Automated Assessment of COVID-19 Reporting and Data System and Chest CT Severity Scores in Patients Suspected of Having COVID-19 Using Artificial Intelligence

Artificial intelligence with deep learning in nuclear medicine and radiology

The use of deep learning in medical imaging has increased rapidly over the past few years, finding applications throughout the entire radiology pipeline, from improved scanner performance to automatic disease detection and diagnosis. These advancements have resulted in a wide variety of deep learning approaches being developed, solving unique challenges for various imaging modalities. This paper provides a review on these developments from a technical point of view, categorizing the different methodologies and summarizing their implementation. We provide an introduction to the design of neural networks and their training procedure, after which we take an extended look at their uses in medical imaging. We cover the different sections of the radiology pipeline, highlighting some influential works and discussing the merits and limitations of deep learning approaches compared to other traditional methods. As such, this review is intended to provide a broad yet concise overview for the interested reader, facilitating adoption and interdisciplinary research of deep learning in the field of medical imaging.

Chest X-ray lung imaging features in pediatric COVID-19 and comparison with viral lower respiratory infections in young children

RATIONALE

Chest radiography (CXR) is a noninvasive imaging approach commonly used to evaluate lower respiratory tract infections (LRTIs) in children. However, the specific imaging patterns of pediatric coronavirus disease 2019 (COVID-19) on CXR, their relationship to clinical outcomes, and the possible differences from LRTIs caused by other viruses in children remain to be defined.

METHODS

This is a cross-sectional study of patients seen at a pediatric hospital with polymerase chain reaction (PCR)-confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (n = 95). Patients were subdivided in infants (0-2 years, n = 27), children (3-10 years, n = 27), and adolescents (11-19 years, n = 41). A sample of young children (0-2 years, n = 68) with other viral lower respiratory infections (LRTI) was included to compare their CXR features with the subset of infants (0-2 years) with COVID-19.

RESULTS

Forty-five percent of pediatric patients with COVID-19 were hospitalized and 20% required admission to intensive care unit (ICU). The most common abnormalities identified were ground-glass opacifications (GGO)/consolidations (35%) and increased peribronchial markings/cuffing (33%). GGO/consolidations were more common in older individuals and perihilar markings were more common in younger subjects. Subjects requiring hospitalization or ICU admission had significantly more GGO/consolidations in CXR (p < .05). Typical CXR features of pediatric viral LRTI (e.g., hyperinflation) were more common in non-COVID-19 viral LRTI cases than in COVID-19 cases (p < .05).

CONCLUSIONS

CXR may be a complemental exam in the evaluation of moderate or severe pediatric COVID-19 cases. The severity of GGO/consolidations seen in CXR is predictive of clinically relevant outcomes. Hyperinflation could potentially aid clinical assessment in distinguishing COVID-19 from other types of viral LRTI in young children.

Semi-quantitative CT severity scoring as a predictor of development of post-COVID syndrome

Background

Following COVID-19 pandemic, clinical description focused on the clinical presentation of patients in the acute stage of the disease. More recently, data have emerged that some patients continue to experience symptoms related to COVID-19 after the acute phase of infection has subsided (post-COVID syndrome). Although characteristics of post-COVID syndrome have been well described, less is known about the possible invitations during acute illnesses that can predict such syndrome. Our study is a prospective study aiming at assessment of CT severity scoring in the acute phase of COVID-19 pneumonia as a predictor for development of post-COVID syndrome in recovering patients.

Results

A total of 192 symptomatic COVID-19 patients between April 2020 and October 2020 were enrolled in this single-center study, and high-resolution chest CT examinations were evaluated for CT severity scoring. Data were matched with the long-term clinical outcome. CT severity score was significantly higher in patients who developed post-COVID symptoms (p < 0.001). A CT score of > 7 was associated with an increased risk and was found to be predictive of condition development with sensitivity (95.9%), specificity (96%), positive predictive value (95.92%), negative predictive value (96%), and accuracy (95.96%).

