ABSTRACT
During the early phase of the COVID-19 pandemic, reverse transcriptase-polymerase chain reaction (RT-PCR) testing faced limitations, prompting the exploration of machine learning (ML) alternatives for diagnosis and prognosis. Providing a comprehensive appraisal of such decision support systems and their use in COVID-19 management can aid the medical community in making informed decisions during the risk assessment of their patients, especially in low-resource settings. Therefore, the objective of this study was to systematically review the studies that predicted the diagnosis of COVID-19 or the severity of the disease using ML. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA), we conducted a literature search of MEDLINE (OVID), Scopus, EMBASE, and IEEE Xplore from January 1 to June 31, 2020. The outcomes were COVID-19 diagnosis or prognostic measures such as death, need for mechanical ventilation, admission, and acute respiratory distress syndrome. We included peer-reviewed observational studies, clinical trials, research letters, case series, and reports. We extracted data about the study's country, setting, sample size, data source, dataset, diagnostic or prognostic outcomes, prediction measures, type of ML model, and measures of diagnostic accuracy. Bias was assessed using the Prediction model Risk Of Bias ASsessment Tool (PROBAST). This study was registered in the International Prospective Register of Systematic Reviews (PROSPERO), with the number CRD42020197109. The final records included for data extraction were 66. Forty-three (64%) studies used secondary data. The majority of studies were from Chinese authors (30%). Most of the literature (79%) relied on chest imaging for prediction, while the remainder used various laboratory indicators, including hematological, biochemical, and immunological markers. Thirteen studies explored predicting COVID-19 severity, while the rest predicted diagnosis. Seventy percent of the articles used deep learning models, while 30% used traditional ML algorithms. Most studies reported high sensitivity, specificity, and accuracy for the ML models (exceeding 90%). The overall concern about the risk of bias was "unclear" in 56% of the studies. This was mainly due to concerns about selection bias. ML may help identify COVID-19 patients in the early phase of the pandemic, particularly in the context of chest imaging. Although these studies reflect that these ML models exhibit high accuracy, the novelty of these models and the biases in dataset selection make using them as a replacement for the clinicians' cognitive decision-making questionable. Continued research is needed to enhance the robustness and reliability of ML systems in COVID-19 diagnosis and prognosis.
ABSTRACT
INTRODUCTION: Computed tomography (CT) imaging is one of the most commonly used diagnostic tools. Iodine-based contrast media (IBCM) are frequently administered intravenously to improve soft tissue contrast in a wide range of CT scans. Supply chain disruptions triggered by the SARS-CoV-19 pandemic led to a global shortage of IBCM in mid-2022. The purpose of this study was to explore the impact of this shortage on the delivery of healthcare in Western Australia. METHODS: We performed a single-centre retrospective analysis of the provision of CT studies, comparing historical patterns to the shortage period. We focussed our attention on the total number of CT scans (noncontrast CT [NCCT] and contrast-enhanced CT [CECT]) and also specifically CT pulmonary angiogram (CTPA) and CT neck angiogram with or without inclusion of circle of Willis (CTNA) examinations. We also examined whether a decrease was compensated by increasing frequency of alternate examinations such as ventilation/perfusion (V/Q) scans, carotid Doppler ultrasound studies and Magnetic Resonance Angiograms (MRAs). RESULTS: Since 2012, there has been an approximate linear increase in the frequency of CT examinations. During the period of contrast shortage, there was an abrupt drop-off by approximately 50% in the CECT, CTPA and CTNA groups compared with the preceding 6 weeks (49%, 55% and 44%, respectively, with P < 0.001 in all cases). During the contrast shortage, the frequency of V/Q scans increased fivefold (from 13 to 65; P < 0.001). However, the provision of carotid Doppler ultrasound studies and MRAs remained approximately stable in frequency across recent time intervals. CONCLUSION: Our findings demonstrate that the IBCM shortage crisis had a very significant impact on the delivery of healthcare. While V/Q scans could (partially) substitute for CTPA studies in suspected pulmonary emboli, there appeared to be no valid alternative for CTNA studies in stroke calls. The unexpected and critical shortage of IBCM forced healthcare professionals to conserve resources, prioritise indications, triage patients based on risk, explore alternate imaging strategies and prepare for similar events recurring in the future.
