Chest Computed Tomography and Lung Ultrasound Findings in COVID-19 Pneumonia: A Pocket Review for Non-radiologists

Diagnostic Tools for Rapid Screening and Detection of SARS-CoV-2 Infection

The novel coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has severely impacted human health and the health management system globally. The ongoing pandemic has required the development of more effective diagnostic strategies for restricting deadly disease. For appropriate disease management, accurate and rapid screening and isolation of the affected population is an efficient means of containment and the decimation of the disease. Therefore, considerable efforts are being directed toward the development of rapid and robust diagnostic techniques for respiratory infections, including SARS-CoV-2. In this article, we have summarized the origin, transmission, and various diagnostic techniques utilized for the detection of the SARS-CoV-2 virus. These higher-end techniques can also detect the virus copy number in asymptomatic samples. Furthermore, emerging rapid, cost-effective, and point-of-care diagnostic devices capable of large-scale population screening for COVID-19 are discussed. Finally, some breakthrough developments based on spectroscopic diagnosis that could revolutionize the field of rapid diagnosis are discussed.

Review of Machine Learning in Lung Ultrasound in COVID-19 Pandemic

Ultrasound imaging of the lung has played an important role in managing patients with COVID-19-associated pneumonia and acute respiratory distress syndrome (ARDS). During the COVID-19 pandemic, lung ultrasound (LUS) or point-of-care ultrasound (POCUS) has been a popular diagnostic tool due to its unique imaging capability and logistical advantages over chest X-ray and CT. Pneumonia/ARDS is associated with the sonographic appearances of pleural line irregularities and B-line artefacts, which are caused by interstitial thickening and inflammation, and increase in number with severity. Artificial intelligence (AI), particularly machine learning, is increasingly used as a critical tool that assists clinicians in LUS image reading and COVID-19 decision making. We conducted a systematic review from academic databases (PubMed and Google Scholar) and preprints on arXiv or TechRxiv of the state-of-the-art machine learning technologies for LUS images in COVID-19 diagnosis. Openly accessible LUS datasets are listed. Various machine learning architectures have been employed to evaluate LUS and showed high performance. This paper will summarize the current development of AI for COVID-19 management and the outlook for emerging trends of combining AI-based LUS with robotics, telehealth, and other techniques.

Point-of-care ultrasound (PoCUS) in the early diagnosis of novel coronavirus 2019 disease (COVID-19) in a first-level emergency department during a SARS-CoV-2 outbreak in Italy: a real-life analysis

In December 2019, the severe acute respiratory syndrome coronavirus 2 (SARSCoV-2) spread worldwide, challenging emergency departments (EDs) with the need of rapid diagnosis for appropriate allocation in dedicated setting. Many authors highlighted the role of lung ultrasound (LUS) in management of the novel coronavirus disease 2019 (COVID-19). The study aims to analyze the performance of LUS in the early identification of COVID-19 patients in ED during a SARS-CoV-2 outbreak. We prospectively collected consecutive adult patients admitted to a first-level ED in Powered by Editorial Manager^® and ProduXion Manager^® from Aries Systems Corporation Florence with history or symptoms suggestive for COVID-19 that underwent LUS during the ED management. LUS findings were categorized in 6 discrete main etiological patterns. "A", "Cardiogenic B" and "Typical C" patterns were referred as non-COVID-19-suggestive, while "Atypical" B or C patterns, "Multiple Consolidations" pattern and "ARDS" pattern were referred as COVID-19-suggestive. The primary outcome was the diagnosis of SARS-CoV-2 infection. From 12 March to 12 May 2020, 360 patients were enrolled. COVID-19 suggestive LUS findings were significantly associated with final COVID-19 diagnosis (86% in COVID-19 vs 29% in non-COVID-19, p < 0.001). The presence in ED of at least one in positive swab OR a COVID-19-suggestive LUS showed a sensitivity of 97% and a negative predictive value (NPV) of 98%. In patients with known SARS-CoV-2 exposition in the last 14 days, a COVID-19-suggestive pattern at LUS had a positive predictive value (PPV) of 97% for COVID-19 diagnosis. Point-of-care ultrasound (PoCUS) is a valuable tool for diagnostic stratification during COVID-19 outbreaks. LUS can help physicians in identifying false-negative RT-PCR, improving its diagnostic sensitivity in ED.