Conclusions

CT severity scoring can help in predicting the long-term outcome of COVID-19 patients with cutoff value of CT-SSS > 7 showing highest sensitivity and specificity for predicting development of post-COVID syndrome.

COVID-19 clinico-radiological mismatch: a proposal for a novel combined morphologic/volumetric CT severity score with blinded validation

Background

Some COVID-19 patients with similar quantitative CT measurements had variable clinical presentation and outcome. The absence of reasonable clinical explanations, such as pre-existing comorbidities or vascular complications, adds to the confusion. The authors believed that neglecting the impact of certain severe morphologic features could be an alternative radiological explanation. This study aims to optimize the initial CT staging of COVID-19 and propose a new combined morphologic/volumetric CT severity index (CTSI) to solve this clinico-radiological mismatch.

Results

This multi-center study included two major steps. The first step of the study entailed a standardized combined morphologic/volumetric CT severity analyses to propose a new optimized CTSI. This was conducted retrospectively during the period from June till September 2020. It included 379 acutely symptomatic COVID-19 patients. They were clinically classified according to their oxygen saturation and respiratory therapeutic requirements into three groups: group A (mild 298/79%), group B (borderline severity 57/15%), and group C (severe/critical 24/6%). The morphologic and volumetric assessment of their HRCT was analyzed according to severity, by two consultant radiologists in consensus. A new 25 point-CTSI has been created, combining eight morphological CT patterns [M1:M8; 8 points] and four grades of volumetric scores [S1:S4; 17 points]. The addition of the M5 pattern (air bubble sign), M6 pattern (early fibrosis and architectural distortion), or M7 pattern (crazy-paving) proved to increase the clinical severity. The second step of the study entailed a standardized blinded/independent validation analysis for the proposed CTSI. This was prospectively conducted on other 132 patients during October 2020 and independently performed by other two consultant radiologists. Validation results reached 80.2% sensitivity, 91.8% specificity, AUROC-curve = 0.8356, and 90.9% accuracy.

Conclusion

A new optimized CTSI with accepted validation is proposed for initial staging of COVID-19 patients, using combined morphologic/volumetric assessment instead of the quantitative assessment alone. It could solve the clinico-radiological mismatch among patients with similar quantitative CT results and variable clinical presentation during the absence of pre-existing comorbidities or vascular complications.

Artificial intelligence for imaging-based COVID-19 detection: Systematic review comparing added value of AI versus human readers

PURPOSE

A growing number of studies have examined whether Artificial Intelligence (AI) systems can support imaging-based diagnosis of COVID-19-caused pneumonia, including both gains in diagnostic performance and speed. However, what is currently missing is a combined appreciation of studies comparing human readers and AI.

METHODS

We followed PRISMA-DTA guidelines for our systematic review, searching EMBASE, PUBMED and Scopus databases. To gain insights into the potential value of AI methods, we focused on studies comparing the performance of human readers versus AI models or versus AI-supported human readings.

RESULTS

Our search identified 1270 studies, of which 12 fulfilled specific selection criteria. Concerning diagnostic performance, in testing datasets reported sensitivity was 42-100% (human readers, n = 9 studies), 60-95% (AI systems, n = 10) and 81-98% (AI-supported readers, n = 3), whilst reported specificity was 26-100% (human readers, n = 8), 61-96% (AI systems, n = 10) and 78-99% (AI-supported readings, n = 2). One study highlighted the potential of AI-supported readings for the assessment of lung lesion burden changes, whilst two studies indicated potential time savings for detection with AI.

CONCLUSIONS

Our review indicates that AI systems or AI-supported human readings show less performance variability (interquartile range) in general, and may support the differentiation of COVID-19 pneumonia from other forms of pneumonia when used in high-prevalence and symptomatic populations. However, inconsistencies related to study design, reporting of data, areas of risk of bias, as well as limitations of statistical analyses complicate clear conclusions. We therefore support efforts for developing critical elements of study design when assessing the value of AI for diagnostic imaging.