ABSTRACT
Introduction and Aim: Coronavirus disease (COVID-19) is a viral infection that can lead to severe respiratory disease. Radiological examinations mainly computed tomography (CT) and Chest x-ray (CXR) play a role in diagnosis, follow-up, and management of COVID-19 infection. The purpose of this study was to look into the extent of using chest imaging in COVID-19 infection, as well to see if chest imaging in COVID-19 infections is justified and guided by clinical recommendation in Mosul, Iraq. Material(s) and Method(s): This cross-sectional study involved 245 people (93 males and 152 females), infected previously with COVID-19 infection in Mosul, Iraq. The participants were asked to self-complete an anonymous questionnaire. Data obtained was subjected to statistical analysis. Result(s): The 245 participants had an average age of 25.7 +/-8.44 years. The study sample included 57 (23.2%) with low education and 188 (76.7%) with moderate to high education. Among the radiological examinations undergone by these participants, chest X-ray (CXR) was the most common followed by chest computed tomography scan (CT scan). The CXR and the CT scan were done during the patient's illness either for diagnosis or follow-up of the disease. Non-clinically recommended examinations were reported by 64% and 20% of patients who undertook CXR and CT scan respectively, during COVID-19 illness. Higher education status was associated with a tendency to do non-recommended CXR examination during COVID-19 infection. Conclusion(s): CXR and CT imaging are the most commonly used radiological examinations in the diagnosis and follow-up in COVID-19 infection. However, a non-clinically recommended utilization of these examinations was noted in Mosul, Iraq during the pandemic. Therefore, educating people of this region about the limitation of non-justified uses of imaging is essential for healthy maintenance of individuals, environment, and resources.Copyright © 2023, Indian Association of Biomedical Scientists. All rights reserved.
ABSTRACT
Objectives: Severe acute respiratory syndrome - coronavirus 2 (SARS-CoV-2) is a single-stranded positive ribonucleic acid virus of the coronaviridae family. The disease caused by this virus has been named by the World Health Organization coronavirus disease 19 (COVID-19), whose main manifestation is interstitial pneumonia. Aim of this study is to describe the radiological features of SARS-CoV-2 infection in its original form, to correlate the high-resolution computed tomography (HRCT) patterns with clinical findings, prognosis and mortality, and to establish the need for treatment and admission to the intensive care unit. Material and Methods: From March 2020 to May 2020, 193 patients (72 F and 121 M) who were swab positive for SARS-CoV-2 were retrospectively selected for our study. These patients underwent HRCT in the clinical suspicion of SARS-CoV-2 interstitial pneumonia. Results: Our results confirm the role of radiology and, in particular, of chest HRCT as a technique with high sensitivity in the recognition of the most peculiar features of COVID-19 pneumonia, in the evaluation of severity of the disease, in the correct interpretation of temporal changes of the radiological picture during the follow-up until the resolution, and in obtaining prognostic information, also to direct the treatment. Conclusion: Chest computed tomography cannot be considered as a substitute for real-time - polymerase chain reaction in the diagnosis of COVID-19, but rather supplementary to it in the diagnostic process as it can detect parenchymal changes at an early stage and even before the positive swab, at least for patients who have been symptomatic for more than 3 days.
ABSTRACT
COVID-19 has been raging for almost three years ever since its first outbreak. It is without a doubt that it is a common human goal to end the pandemic and how it was before it started. Many efforts have been made to work toward this goal. In computer vision, works have been done to aid medical professionals into faster and more effective procedures when dealing with the disease. For example, disease diagnosis and severity prediction using chest imaging. At the same time, vision transformer is introduced and quickly stormed its way into one of the best deep learning models ever developed due to its ability to achieve good performance while being resources friendly. In this study, we investigated the performance of ViT on COVID19 severity classification using an open-source CXR images dataset. We applied different augmentation and transformation techniques to the dataset to see ViT's ability to learn the features of the different severity levels of the disease. It is concluded that training ViT using the horizontally flipped images added to the original dataset gives the best overall accuracy of 0.862. To achieve explainability, we have also applied Grad-CAM to the best performing model to make sure it is looking at relevant region of the CXR image upon predicting the class label. © 2022 IEEE.