Recent ultrasound advancements for the manipulation of nanobiomaterials and nanoformulations for drug delivery

Recent advances in ultrasound (US) have shown its great potential in biomedical applications as diagnostic and therapeutic tools. The coupling of US-assisted drug delivery systems with nanobiomaterials possessing tailor-made functions has been shown to remove the limitations of conventional drug delivery systems. The low-frequency US has significantly enhanced the targeted drug delivery effect and efficacy, reducing limitations posed by conventional treatments such as a limited therapeutic window. The acoustic cavitation effect induced by the US-mediated microbubbles (MBs) has been reported to replace drugs in certain acute diseases such as ischemic stroke. This review briefly discusses the US principles, with particular attention to the recent advancements in drug delivery applications. Furthermore, US-assisted drug delivery coupled with nanobiomaterials to treat different diseases (cancer, neurodegenerative disease, diabetes, thrombosis, and COVID-19) are discussed in detail. Finally, this review covers the future perspectives and challenges on the applications of US-mediated nanobiomaterials.

Classification of Lung Disease in Children by Using Lung Ultrasound Images and Deep Convolutional Neural Network

Bronchiolitis is the most common cause of hospitalization of children in the first year of life and pneumonia is the leading cause of infant mortality worldwide. Lung ultrasound technology (LUS) is a novel imaging diagnostic tool for the early detection of respiratory distress and offers several advantages due to its low-cost, relative safety, portability, and easy repeatability. More precise and efficient diagnostic and therapeutic strategies are needed. Deep-learning-based computer-aided diagnosis (CADx) systems, using chest X-ray images, have recently demonstrated their potential as a screening tool for pulmonary disease (such as COVID-19 pneumonia). We present the first computer-aided diagnostic scheme for LUS images of pulmonary diseases in children. In this study, we trained from scratch four state-of-the-art deep-learning models (VGG19, Xception, Inception-v3 and Inception-ResNet-v2) for detecting children with bronchiolitis and pneumonia. In our experiments we used a data set consisting of 5,907 images from 33 healthy infants, 3,286 images from 22 infants with bronchiolitis, and 4,769 images from 7 children suffering from bacterial pneumonia. Using four-fold cross-validation, we implemented one binary classification (healthy vs. bronchiolitis) and one three-class classification (healthy vs. bronchiolitis vs. bacterial pneumonia) out of three classes. Affine transformations were applied for data augmentation. Hyperparameters were optimized for the learning rate, dropout regularization, batch size, and epoch iteration. The Inception-ResNet-v2 model provides the highest classification performance, when compared with the other models used on test sets: for healthy vs. bronchiolitis, it provides 97.75% accuracy, 97.75% sensitivity, and 97% specificity whereas for healthy vs. bronchiolitis vs. bacterial pneumonia, the Inception-v3 model provides the best results with 91.5% accuracy, 91.5% sensitivity, and 95.86% specificity. We performed a gradient-weighted class activation mapping (Grad-CAM) visualization and the results were qualitatively evaluated by a pediatrician expert in LUS imaging: heatmaps highlight areas containing diagnostic-relevant LUS imaging-artifacts, e.g., A-, B-, pleural-lines, and consolidations. These complex patterns are automatically learnt from the data, thus avoiding hand-crafted features usage. By using LUS imaging, the proposed framework might aid in the development of an accessible and rapid decision support-method for diagnosing pulmonary diseases in children using LUS imaging.

