected pathologies are often imbalanced. The aim of this study was to automatically construct a public, weakly-labeled CXR database from articles in PubMed Central Open Access (PMC-OA) and to assess model performance on CXR pathology classification by using this database as additional training data. Our framework includes text extraction, CXR pathology verification, subfigure separation, and image modality classification. We have extensively validated the utility of the automatically generated image database on thoracic disease detection tasks, including Hernia, Lung Lesion, Pneumonia, and pneumothorax. We pick these diseases due to their historically poor performance in existing datasets: the NIH-CXR dataset (112,120 CXR) and the MIMIC-CXR dataset (243,324 CXR). We find that classifiers fine-tuned with additional PMC-CXR extracted by the proposed framework consistently and significantly achieved better performance than those without (e.g., Hernia: 0.9335 vs 0.9154; Lung Lesion: 0.7394 vs. 0.7207; Pneumonia: 0.7074 vs. 0.6709; Pneumothorax 0.8185 vs. 0.7517, all in AUC with p< 0.0001) for CXR pathology detection. In contrast to previous approaches that manually submit the medical images to the repository, our framework can automatically collect figures and their accompanied figure legends. Compared to previous studies, the proposed framework improved subfigure segmentation and incorporates our advanced self-developed NLP technique for CXR pathology verification. We hope it complements existing resources and improves our ability to make biomedical image data findable, accessible, interoperable, and reusable.

Adaptive Feature Selection Guided Deep Forest for COVID-19 Classification With Chest CT.

Chest computed tomography (CT) becomes an effective tool to assist the diagnosis of coronavirus disease-19 (COVID-19). Due to the outbreak of COVID-19 worldwide, using the computed-aided diagnosis technique for COVID-19 classification based on CT images could largely alleviate the burden of clinicians. In this paper, we propose an Adaptive Feature Selection guided Deep Forest (AFS-DF) for COVID-19 classification based on chest CT images. Specifically, we first extract location-specific features from CT images. Then, in order to capture the high-level representation of these features with the relatively small-scale data, we leverage a deep forest model to learn high-level representation of the features. Moreover, we propose a feature selection method based on the trained deep forest model to reduce the redundancy of features, where the feature selection could be adaptively incorporated with the COVID-19 classification model. We evaluated our proposed AFS-DF on COVID-19 dataset with 1495 patients of COVID-19 and 1027 patients of community acquired pneumonia (CAP). The accuracy (ACC), sensitivity (SEN), specificity (SPE), AUC, precision and F1-score achieved by our method are 91.79%, 93.05%, 89.95%, 96.35%, 93.10% and 93.07%, respectively. Experimental results on the COVID-19 dataset suggest that the proposed AFS-DF achieves superior performance in COVID-19 vs. CAP classification, compared with 4 widely used machine learning methods.

COVID-19 in solid organ transplant recipients: A single-center case series from Spain.

The clinical characteristics, management, and outcome of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) after solid organ transplant (SOT) remain unknown. We report our preliminary experience with 18 SOT (kidney [44.4%], liver [33.3%], and heart [22.2%]) recipients diagnosed with COVID-19 by March 23, 2020 at a tertiary-care center at Madrid. Median age at diagnosis was 71.0 ± 12.8 years, and the median interval since transplantation was 9.3 years. Fever (83.3%) and radiographic abnormalities in form of unilateral or bilateral/multifocal consolidations (72.2%) were the most common presentations. Lopinavir/ritonavir (usually associated with hydroxychloroquine) was used in 50.0% of patients and had to be prematurely discontinued in 2 of them. Other antiviral regimens included hydroxychloroquine monotherapy (27.8%) and interferon-ß (16.7%). As of April 4, the case-fatality rate was 27.8% (5/18). After a median follow-up of 18 days from symptom onset, 30.8% (4/13) of survivors developed progressive respiratory failure, 7.7% (1/13) showed stable clinical condition or improvement, and 61.5% (8/13) had been discharged home. C-reactive protein levels at various points were significantly higher among recipients who experienced unfavorable outcome. In conclusion, this frontline report suggests that SARS-CoV-2 infection has a severe course in SOT recipients.