ABSTRACT
The dreadful coronavirus has not only shattered the lives of millions of people, but it has also placed enormous strain on the whole healthcare system. In order to isolate positive cases and stop the disease from spreading, early detection of COVID-19 is crucial. Currently, a laboratory test (RT-PCR) on samples collected from the throat and nose is required for the official diagnosis of COVID-19. Specialized tools are needed for the RT-PCR test, which takes at least 24 hours to complete. It may often provide more false negative and false positive results than expected. Therefore, using X-ray and CT scan images of the individual's lung, COVID-19 screening can be used to support the conventional RTPCR methods for an accurate clinical diagnosis. The importance of chest imaging in the emergence of this lung illness has been recognized. Images from the computed tomography (CT) scan and chest X-ray (CXR) can be used to quickly and accurately diagnose COVID-19. However, CT scan pictures have their own drawbacks. In order to assess the effectiveness of chest imaging approaches and demonstrates that CXR as an input may compete with CT scan pictures in the diagnosis of COVID-19 infection using various CNN based models, this article thoroughly covers modern deep learning techniques (CNN). For CXR and CT scan pictures, we have evaluated with ResNet, MobileNet, VGG 16, and EfficientNet. Both chest X-ray (3604 Images) and CT scans (3227 images) from publicly accessible databases have been evaluated, and the experimental outcomes are also contrasted. © 2022 IEEE.
ABSTRACT
Objective: To investigate the clinical features and chest CT findings in moderate and severe COVID-19 patients. Method(s): A total of 506 patients with COVID-19 treated in Wuhan Huoshenshan Hospital during February 9 to March 9, 2020 were enrolled in the study, including 365 moderate cases and 141 severe cases. The clinical features and chest CT findings were retrospectively analyzed. Chi-square test and Fisher's exact probability were used for data analysis. Result(s): The proportions of patients with diabetes and hypertension in severe group were significantly higher than those in moderate group (chi2=9.377 and 15.085, P<0.01). Compared with the severe patients, the white blood cell counts and lymphocyte counts of moderate patients were statistically significant (chi2=14.816 and 30.097, P<0.01). The protortion of increased CRP in severe patients was higher than that in moderate patients (chi2=21.639, P<0.01). The cure rate and discharge rate of severe patients were significantly lower than those of moderate patients (P<0.01). Compared with the moderate cases of COVID-19, the CT images in severe patients mainly showed lesions of diffuse distribution, mixed density, with maximum diameter>10 cm and involved all five lung lobes (P<0.01). The severe patients had more imaging signs of air bronchogram, bronchovascular thickening, pleural thickening, mediastinal or hilar lymphnodes enlargement, pleural effusion and pericardial effusion than moderate patients (chi2=33.357, 11.114, 14.580, 5.978, 45.731 and 6.623, P<0.05 or <0.01). Conclusion(s): There are differences in clinical features and chest CT findings between moderate and severe patients, and chest CT findings can be used as important criteria for clinical classification.Copyright © 2020 by the Chinese Medical Association.