A Comparison of Lung Ultrasound and Computed Tomography in the Diagnosis of Patients with COVID-19: A Systematic Review and Meta-Analysis

Background Lung ultrasound (LUS) and computed tomography (CT) can both be used for diagnosis of interstitial pneumonia caused by coronavirus disease 2019 (COVID-19), but the agreement between LUS and CT is unknown. Purpose to compare the agreement of LUS and CT in the diagnosis of interstitial pneumonia caused by COVID-19. Materials and Methods We searched PubMed, Cochrane library, Embase, Chinese Biomedicine Literature, and WHO COVID-19 databases to identify studies that compared LUS with CT in the diagnosis of interstitial pneumonia caused by COVID-19. We calculated the pooled overall, positive and negative percent agreements, diagnostic odds ratio (DOR) and the area under the standard receiver operating curve (SROC) for LUS in the diagnosis of COVID-19 compared with CT. Results We identified 1896 records, of which nine studies involving 531 patients were finally included. The pooled overall, positive and negative percentage agreements of LUS for the diagnosis of interstitial pneumonia caused by COVID-19 compared with CT were 81% (95% confidence interval [CI] 43–99%), 96% (95% CI, 80–99%, I² = 92.15%) and 80% (95%CI, 60–92%, I² = 92.85%), respectively. DOR was 37.41 (95% CI, 9.43–148.49, I² = 63.9%), and the area under the SROC curve was 0.94 (95% CI, 0.92–0.96). The quality of evidence for both specificity and sensitivity was low because of heterogeneity and risk of bias. Conclusion The level of diagnostic agreement between LUS and CT in the diagnosis of interstitial pneumonia caused by COVID-19 is high. LUS can be therefore considered as an equally accurate alternative for CT in situations where molecular tests are not available.

Clinical features and follow-up of pediatric patients hospitalized for COVID-19

OBJECTIVE

This report summarizes the clinical features and 1-month follow-up observations for pediatric patients who were hospitalized with coronavirus disease 2019 (COVID-19) in Wuhan Women and Children's Hospital.

METHODS

The 1-month follow-up data included clinical manifestations and results from serum severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) IgG and IgM tests, reverse-transcription polymerase chain reaction (RT-PCR) for SARS-CoV-2, lung computed tomography (CT) scans, and laboratory tests.

RESULTS

Between January 20 and March 15, 2020, 127 patients aged 0-15 years were hospitalized for COVID-19 treatment, including 3 severe cases and 124 mild or moderate cases. The main therapies included inhalation of aerosolized interferon-α (122/127) and additional antiviral drugs (28/127). Among the 81 patients who had pneumonia at admission, 35 with right lobe pneumonia had the longest hospital stay (mean 14.5 ± 7 days); 17 with left lobe pneumonia had the highest creatine kinase (154 ± 106 U/L) and creatine kinase myocardial band  (CK-MB, 43 ± 48 U/L) levels; and 29 with bilateral pneumonia had the highest white blood cell counts (8.3 ± 4 × 10⁹ /L). Among the 46 patients who were successfully followed up 1 month after discharge, two notable findings were right lobe pneumonia in 22% (95% confidence interval [CI]: 11%-37%) of patients and persistently elevated serum creatine kinase and CK-MB levels. The median duration of elevated CK-MB was 45 days. The mean concentrations of serum SARS-CoV-2 IgG and IgM in 41 patients were 8.0 ± 7.5 and 98 ± 40 ng/ml, respectively. At follow-up, four patients retested positive for SARS-CoV-2.

CONCLUSIONS

The involvement of different lung lobes in patients with COVID-19 was associated with variations in the persistence of pneumonia and elevation of CK-MB levels and body temperature.