Robotics and artificial intelligence in healthcare during COVID-19 pandemic: A systematic review

The outbreak of the COVID-19 pandemic is unarguably the biggest catastrophe of the 21st century, probably the most significant global crisis after the second world war. The rapid spreading capability of the virus has compelled the world population to maintain strict preventive measures. The outrage of the virus has rampaged through the healthcare sector tremendously. This pandemic created a huge demand for necessary healthcare equipment, medicines along with the requirement for advanced robotics and artificial intelligence-based applications. The intelligent robot systems have great potential to render service in diagnosis, risk assessment, monitoring, telehealthcare, disinfection, and several other operations during this pandemic which has helped reduce the workload of the frontline workers remarkably. The long-awaited vaccine discovery of this deadly virus has also been greatly accelerated with AI-empowered tools. In addition to that, many robotics and Robotics Process Automation platforms have substantially facilitated the distribution of the vaccine in many arrangements pertaining to it. These forefront technologies have also aided in giving comfort to the people dealing with less addressed mental health complicacies. This paper investigates the use of robotics and artificial intelligence-based technologies and their applications in healthcare to fight against the COVID-19 pandemic. A systematic search following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) method is conducted to accumulate such literature, and an extensive review on 147 selected records is performed.

Gender Bias in Artificial Intelligence: Severity Prediction at an Early Stage of COVID-19

Artificial intelligence (AI) technologies have been applied in various medical domains to predict patient outcomes with high accuracy. As AI becomes more widely adopted, the problem of model bias is increasingly apparent. In this study, we investigate the model bias that can occur when training a model using datasets for only one particular gender and aim to present new insights into the bias issue. For the investigation, we considered an AI model that predicts severity at an early stage based on the medical records of coronavirus disease (COVID-19) patients. For 5,601 confirmed COVID-19 patients, we used 37 medical records, namely, basic patient information, physical index, initial examination findings, clinical findings, comorbidity diseases, and general blood test results at an early stage. To investigate the gender-based AI model bias, we trained and evaluated two separate models-one that was trained using only the male group, and the other using only the female group. When the model trained by the male-group data was applied to the female testing data, the overall accuracy decreased-sensitivity from 0.93 to 0.86, specificity from 0.92 to 0.86, accuracy from 0.92 to 0.86, balanced accuracy from 0.93 to 0.86, and area under the curve (AUC) from 0.97 to 0.94. Similarly, when the model trained by the female-group data was applied to the male testing data, once again, the overall accuracy decreased-sensitivity from 0.97 to 0.90, specificity from 0.96 to 0.91, accuracy from 0.96 to 0.91, balanced accuracy from 0.96 to 0.90, and AUC from 0.97 to 0.95. Furthermore, when we evaluated each gender-dependent model with the test data from the same gender used for training, the resultant accuracy was also lower than that from the unbiased model.

How does DICOM support big data management? Investigating its use in medical imaging community

The diagnostic imaging field is experiencing considerable growth, followed by increasing production of massive amounts of data. The lack of standardization and privacy concerns are considered the main barriers to big data capitalization. This work aims to verify whether the advanced features of the DICOM standard, beyond imaging data storage, are effectively used in research practice. This issue will be analyzed by investigating the publicly shared medical imaging databases and assessing how much the most common medical imaging software tools support DICOM in all its potential. Therefore, 100 public databases and ten medical imaging software tools were selected and examined using a systematic approach. In particular, the DICOM fields related to privacy, segmentation and reporting have been assessed in the selected database; software tools have been evaluated for reading and writing the same DICOM fields. From our analysis, less than a third of the databases examined use the DICOM format to record meaningful information to manage the images. Regarding software, the vast majority does not allow the management, reading and writing of some or all the DICOM fields. Surprisingly, if we observe chest computed tomography data sharing to address the COVID-19 emergency, there are only two datasets out of 12 released in DICOM format. Our work shows how the DICOM can potentially fully support big data management; however, further efforts are still needed from the scientific and technological community to promote the use of the existing standard, encouraging data sharing and interoperability for a concrete development of big data analytics.