ABSTRACT
Objective: To investigate the clinical features and chest CT findings in moderate and severe COVID-19 patients. Method(s): A total of 506 patients with COVID-19 treated in Wuhan Huoshenshan Hospital during February 9 to March 9, 2020 were enrolled in the study, including 365 moderate cases and 141 severe cases. The clinical features and chest CT findings were retrospectively analyzed. Chi-square test and Fisher's exact probability were used for data analysis. Result(s): The proportions of patients with diabetes and hypertension in severe group were significantly higher than those in moderate group (chi2=9.377 and 15.085, P<0.01). Compared with the severe patients, the white blood cell counts and lymphocyte counts of moderate patients were statistically significant (chi2=14.816 and 30.097, P<0.01). The protortion of increased CRP in severe patients was higher than that in moderate patients (chi2=21.639, P<0.01). The cure rate and discharge rate of severe patients were significantly lower than those of moderate patients (P<0.01). Compared with the moderate cases of COVID-19, the CT images in severe patients mainly showed lesions of diffuse distribution, mixed density, with maximum diameter>10 cm and involved all five lung lobes (P<0.01). The severe patients had more imaging signs of air bronchogram, bronchovascular thickening, pleural thickening, mediastinal or hilar lymphnodes enlargement, pleural effusion and pericardial effusion than moderate patients (chi2=33.357, 11.114, 14.580, 5.978, 45.731 and 6.623, P<0.05 or <0.01). Conclusion(s): There are differences in clinical features and chest CT findings between moderate and severe patients, and chest CT findings can be used as important criteria for clinical classification.Copyright © 2020 by the Chinese Medical Association.
ABSTRACT
Objective: To investigate the clinical features and chest CT findings in moderate and severe COVID-19 patients. Method(s): A total of 506 patients with COVID-19 treated in Wuhan Huoshenshan Hospital during February 9 to March 9, 2020 were enrolled in the study, including 365 moderate cases and 141 severe cases. The clinical features and chest CT findings were retrospectively analyzed. Chi-square test and Fisher's exact probability were used for data analysis. Result(s): The proportions of patients with diabetes and hypertension in severe group were significantly higher than those in moderate group (chi2=9.377 and 15.085, P<0.01). Compared with the severe patients, the white blood cell counts and lymphocyte counts of moderate patients were statistically significant (chi2=14.816 and 30.097, P<0.01). The protortion of increased CRP in severe patients was higher than that in moderate patients (chi2=21.639, P<0.01). The cure rate and discharge rate of severe patients were significantly lower than those of moderate patients (P<0.01). Compared with the moderate cases of COVID-19, the CT images in severe patients mainly showed lesions of diffuse distribution, mixed density, with maximum diameter>10 cm and involved all five lung lobes (P<0.01). The severe patients had more imaging signs of air bronchogram, bronchovascular thickening, pleural thickening, mediastinal or hilar lymphnodes enlargement, pleural effusion and pericardial effusion than moderate patients (chi2=33.357, 11.114, 14.580, 5.978, 45.731 and 6.623, P<0.05 or <0.01). Conclusion(s): There are differences in clinical features and chest CT findings between moderate and severe patients, and chest CT findings can be used as important criteria for clinical classification.Copyright © 2020 by the Chinese Medical Association.
ABSTRACT
The epidemic of coronavirus disease 2019 (COVID-19) has caused an ever-growing demand for treatment, testing, and diagnosis. Chest x-rays are a fast and low-cost test that can detect COVID19 but chest imaging is not a first-line test for COVID19 because of lower diagnosis performance and confounding with other viral pneumonia. Current studies using deep learning (DL) might assist in overcoming these issues as convolution neural networks (CNN) have illustrated higher performance of COVID19 diagnoses at the earlier phase. This study develops a new Firefly Optimization with Bidirectional Gated Recurrent Unit (FFO-BGRU) for COVID19 diagnoses on Chest Radiographs. The main intention of the FFO-BGRU technique lies in the recognition and classification of COVID-19 on Chest X-ray images. At the initial stage, the presented FFO-BGRU technique applies Wiener filtering (WF) technique for noise removal process. Followed, the hyperparameter tuning process takes place by using FFO algorithm and SqueezeNet architecture is applied for feature extraction. Lastly, the BGRU model is applied for COVID19 recognition and classification. A wide range of simulations were performed to demonstrate the betterment of the FFO-BGRU model. The comprehensive comparison study highlighted the improved outcomes of the FFO-BGRU algorithm over other recent approaches. © 2022 IEEE.