Did radiation exposure increase with chest computed tomography use among different ages during the COVID-19 pandemic? A multi-center study with 42028 chest computed tomography scans

PURPOSE: To determine whether radiation exposure increased among different ages with chest computed tomography (CT) use during the coronavirus disease-2019 (COVID-19) pandemic. METHODS: Patients with chest CT scans in an 8-month period of the pandemic between March 15, 2020, and November 15, 2020, and the same period of the preceding year were included in the study. Indications of chest CT scans were obtained from the clinical notes and categorized as infectious diseases, neoplastic disorders, trauma, and other diseases. Chest CT scans for infectious diseases during the pandemic were compared with those with the same indications in 2019. The dose-length product values were obtained from the protocol screen individually. RESULTS: The total number of chest CT scans with an indication of infectious disease was 21746 in 2020 and 4318 in 2019. Total radiation exposure increased by 573% with the use of chest CT for infectious indications but decreased by 19% for neoplasia, 12% for trauma, and 43% for other reasons. The mean age of the patients scanned in 2019 was significantly higher than those scanned during the pandemic (64.6 vs. 50.3 years). A striking increase was seen in the 10-59 age group during the pandemic (P < 0.001). The highest increase was seen in the 20-29 age group, being 18.6 fold. One death was recorded per 58 chest CT scans during the pandemic. Chest CT use was substantially higher at the beginning of the pandemic. CONCLUSION: Chest CT was excessively used during the COVID-19 pandemic. Young and middle-aged people were exposed more than others. The impact of COVID-19-pandemic-related radiation exposure on public health should be followed carefully in future years.

Chest radiograph findings in children with COVID-19-A retrospective analysis from a tertiary care paediatric hospital in South India.

OBJECTIVE: The primary aim of this study is to document the chest X-ray findings in children with COVID-19 pneumonia. The secondary aim is to correlate chest X-ray findings to patient outcome. METHODS: We performed a retrospective analysis of children (0-18 years) with SARS-CoV-2 admitted to our hospital from June 2020 to December 2021. The chest radiographs were assessed for: peribronchial cuffing, ground-glass opacities (GGOs), consolidation, pulmonary nodules and pleural effusion. The severity of the pulmonary findings was graded using a modification of the Brixia score. RESULTS: There were a total of 90 patients with SARS-CoV-2 infection; the mean age was 5.8 years (age range 7 days to 17 years). Abnormalities were seen on the CXR in 74 (82%) of the 90 patients. Bilateral peribronchial cuffing was seen in 68% (61/90), consolidation in 11% (10/90), bilateral central GGOs in 2% (2/90) and unilateral pleural effusion in 1% (1/90). Overall the average CXR score in our cohort of patients was 6. The average CXR score in patients with oxygen requirement was 10. The duration of hospital stay was significantly longer in those patients with CXR score >9. CONCLUSION: The CXR score has the potential to serve as tool to identify children at high risk and may aid planning of clinical management in such patients.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) created a global pandemic in early March 2020. There are very few studies describing the lung changes in affected children. We performed a retrospective study in children, aged between 0 days and 18 years, who tested positive for this virus. This study was conducted in a paediatric tertiary care hospital in South India. Chest X-ray (CXR) was done in children with moderate and severe SARS-CoV-2 infection; these X-rays were reviewed and scoring was done to assess the degree of abnormality. It was seen that the duration of hospital stay was longer in children with a high CXR score. Amongst the children with score >9, 60% needed oxygen support during their treatment. Thus, CXR score can play a role in the prediction of disease outcome in SARS-CoV-2 infection.

Abdominal Imaging Findings on Computed Tomography as a Tool for COVID-19 Mortality Risk Assessment: Comparison With Chest Radiograph Severity Scores.