The role of chest imaging in the diagnosis, management, and monitoring of coronavirus disease 2019 (COVID-19)

Coronavirus disease 2019 (COVID-19) pandemic has posed a major public health crisis all over the world. The role of chest imaging, especially computed tomography (CT), has evolved during the pandemic paralleling the accumulation of scientific evidence. In the early stage of the pandemic, the performance of chest imaging for COVID-19 has widely been debated especially in the context of comparison to real-time reverse transcription polymerase chain reaction. Current evidence is against the use of chest imaging for routine screening of COVID-19 contrary to the initial expectations. It still has an integral role to play, however, in its work up and staging, especially when assessing complications or disease progression. Chest CT is gold standard imaging modality for COVID-19 pneumonia; in some situations, chest X-ray or ultrasound may be an effective alternative. The most important role of radiologists in this context is to be able to identify those patients at greatest risk of imminent clinical decompensation by learning to stratify cases of COVID-19 on the basis of radiologic imaging in the most efficient and timely fashion possible. The present availability of multiple and more refined CT grading systems and classification is now making this task easier and thereby contributing to the recent improvements achieved in COVID-19 treatment and outcomes. In this article, evidence of chest imaging regarding diagnosis, management and monitoring of COVID-19 will be chronologically reviewed.

The Predictive Role of Artificial Intelligence-Based Chest CT Quantification in Patients with COVID-19 Pneumonia

We sought to analyze the prognostic value of laboratory and clinical data, and an artificial intelligence (AI)-based algorithm for Coronavirus disease 2019 (COVID-19) severity scoring, on CT-scans of patients hospitalized with COVID-19. Moreover, we aimed to determine personalized probabilities of clinical deterioration. Data of symptomatic patients with COVID-19 who underwent chest-CT-examination at the time of hospital admission between April and November 2020 were analyzed. COVID-19 severity score was automatically quantified for each pulmonary lobe as the percentage of affected lung parenchyma with the AI-based algorithm. Clinical deterioration was defined as a composite of admission to the intensive care unit, need for invasive mechanical ventilation, use of vasopressors or in-hospital mortality. In total 326 consecutive patients were included in the analysis (mean age 66.7 ± 15.3 years, 52.1% male) of whom 85 (26.1%) experienced clinical deterioration. In the multivariable regression analysis prior myocardial infarction (OR = 2.81, 95% CI = 1.12–7.04, p = 0.027), immunodeficiency (OR = 2.08, 95% CI = 1.02–4.25, p = 0.043), C-reactive protein (OR = 1.73, 95% CI = 1.32–2.33, p < 0.001) and AI-based COVID-19 severity score (OR = 1.08; 95% CI = 1.02–1.15, p = 0.013) appeared to be independent predictors of clinical deterioration. Personalized probability values were determined. AI-based COVID-19 severity score assessed at hospital admission can provide additional information about the prognosis of COVID-19, possibly serving as a useful tool for individualized risk-stratification.