ABSTRACT
Background: Early in the pandemic, we established COVID-19 Recovery and Engagement (CORE) Clinics in the Bronx and implemented a detailed evaluation protocol to assess physical, emotional, and cognitive function, pulmonary function tests, and imaging for COVID-19 survivors. Here, we report our findings up to five months post-acute COVID-19. Methods: Main outcomes and measures included pulmonary function tests, imaging tests, and a battery of symptom, physical, emotional, and cognitive assessments 5 months post-acute COVID-19. Findings: Dyspnea, fatigue, decreased exercise tolerance, brain fog, and shortness of breath were the most common symptoms but there were generally no significant differences between hospitalized and non-hospitalized cohorts (p > 0.05). Many patients had abnormal physical, emotional, and cognitive scores, but most functioned independently; there were no significant differences between hospitalized and non-hospitalized cohorts (p > 0.05). Six-minute walk tests, lung ultrasound, and diaphragm excursion were abnormal but only in the hospitalized cohort. Pulmonary function tests showed moderately restrictive pulmonary function only in the hospitalized cohort but no obstructive pulmonary function. Newly detected major neurological events, microvascular disease, atrophy, and white-matter changes were rare, but lung opacity and fibrosis-like findings were common after acute COVID-19. Interpretation: Many COVID-19 survivors experienced moderately restrictive pulmonary function, and significant symptoms across the physical, emotional, and cognitive health domains. Newly detected brain imaging abnormalities were rare, but lung imaging abnormalities were common. This study provides insights into post-acute sequelae following SARS-CoV-2 infection in neurological and pulmonary systems which may be used to support at-risk patients and develop effective screening methods and interventions.
ABSTRACT
Recently the chest imaging plays an important role in COVID-19 diagnosis compared to laboratory diagnosis such as RT-PCR. This paper investigates a robust algorithm to detect infected COVID-19 patients using computed tomography scans. The proposed algorithm utilizes a deep learning approach through applying an off shelf pretrained neural network to extract a feature map matrix from the first layer of convolutional neural network, which preserve the basic features related to geometry structures. Later the trained features are classified using a machine learning classifier as support vector machine algorithm to classify the tested images into two classes infected against not infected. The investigated algorithm was trained and tested using two open-source datasets. The results were experimented using five general pretrained off shelf CNN architectures, the performance of the pretrained algorithm was measured and evaluated for the extracted image features with popular five pretrained CNN and classification accuracy 90%. © 2022 IEEE.
ABSTRACT
Background: Diagnosing a coronavirus disease 2019 (COVID-19) infection with high specificity in chest computed tomography (CT) imaging is considered possible due to distinctive imaging features of COVID-19 pneumonia. Since other viral non-COVID pneumonia show mostly a different distribution pattern, it is reasonable to assume that the patterns observed caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are a consequence of its genetically encoded molecular properties when interacting with the respiratory tissue. As more mutations of the initial SARS-CoV-2 wild-type with varying aggressiveness have been detected in the course of 2021, it became obvious that its genome is in a state of transformation and therefore a potential modification of the specific morphological appearance in CT may occur. The aim of this study was to quantitatively analyze the morphological differences of the SARS-CoV-2-B.1.1.7 mutation and wildtype variant in CT scans of the thorax. Methods: We analyzed a dataset of 140 patients, which was divided into pneumonias caused by n=40 wildtype variants, n=40 B.1.1.7 variants, n=20 bacterial pneumonias, n=20 viral (non-COVID) pneumonias, and a test group of n=20 unremarkable CT examinations of the thorax. Semiautomated 3D segmentation of the lung tissue was performed for quantification of lung pathologies. The extent, ratio, and specific distribution of inflammatory affected lung tissue in each group were compared in a multivariate group analysis. Results: Lung segmentation revealed significant difference between the extent of ground glass opacities (GGO) or consolidation comparing SARS-CoV-2 wild-type and B.1.1.7 variant. Wildtype and B.1.1.7 variant showed both a symmetric distribution pattern of stage-dependent GGO and consolidation within matched COVID-19 stages. Viral non-COVID pneumonias had significantly fewer consolidations than the bacterial, but also than the COVID-19 B.1.1.7 variant groups. Conclusions: CT based segmentation showed no significant difference between the morphological appearance of the COVID-19 wild-type variant and the SARS-CoV-2 B.1.1.7 mutation. However, our approach allowed a semiautomatic quantification of bacterial and viral lung pathologies. Quantitative CT image analyses, such as the one presented, appear to be an important component of pandemic preparedness considering an organism with ongoing genetic change, to describe a potential arising change in CT morphological appearance of possible new upcoming COVID-19 variants of concern.