OBJECTIVE: To quantify the association between computed tomography abdomen and pelvis with contrast (CTAP) findings and chest radiograph (CXR) severity score, and the incremental effect of incorporating CTAP findings into predictive models of COVID-19 mortality. METHODS: This retrospective study was performed at a large quaternary care medical center. All adult patients who presented to our institution between March and June 2020 with the diagnosis of COVID-19 and had a CXR up to 48 hours before a CTAP were included. Primary outcomes were the severity of lung disease before CTAP and mortality within 14 and 30 days. Logistic regression models were constructed to quantify the association between CXR score and CTAP findings. Penalized logistic regression models and random forests were constructed to identify key predictors (demographics, CTAP findings, and CXR score) of mortality. The discriminatory performance of these models, with and without CTAP findings, was summarized using area under the characteristic (AUC) curves. RESULTS: One hundred ninety-five patients (median age, 63 years; 119 men) were included. The odds of having CTAP findings was 3.89 times greater when a CXR score was classified as severe compared with mild (P = 0.002). When CTAP findings were included in the feature set, the AUCs for 14-day mortality were 0.67 (penalized logistic regression) and 0.71 (random forests). Similar values for 30-day mortality were 0.76 and 0.75. When CTAP findings were omitted, all AUC values were attenuated. CONCLUSIONS: The CTAP findings were associated with more severe CXR score and may serve as predictors of COVID-19 mortality.

Diagnostic yield, safety, and advantages of ultra-low dose chest CT compared to chest radiography in early stage suspected SARS-CoV-2 pneumonia: A retrospective observational study.

ABSTRACT: To determine the role of ultra-low dose chest computed tomography (uld CT) compared to chest radiographs in patients with laboratory-confirmed early stage SARS-CoV-2 pneumonia.Chest radiographs and uld CT of 12 consecutive suspected SARS-CoV-2 patients performed up to 48 hours from hospital admission were reviewed by 2 radiologists. Dosimetry and descriptive statistics of both modalities were analyzed.On uld CT, parenchymal abnormalities compatible with SARS-CoV-2 pneumonia were detected in 10/12 (83%) patients whereas on chest X-ray in, respectively, 8/12 (66%) and 5/12 (41%) patients for reader 1 and 2. The average increment of diagnostic performance of uld CT compared to chest X-ray was 29%. The average effective dose was, respectively, of 0.219 and 0.073 mSv.Uld CT detects substantially more lung injuries in symptomatic patients with suspected early stage SARS-CoV-2 pneumonia compared to chest radiographs, with a significantly better inter-reader agreement, at the cost of a slightly higher equivalent radiation dose.

Chest x-ray imaging score is associated with severity of COVID-19 pneumonia: the MBrixia score.

Spatial resolution in existing chest x-ray (CXR)-based scoring systems for coronavirus disease 2019 (COVID-19) pneumonia is low, and should be increased for better representation of anatomy, and severity of lung involvement. An existing CXR-based system, the Brixia score, was modified to increase the spatial resolution, creating the MBrixia score. The MBrixia score is the sum, of a rule-based quantification of CXR severity on a scale of 0 to 3 in 12 anatomical zones in the lungs. The MBrixia score was applied to CXR images from COVID-19 patients at a single tertiary hospital in the period May 4th-June 5th, 2020. The relationship between MBrixia score, and level of respiratory support at the time of performed CXR imaging was investigated. 37 hospitalized COVID-19 patients with 290 CXRs were identified, 22 (59.5%) were admitted to the intensive care unit and 10 (27%) died during follow-up. In a Poisson regression using all 290 MBrixia scored CXRs, a higher MBrixia score was associated with a higher level of respiratory support at the time of performed CXR. The MBrixia score could potentially be valuable as a quantitative surrogate measurement of COVID-19 pneumonia severity, and future studies should investigate the score's validity and capabilities of predicting clinical outcomes.

Predicting the risk of chest radiograph abnormality 12-weeks post hospitalisation with SARS CoV-2 PCR confirmed COVID-19.