Mini-COVIDNet: Efficient Lightweight Deep Neural Network for Ultrasound Based Point-of-Care Detection of COVID-19

Lung ultrasound (US) imaging has the potential to be an effective point-of-care test for detection of COVID-19, due to its ease of operation with minimal personal protection equipment along with easy disinfection. The current state-of-the-art deep learning models for detection of COVID-19 are heavy models that may not be easy to deploy in commonly utilized mobile platforms in point-of-care testing. In this work, we develop a lightweight mobile friendly efficient deep learning model for detection of COVID-19 using lung US images. Three different classes including COVID-19, pneumonia, and healthy were included in this task. The developed network, named as Mini-COVIDNet, was bench-marked with other lightweight neural network models along with state-of-the-art heavy model. It was shown that the proposed network can achieve the highest accuracy of 83.2% and requires a training time of only 24 min. The proposed Mini-COVIDNet has 4.39 times less number of parameters in the network compared to its next best performing network and requires a memory of only 51.29 MB, making the point-of-care detection of COVID-19 using lung US imaging plausible on a mobile platform. Deployment of these lightweight networks on embedded platforms shows that the proposed Mini-COVIDNet is highly versatile and provides optimal performance in terms of being accurate as well as having latency in the same order as other lightweight networks. The developed lightweight models are available at https://github.com/navchetan-awasthi/Mini-COVIDNet.

Coronavirus Disease 2019: An Overview of the Complications and Management

The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is the causative agent of coronavirus disease 2019 (COVID-19). Since the first report of COVID-19 emerging in Wuhan, China, authorities in 216 countries and territories have reported about 47.3 million COVID-19 cases and 1.2 million deaths. The WHO guidelines for the management of COVID-19 are very limited to recommendations for managing symptoms and advice on careful management of pediatric patients, pregnant women, and patients with underlying comorbidities. There is no approved treatment for COVID-19 and guidelines vary between countries. In this review, first, a brief overview is provided on the basic knowledge about the virus, clinical features of the disease, and different diagnostic methods. Then, the relationship between COVID-19, various body systems, and other complications is discussed. Finallly, different management strategies are discussed, including those drawn on computational chemistry analyses, pre-clinical investigations, and clinical trials which involve pharmacological and non-pharmacological interventions. In conclusion, despite the recent approval of different vaccine candidates, more virological characteristics of SARS-CoV-2 are required to be explored, which may result in the discovery of more potential therapeutic targets leading to safer and more effective treatment to COVID-19.

Longitudinal Chest CT Features in Severe/Critical COVID-19 Cases and the Predictive Value of the Initial CT for Mortality

Purpose

To evaluate longitudinal computed tomography (CT) features and the predictive value of the initial CT and clinical characteristics for mortality in patients with severe/critical coronavirus disease 2019 (COVID-19) pneumonia.

Methods

A retrospective analysis was performed on patients with COVID-19 pneumonia confirmed by laboratory. By excluding mild and common patients, 155 severe/critical patients with definite outcome were finally enrolled. A total of 516 CTs of 147 patients were divided into four stages according to the time after onset (stage 1, 1–7 days; stage 2, 8–14 days; stage 3, 15–21 days, and stage 4, >21 days). The evolving imaging features between the survival and non-survival groups were compared by using Chi-square, Fisher’s exact test, student’s t-test or Mann–Whitney U-test, as appropriate. The predictive value of clinical and CT features at admission for mortality was analysed through logistic regression analysis. To avoid overfitting caused by CT scores, CT scores were divided into two parts, which were combined with clinical variables, respectively, to construct the models.

Results

Ground-glass opacities (GGO) patterns were predominant for stages 1 and 2 for both groups (both P>0.05). The numbers of consolidation lesions increased in stage 3 in both groups (P=0.857), whereas the linear opacity increased in the survival group but decreased in the non-survival group (P=0.0049). In stage 4, the survival group predominantly presented linear opacity patterns, whereas the non-survival group mainly showed consolidation patterns (P=0.007). Clinical and imaging characteristics correlated with mortality; multivariate analyses revealed age >71 years, neutrophil count >6.38 × 10⁹/L, aspartate aminotransferase (AST) >58 IU/L, and CT score (total lesions score >17 in model 1, GGO score >14 and consolidation score >2 in model 2) as independent risk factors (all P<0.05). The areas under the curve of the six independent risk factors alone ranged from 0.65 to 0.75 and were 0.87 for model 2, 0.89 for model 1, and 0.92 for the six variables combined. Statistical differences were observed between Kaplan Meier curves of groups separated by cut-off values of these six variables (all P<0.01).