ABSTRACT
(1) Background: Currently, multisystem inflammatory syndrome in children (MIS-C) is diagnosed based on clinical symptoms and laboratory findings of inflammation in the body. Once MIS-C is diagnosed, children will need to be followed over time. The imaging modalities most commonly used in the evaluation of patients with MIS-C include radiographs, ultrasound (US), and computed tomography (CT). Our study aims to summarise the literature data for the main gastrointestinal and pulmonary imaging features in children diagnosed with MIS-C and to share a single-centre experience. (2) Methods: We present the imaging findings in a cohort of 51 children diagnosed with MIS-C, admitted between December 2020 and February 2022. Imaging studies include chest and abdominal radiographs, thoracic, abdominal, and neck US and echocardiography (ECHO), and CT of the chest, abdomen, and pelvis. (3) Results: In accordance with the results in other studies, our observations show predominantly gastrointestinal involvement (GI) with ascites (33/51, 65%) and lymphadenopathy (19/51, 37%), ileitis or colitis (18/51, 35%), some cases of splenomegaly (9/51, 18%), hepatomegaly (8/51, 16%), and a few cases of renal enlargement (3/51, 6%) and gallbladder fossa oedema/wall thickening (2/51, 4%). Most common among the thoracic findings are posterior-basal consolidations (16/51, 31%), pleural effusion (14/51, 27%), and ground-glass opacities (12/51, 24%). We also register the significant involvement of the cardiovascular system with pericarditis (30/51, 58%), pericardial effusion (16/51, 31%), and myocarditis (6/51, 12%). (4) Conclusions: Radiologists should be aware of those imaging findings in order to take an important and active role not only in applying an accurate diagnosis, but also in the subsequent management of children with MIS-C. Radiological findings are not the primary diagnostic tool, but can assist in the evaluation of the affected systems and guide treatment.
ABSTRACT
BACKGROUND: We investigated whether COVID-19 leads to persistent impaired pulmonary function, fibrotic-like abnormalities or psychological symptoms 12 months after discharge and whether severely ill patients (ICU admission) recover differently than moderately ill patients. METHODS: This single-centre cohort study followed adult COVID-19 survivors for a period of one year after discharge. Patients underwent pulmonary function tests 6 weeks, 3 months and 12 months after discharge and were psychologically evaluated at 6 weeks and 12 months. Computed tomography (CT) was performed after 3 months and 12 months. RESULTS: 66 patients were analysed, their median age was 60.5 (IQR: 54-69) years, 46 (70%) patients were male. 38 (58%) patients had moderate disease and 28 (42%) patients had severe disease. Most patients had spirometric values within normal range after 12 months of follow-up. 12 (23%) patients still had an impaired lung diffusion after 12 months. Impaired pulmonary diffusion capacity was associated with residual CT abnormalities (OR 5.1,CI-95: 1.2-22.2), shortness of breath (OR 7.0, CI-95: 1.6-29.7) and with functional limitations (OR 5.8, CI-95: 1.4-23.8). Ground-glass opacities resolved in most patients during follow-up. Resorption of reticulation, bronchiectasis and curvilinear bands was rare and independent of disease severity. 81% of severely ill patients and 37% of moderately ill patients showed residual abnormalities after 12 months (OR 8.1, CI-95: 2.5-26.4). A minority of patients had symptoms of post-traumatic stress disorder, anxiety, depression and cognitive failure during follow-up. CONCLUSION: Some patients still had impaired lung diffusion 12 months after discharge and fibrotic-like residual abnormalities were notably prevalent, especially in severely ill patients.