BACKGROUND: Routine follow-up of patients hospitalised with COVID-19 is recommended, however due to the ongoing high number of infections this is not without significant health resource and economic burden. In a previous study we investigated the prevalence of, and risk factors for, persistent chest radiograph (CXR) abnormalities post-hospitalisation with COVID-19 and identified a 5-point composite score that strongly predicted risk of persistent CXR abnormality at 12-weeks. Here we sought to validate and refine our findings in an independent cohort of patients. METHODOLOGY: A single-centre prospective study of consecutive patients attending a virtual post-hospitalisation COVID-19 clinic and CXR as part of their standard clinical care between 2nd March - 22nd June 2021. Inpatient and follow-up CXRs were scored by the assessing clinician for extent of pulmonary infiltrates (0-4 in each lung) with complete resolution defined as a follow-up score of zero. RESULTS: 182 consecutive patients were identified of which 31% had persistent CXR abnormality at 12-weeks. Patients with persistent CXR abnormality were significantly older (p < 0.001), had a longer hospital length of stay (p = 0.005), and had a higher incidence of both level 2 or 3 facility admission (level 2/3 care) (p = 0.003) and ever-smoking history (p = 0.038). Testing our composite score in the present cohort we found it predicted persistent CXR abnormality with reasonable accuracy (area under the receiver operator curve [AUROC 0.64]). Refining this score replacing obesity with Age ≥ 50 years, we identify the SHADE-750 score (1-point each for; Smoking history, Higher-level care (level 2/3 admission), Age ≥ 50 years, Duration of admission ≥ 15 days and Enzyme-lactate dehydrogenase (LDH ≥ 750U/L), that accurately predicted risk of persistent CXR abnormality, both in the present cohort (AUROC 0.73) and when retrospectively applied to our 1st cohort (AUROC 0.79). Applied to both cohorts combined (n = 213) it again performed strongly (AUROC 0.75) with all patients with a score of zero (n = 18) having complete CXR resolution at 12-weeks. CONCLUSIONS: In two independent cohorts of patients hospitalised with COVID-19, we identify a 5-point score which accurately predicts patients at risk of persistent CXR abnormality at 12-weeks. This tool could be used by clinicians to identify patients in which radiological follow-up may not be required.

Implementation of a Deep Learning-Based Computer-Aided Detection System for the Interpretation of Chest Radiographs in Patients Suspected for COVID-19.

OBJECTIVE: To describe the experience of implementing a deep learning-based computer-aided detection (CAD) system for the interpretation of chest X-ray radiographs (CXR) of suspected coronavirus disease (COVID-19) patients and investigate the diagnostic performance of CXR interpretation with CAD assistance. MATERIALS AND METHODS: In this single-center retrospective study, initial CXR of patients with suspected or confirmed COVID-19 were investigated. A commercialized deep learning-based CAD system that can identify various abnormalities on CXR was implemented for the interpretation of CXR in daily practice. The diagnostic performance of radiologists with CAD assistance were evaluated based on two different reference standards: 1) real-time reverse transcriptase-polymerase chain reaction (rRT-PCR) results for COVID-19 and 2) pulmonary abnormality suggesting pneumonia on chest CT. The turnaround times (TATs) of radiology reports for CXR and rRT-PCR results were also evaluated. RESULTS: Among 332 patients (male:female, 173:159; mean age, 57 years) with available rRT-PCR results, 16 patients (4.8%) were diagnosed with COVID-19. Using CXR, radiologists with CAD assistance identified rRT-PCR positive COVID-19 patients with sensitivity and specificity of 68.8% and 66.7%, respectively. Among 119 patients (male:female, 75:44; mean age, 69 years) with available chest CTs, radiologists assisted by CAD reported pneumonia on CXR with a sensitivity of 81.5% and a specificity of 72.3%. The TATs of CXR reports were significantly shorter than those of rRT-PCR results (median 51 vs. 507 minutes; p < 0.001). CONCLUSION: Radiologists with CAD assistance could identify patients with rRT-PCR-positive COVID-19 or pneumonia on CXR with a reasonably acceptable performance. In patients suspected with COVID-19, CXR had much faster TATs than rRT-PCRs.