Conclusion

Longitudinal imaging features demonstrated differences between the two groups, which may help determine the patient’s prognosis. The initial CT score combined with age, AST, and neutrophil count is an excellent predictor for mortality in COVID-19 patients.

Ultrasound-responsive polymer-based drug delivery systems

Ultrasound-responsive polymeric materials have received a tremendous amount of attention from scientists for several decades. Compared to other stimuli-responsive materials (such as UV-, thermal-, and pH-responsive materials), these smart materials are more applicable since they allow more efficient drug delivery and targeted treatment by fairly non-invasive means. This review describes the recent advances of such ultrasound-responsive polymer-based drug delivery systems and illustrates various applications. More specifically, the mechanism of ultrasound-induced drug delivery, typical formulations, and biomedical applications (tumor therapy, disruption of blood-brain barrier, fighting infectious diseases, transdermal drug delivery, and enhanced thrombolysis) are summarized. Finally, a perspective on the future research directions for the development of ultrasound-responsive polymeric materials to facilitate a clinical translation is given.

Ferritin is associated with the severity of lung involvement but not with worse prognosis in patients with COVID-19: data from two Italian COVID-19 units

The coronavirus 2019 disease (COVID-19) is characterised by a heterogeneous clinical presentation, a complex pathophysiology and a wide range of imaging findings, depending on disease severity and time course. We conducted a retrospective evaluation of hospitalized patients with proven SARS-CoV-2 infection, clinical signs of COVID-19 and computed tomography (CT) scan-proven pulmonary involvement, in order to identify relationships between clinical, serological, imaging data and disease outcomes in patients with COVID-19. Clinical and serological records of patients admitted to two COVID-19 Units of the Abruzzo region in Italy with proven SARS-CoV-2 pulmonary involvement investigated with CT scan, assessed at the time of admission to the hospital, were retrospectively evaluated. Sixty-one patients (22 females and 39 males) of median age 65 years were enrolled. Fifty-six patients were discharged while death occurred in 5 patients. None of the lung abnormalities detected by CT was different between discharged and deceased patients. No differences were observed in the features and extent of pulmonary involvement according to age and gender. Logistic regression analysis with age and gender as covariates demonstrated that ferritin levels over the 25th percentile were associated with the involvement of all 5 pulmonary lobes (OR = 14.5, 95% CI 2.3–90.9, p = 0.004), the presence of septal thickening (OR = 8.2, 95% CI 1.6–40.9, p = 0.011) and the presence of mediastinal lymph node enlargement (OR = 12.0, 95% CI 1.1–127.5, p = 0.039) independently of age and gender. We demonstrated that ferritin levels over the 25th percentile are associated with a more severe pulmonary involvement, independently of age and gender and not associated with disease outcomes. The identification of reliable biomarkers in patients with COVID-19 may help guiding clinical decision, tailoring therapeutic approaches and ultimately improving the care and prognosis of patients with this disease.

Ultrasound assessment of pulmonary fibroproliferative changes in severe COVID-19: a quantitative correlation study with histopathological findings

PURPOSE

This study was designed to evaluate the usefulness of lung ultrasound (LUS) imaging to characterize the progression and severity of lung damage in cases of COVID-19.

METHODS

We employed a set of combined ultrasound parameters and histopathological images obtained simultaneously in 28 patients (15 women, 0.6-83 years) with fatal COVID-19 submitted to minimally invasive autopsies, with different times of disease evolution from initial symptoms to death (3-37 days, median 18 days). For each patient, we analysed eight post-mortem LUS parameters and the proportion of three histological patterns (normal lung, exudative diffuse alveolar damage [DAD] and fibroproliferative DAD) in eight different lung regions. The relationship between histopathological and post-mortem ultrasonographic findings was assessed using various statistical approaches.