Subject(s)
COVID-19 , Adult , Humans , Male , Middle Aged , Female , COVID-19/complications , COVID-19/epidemiology , Cohort Studies , Hospitalization , Patient Discharge , Patient Acuity , Disease ProgressionABSTRACT
BACKGROUND: Radiology requests and reports contain valuable information about diagnostic findings and indications, and transformer-based language models are promising for more accurate text classification. METHODS: In a retrospective study, 2256 radiologist-annotated radiology requests (8 classes) and reports (10 classes) were divided into training and testing datasets (90% and 10%, respectively) and used to train 32 models. Performance metrics were compared by model type (LSTM, Bertje, RobBERT, BERT-clinical, BERT-multilingual, BERT-base), text length, data prevalence, and training strategy. The best models were used to predict the remaining 40,873 cases' categories of the datasets of requests and reports. RESULTS: The RobBERT model performed the best after 4000 training iterations, resulting in AUC values ranging from 0.808 [95% CI (0.757-0.859)] to 0.976 [95% CI (0.956-0.996)] for the requests and 0.746 [95% CI (0.689-0.802)] to 1.0 [95% CI (1.0-1.0)] for the reports. The AUC for the classification of normal reports was 0.95 [95% CI (0.922-0.979)]. The predicted data demonstrated variability of both diagnostic yield for various request classes and request patterns related to COVID-19 hospital admission data. CONCLUSION: Transformer-based natural language processing is feasible for the multilabel classification of chest imaging request and report items. Diagnostic yield varies with the information in the requests.
Subject(s)
COVID-19 , Radiology , COVID-19/diagnostic imaging , Humans , Natural Language Processing , Research Report , Retrospective StudiesABSTRACT
OBJECTIVES: Pulmonary disease is a significant cause of morbidity and mortality in adults and children, but most of the world lacks diagnostic imaging for its assessment. Lung ultrasound is a portable, low-cost, and highly accurate imaging modality for assessment of pulmonary pathology including pneumonia, but its deployment is limited secondary to a lack of trained sonographers. In this study, we piloted a low-cost lung teleultrasound system in rural Peru during the COVID-19 pandemic using lung ultrasound volume sweep imaging (VSI) that can be operated by an individual without prior ultrasound training circumventing many obstacles to ultrasound deployment. DESIGN: Pilot study. SETTING: Study activities took place in five health centres in rural Peru. PARTICIPANTS: There were 213 participants presenting to rural health clinics. INTERVENTIONS: Individuals without prior ultrasound experience in rural Peru underwent brief training on how to use the teleultrasound system and perform lung ultrasound VSI. Subsequently, patients attending clinic were scanned by these previously ultrasound-naïve operators with the teleultrasound system. PRIMARY AND SECONDARY OUTCOME MEASURES: Radiologists examined the ultrasound imaging to assess its diagnostic value and identify any pathology. A random subset of 20% of the scans were analysed for inter-reader reliability. RESULTS: Lung VSI teleultrasound examinations underwent detailed analysis by two cardiothoracic attending radiologists. Of the examinations, 202 were rated of diagnostic image quality (94.8%, 95% CI 90.9% to 97.4%). There was 91% agreement between radiologists on lung ultrasound interpretation among a 20% sample of all examinations (κ=0.76, 95% CI 0.53 to 0.98). Radiologists were able to identify sequelae of COVID-19 with the predominant finding being B-lines. CONCLUSION: Lung VSI teleultrasound performed by individuals without prior training allowed diagnostic imaging of the lungs and identification of sequelae of COVID-19 infection. Deployment of lung VSI teleultrasound holds potential as a low-cost means to improve access to imaging around the world.
Subject(s)
COVID-19 , Adult , COVID-19/diagnostic imaging , Child , Humans , Lung/diagnostic imaging , Pandemics , Peru/epidemiology , Pilot Projects , Reproducibility of Results , Ultrasonography/methodsABSTRACT
Pneumonia is an acute lung infection caused by a variety of organisms, such as viruses, bacteria, or fungi, that poses a serious risk to vulnerable populations. The first step in the diagnosis and treatment of pneumonia is a prompt and accurate diagnosis, especially in the context of an epidemic outbreak such as COVID-19, where pneumonia is an important symptom. To provide tools for this purpose, this article evaluates the potential of three textural image characterisation methods, fractal dimension, radiomics, and superpixel-based histon, as biomarkers both to distinguish between healthy individuals and patients affected by pneumonia and to differentiate between potential pneumonia causes. The results show the ability of the textural characterisation methods tested to discriminate between nonpathological images and images with pneumonia, and how some of the generated models show the potential to characterise the general textural patterns that define viral and bacterial pneumonia, and the specific features associated with a COVID-19 infection. © 2022 IEEE.