Association of Chest Xray (CXR) findings with outcomes in COVID-19 Patients.

Objective: To investigate chest radiography findings in suspected coronavirus disease-2019 patients in a tertiary care setting. METHODS: The retrospective study was conducted at the Aga Khan University Hospital, Karachi, and comprised data of coronavirus disease-2019 cases admitted to the tertiary care centre from March 1 to March 30, 2020. A predesigned proforma was used to gather data, including demographics, like age and gender, co-morbidities, presenting symptoms and chest radiography findings during the admission. Length of stay and mortality were the outcome measures. Data was analysed using SPSS 22. RESULTS: Of the 154 suspected cases, 46(29.8%) tested positive for coronavirus disease-2019; 29(63%) males and 17(37%) females with a mean age of 50.7±19.1 years. Abnormal chest radiography was noted in 25(54.3%) cases, with bilateral pulmonary infiltrates being the most common finding 19(41.3%). Mortality was the outcome in 7(28%) of these cases, and the mean length of hospital stay was 9.3±7.3 days. Abnormal chest radiography findings were associated with an increased risk of mortality (p=0.009) and a longer hospital stay (p=0.017). Conclusion: Abnormal chest radiography findings were frequently seen in coronavirus disease-2019 patients and were also associated with increased risk of mortality and prolonged hospital stay.

Deep learning models for COVID-19 chest x-ray classification: Preventing shortcut learning using feature disentanglement.

In response to the COVID-19 global pandemic, recent research has proposed creating deep learning based models that use chest radiographs (CXRs) in a variety of clinical tasks to help manage the crisis. However, the size of existing datasets of CXRs from COVID-19+ patients are relatively small, and researchers often pool CXR data from multiple sources, for example, using different x-ray machines in various patient populations under different clinical scenarios. Deep learning models trained on such datasets have been shown to overfit to erroneous features instead of learning pulmonary characteristics in a phenomenon known as shortcut learning. We propose adding feature disentanglement to the training process. This technique forces the models to identify pulmonary features from the images and penalizes them for learning features that can discriminate between the original datasets that the images come from. We find that models trained in this way indeed have better generalization performance on unseen data; in the best case we found that it improved AUC by 0.13 on held out data. We further find that this outperforms masking out non-lung parts of the CXRs and performing histogram equalization, both of which are recently proposed methods for removing biases in CXR datasets.

Deep Convolutional Neural Network Mechanism Assessment of COVID-19 Severity.

As an epidemic, COVID-19's core test instrument still has serious flaws. To improve the present condition, all capabilities and tools available in this field are being used to combat the pandemic. Because of the contagious characteristics of the unique coronavirus (COVID-19) infection, an overwhelming comparison with patients queues up for pulmonary X-rays, overloading physicians and radiology and significantly impacting the quality of care, diagnosis, and outbreak prevention. Given the scarcity of clinical services such as intensive care and motorized ventilation systems in the aspect of this vastly transmissible ailment, it is critical to categorize patients as per their risk categories. This research describes a novel use of the deep convolutional neural network (CNN) technique to COVID-19 illness assessment seriousness. Utilizing chest X-ray images as contribution, an unsupervised DCNN model is constructed and suggested to split COVID-19 individuals into four seriousness classrooms: low, medium, serious, and crucial with an accuracy level of 96 percent. The efficiency of the DCNN model developed with the proposed methodology is demonstrated by empirical findings on a suitably huge sum of chest X-ray scans. To the evidence relating, it is the first COVID-19 disease incidence evaluation research with four different phases, to use a reasonably high number of X-ray images dataset and a DCNN with nearly all hyperparameters dynamically adjusted by the variable selection optimization task.