RESULTS

Statistically significant positive correlations were observed between fibroproliferative DAD and peripheral consolidation (coefficient 0.43, p = 0.02) and pulmonary consolidation (coefficient 0.51, p = 0.005). A model combining age, time of evolution, sex and ultrasound score predicted reasonably well (r = 0.66) the proportion of pulmonary parenchyma with fibroproliferative DAD.

CONCLUSION

The present study adds information to previous studies related to the use of LUS as a tool to assess the severity of acute pulmonary damage. We provide a histological background that supports the concept that LUS can be used to characterize the progression and severity of lung damage in severe COVID-19.

Lung ultrasound findings in pediatric patients with COVID-19

During the pandemic caused by the novel coronavirus (COVID-19), lung ultrasound has been used to diagnose and monitor respiratory condition. The aim of the study was to describe lung ultrasound findings in children with a COVID-19 infection. Patients younger than 18 years old and positive for COVID-19, admitted to pediatric tertiary referral hospital were included. They were divided into two groups depending on the presence of respiratory symptoms. Lung ultrasound results were categorized into four degrees according to Soldati et al. score (J Ultrasound Med 39:1-7, 2020) and it was also described the presence and type of consolidation. Sixteen patients were recruited. The median age was 11 years old (IQR 2.8-12). Four children (25%) required admission to the intensive care unit. Six patients (37.5%) presented with respiratory symptoms. Most of them showed S.score of 2 and subpleural consolidations were observed in four cases (66.6%). Ten patients (62.5%) presented with non-respiratory symptoms, lung ultrasound showed S.score from 0 to 2. Three (30%) were diagnosed of multisystem inflammatory syndrome and lung ultrasounds showed S.score of 2 with bilateral pleural effusion.Conclusions: Children with COVID-19 and respiratory symptoms mostly showed a S.score of 2 and 3 with subpleural consolidations, upon the lung ultrasound assessment. What is Known: • Lung ultrasound is a useful tool for monitoring patients with respiratory symptoms in both adults and children. Lung ultrasounds are altered in adult patients with COVID-19. What is New: • Lung ultrasound might improve COVID-19 assessment, it could be a useful tool to diagnose and monitor patients throughout the COVID-19 pandemic. Even COVID-19 patients with non-respiratory symptoms have lung alterations that are visible on lung ultrasound.

How to perform and interpret the lung ultrasound by the obstetricians in pregnant women during the SARS-CoV-2 pandemic

Objective:

Evidence for the use of lung ultrasound scan (LUS) examinations in coronavirus 2019 pneumonia is rapidly growing. The safe and non-ionizing nature of LUS drew attention, particularly for pregnant women. This study aimed to contribute to the interpretation of LUS findings in pregnant women for the obstetricians.

Materials and Methods:

LUS was performed to pregnant women suspected of or diagnosed as having Severe Acute Respiratory syndrome coronavirus-2 (SARS-CoV-2) in the first 24 hours of admission. Fourteen areas (3 posterior, 2 lateral, and 2 anterior) were scanned per patient for at least 10 seconds along the indicated anatomical landmarks. The scan was performed in supine, right-sided and left-sided positions, respectively. Each area was given a score between 0 and 3 according to the specific pattern.

Results:

In this study, 21 still images and 21 videoclips that enabled dynamic and real-time evaluation were provided. Pleural line assessment, physiologic A-lines, pathologic B-lines, light beam pattern, white lung pattern, and specific patterns for quick recognition and evaluation are described.

Conclusion:

The potential advantages and limitations of LUS and its areas of use for obstetricians are discussed. LUS is a promising supplementary imaging tool during the SARS-CoV-2 pandemic. It is easy to perform and may be feasible in the hands of obstetricians after a brief didactic course. It may be a firstline imaging modality for pregnant women.