Integrated Clinical and CT Based Artificial Intelligence Nomogram for Predicting Severity and Need for Ventilator Support in COVID-19 Patients: A Multi-Site Study

Almost 25% of COVID-19 patients end up in ICU needing critical mechanical ventilation support. There is currently no validated objective way to predict which patients will end up needing ventilator support, when the disease is mild and not progressed. N = 869 patients from two sites (D1: N = 822, D2: N = 47) with baseline clinical characteristics and chest CT scans were considered for this study. The entire dataset was randomly divided into 70% training, D1train (N = 606) and 30% test-set (Dtest: D1test (N = 216) + D2 (N = 47)). An expert radiologist delineated ground-glass-opacities (GGOs) and consolidation regions on a subset of D1train, (D1train_sub, N = 88). These regions were automatically segmented and used along with their corresponding CT volumes to train an imaging AI predictor (AIP) on D1train to predict the need of mechanical ventilators for COVID-19 patients. Finally, top five prognostic clinical factors selected using univariate analysis were integrated with AIP to construct an integrated clinical and AI imaging nomogram (ClAIN). Univariate analysis identified lactate dehydrogenase, prothrombin time, aspartate aminotransferase, %lymphocytes, albumin as top five prognostic clinical features. AIP yielded an AUC of 0.81 on Dtest and was independently prognostic irrespective of other clinical parameters on multivariable analysis (p<0.001). ClAIN improved the performance over AIP yielding an AUC of 0.84 (p = 0.04) on Dtest. ClAIN outperformed AIP in predicting which COVID-19 patients ended up needing a ventilator. Our results across multiple sites suggest that ClAIN could help identify COVID-19 with severe disease more precisely and likely to end up on a life-saving mechanical ventilation.

CARes-UNet: Content-aware residual UNet for lesion segmentation of COVID-19 from chest CT images

PURPOSE

Coronavirus disease 2019 (COVID-19) has caused a serious global health crisis. It has been proven that the deep learning method has great potential to assist doctors in diagnosing COVID-19 by automatically segmenting the lesions in computed tomography (CT) slices. However, there are still several challenges restricting the application of these methods, including high variation in lesion characteristics and low contrast between lesion areas and healthy tissues. Moreover, the lack of high-quality labeled samples and large number of patients lead to the urgency to develop a high accuracy model, which performs well not only under supervision but also with semi-supervised methods.

METHODS

We propose a content-aware lung infection segmentation deep residual network (content-aware residual UNet (CARes-UNet)) to segment the lesion areas of COVID-19 from the chest CT slices. In our CARes-UNet, the residual connection was used in the convolutional block, which alleviated the degradation problem during the training. Then, the content-aware upsampling modules were introduced to improve the performance of the model while reducing the computation cost. Moreover, to achieve faster convergence, an advanced optimizer named Ranger was utilized to update the model's parameters during training. Finally, we employed a semi-supervised segmentation framework to deal with the problem of lacking pixel-level labeled data.

RESULTS

We evaluated our approach using three public datasets with multiple metrics and compared its performance to several models. Our method outperforms other models in multiple indicators, for instance in terms of Dice coefficient on COVID-SemiSeg Dataset, CARes-UNet got the score 0.731, and semi-CARes-UNet further boosted it to 0.776. More ablation studies were done and validated the effectiveness of each key component of our proposed model.

CONCLUSIONS

Compared with the existing neural network methods applied to the COVID-19 lesion segmentation tasks, our CARes-UNet can gain more accurate segmentation results, and semi-CARes-UNet can further improve it using semi-supervised learning methods while presenting a possible way to solve the problem of lack of high-quality annotated samples. Our CARes-UNet and semi-CARes-UNet can be used in artificial intelligence-empowered computer-aided diagnosis system to improve diagnostic accuracy in this ongoing COVID-19 pandemic.

Quantification of pulmonary involvement in COVID-19 pneumonia by means of a cascade of two U-nets: training and assessment on multiple datasets using different annotation criteria

Purpose

This study aims at exploiting artificial intelligence (AI) for the identification, segmentation and quantification of COVID-19 pulmonary lesions. The limited data availability and the annotation quality are relevant factors in training AI-methods. We investigated the effects of using multiple datasets, heterogeneously populated and annotated according to different criteria.