ABSTRACT
BACKGROUND: In this study, we aimed to evaluate the thorax Computed Tomography (CT) findings of pediatric patients diagnosed with coronavirus disease-19 (COVID-19) and to discuss these findings in light of the results of adult patients from the literature. METHODS: The CT scans of pediatric patients (1-18 years old) with a diagnosis of COVID-19 by reverse transcriptase-polymerase chain reaction (RT-PCR) in our hospital between March 2020 and January 2021 were retrospectively reviewed. The scans were interpreted regarding the distribution and localization features, and involvement patterns including ground-glass opacity, consolidation, halo/reversed halo sign, interlobular septal thickening, air bronchograms and bronchiectasis. The frequencies of these findings in pediatric cases in our study were recorded. RESULTS: A total of 95 patients with a mean age of 13±4.6 years were included in this study. Among them, 34 (36%) had lesions associated with COVID-19 on CT scans. Bilateral involvement was detected in 15 (44%) while unilateral in 19 (56%) patients. Eighteen (53%) patients had single lobe involvement. In 16 (47%) patients a solitary lesion was detected and in 18 (53%) multiple lesions were present. Ground-glass opacity appearance was observed in 28 (82%), consolidation in 9 (26%), and ground-glass opacity with consolidation in 8 (24%), halo sign in 9 (26%), reversed halo sign in 2 (6%), interlobular septal thickening (interstitial thickening) in 1 (3%) patients. CONCLUSIONS: As symptoms are relatively milder in children with COVID-19, CT findings are less extensive than in adults. It is essential to know the thorax CT findings that aid in the diagnosis and follow-up of the disease.
Subject(s)
COVID-19 , Adolescent , Adult , COVID-19/diagnostic imaging , Child , Child, Preschool , Humans , Infant , Lung/diagnostic imaging , Lung/pathology , Retrospective Studies , SARS-CoV-2 , Tomography, X-Ray Computed/methodsABSTRACT
Background: South Africa has experienced multiple waves of the coronavirus disease 2019 (COVID-19) with little research documenting chest imaging features in an human immunodeficiency virus (HIV) and tuberculosis (TB) endemic region. Objectives: Describe the chest imaging features, demographics and clinical characteristics of COVID-19 in an urban population. Method: Retrospective, cross-sectional, review of chest radiographs and computed tomographies (CTs) of adults admitted to a tertiary hospital with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, between 01 May 2020 and 30 June 2020. Imaging was reviewed by three radiologists. Clinical parameters and laboratory data were analysed. Results: A total of 113 adult patients with a mean age of 46 years and 10 months were included. A total of 113 chest radiographs and six CTs were read. Nineteen patients were HIV-positive (16.8%), 40 were hypertensive and diabetic (35.4%), respectively, and one had TB (0.9%). Common symptoms included cough (n = 69; 61.6%), dyspnoea (n = 60; 53.1%) and fever (n = 46; 40.7%). Lower zone predominant ground glass opacities (58.4%) and consolidation (29.2%) were most frequent on chest radiographs. The right lower lobe was most involved (46.9% ground glass opacities and 17.7% consolidation), with relative sparing of the left upper lobe. Bilateral ground glass opacities (66.7%) were most common on CT. Among the HIV-positive, ground glass opacities and consolidation were less common than in HIV-negative or unknown patients (p = 0.037 and p = 0.05, respectively). Conclusion: COVID-19 in South Africa has similar chest imaging findings to those documented globally, with some differences between HIV-positive and HIV-negative or unknown patients. The authors corroborate relative sparing of the left upper lobe; however, further research is required to validate this currently unique local finding.