Methods

We developed an automated analysis pipeline, the LungQuant system, based on a cascade of two U-nets. The first one (U-net\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_1$$\end{document}1) is devoted to the identification of the lung parenchyma; the second one (U-net\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$_2$$\end{document}2) acts on a bounding box enclosing the segmented lungs to identify the areas affected by COVID-19 lesions. Different public datasets were used to train the U-nets and to evaluate their segmentation performances, which have been quantified in terms of the Dice Similarity Coefficients. The accuracy in predicting the CT-Severity Score (CT-SS) of the LungQuant system has been also evaluated.

Results

Both the volumetric DSC (vDSC) and the accuracy showed a dependency on the annotation quality of the released data samples. On an independent dataset (COVID-19-CT-Seg), both the vDSC and the surface DSC (sDSC) were measured between the masks predicted by LungQuant system and the reference ones. The vDSC (sDSC) values of 0.95±0.01 and 0.66±0.13 (0.95±0.02 and 0.76±0.18, with 5 mm tolerance) were obtained for the segmentation of lungs and COVID-19 lesions, respectively. The system achieved an accuracy of 90% in CT-SS identification on this benchmark dataset.

Conclusion

We analysed the impact of using data samples with different annotation criteria in training an AI-based quantification system for pulmonary involvement in COVID-19 pneumonia. In terms of vDSC measures, the U-net segmentation strongly depends on the quality of the lesion annotations. Nevertheless, the CT-SS can be accurately predicted on independent test sets, demonstrating the satisfactory generalization ability of the LungQuant.

Supplementary Information

The online version supplementary material available at 10.1007/s11548-021-02501-2.

Clinical Artificial Intelligence Applications in Radiology: Chest and Abdomen

Organ segmentation, chest radiograph classification, and lung and liver nodule detections are some of the popular artificial intelligence (AI) tasks in chest and abdominal radiology due to the wide availability of public datasets. AI algorithms have achieved performance comparable to humans in less time for several organ segmentation tasks, and some lesion detection and classification tasks. This article introduces the current published articles of AI applied to chest and abdominal radiology, including organ segmentation, lesion detection, classification, and predicting prognosis.

Can Deep Learning-Based Volumetric Analysis Predict Oxygen Demand Increase in Patients with COVID-19 Pneumonia?

Background and Objectives: This study aimed to investigate whether predictive indicators for the deterioration of respiratory status can be derived from the deep learning data analysis of initial chest computed tomography (CT) scans of patients with coronavirus disease 2019 (COVID-19). Materials and Methods: Out of 117 CT scans of 75 patients with COVID-19 admitted to our hospital between April and June 2020, we retrospectively analyzed 79 CT scans that had a definite time of onset and were performed prior to any medication intervention. Patients were grouped according to the presence or absence of increased oxygen demand after CT scan. Quantitative volume data of lung opacity were measured automatically using a deep learning-based image analysis system. The sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) of the opacity volume data were calculated to evaluate the accuracy of the system in predicting the deterioration of respiratory status. Results: All 79 CT scans were included (median age, 62 years (interquartile range, 46–77 years); 56 (70.9%) were male. The volume of opacity was significantly higher for the increased oxygen demand group than for the nonincreased oxygen demand group (585.3 vs. 132.8 mL, p < 0.001). The sensitivity, specificity, and AUC were 76.5%, 68.2%, and 0.737, respectively, in the prediction of increased oxygen demand. Conclusion: Deep learning-based quantitative analysis of the affected lung volume in the initial CT scans of patients with COVID-19 can predict the deterioration of respiratory status to improve treatment and resource management.

A Promising and Challenging Approach: Radiologists' Perspective on Deep Learning and Artificial Intelligence for Fighting COVID-19

Chest X-rays (CXR) and computed tomography (CT) are the main medical imaging modalities used against the increased worldwide spread of the 2019 coronavirus disease (COVID-19) epidemic. Machine learning (ML) and artificial intelligence (AI) technology, based on medical imaging fully extracting and utilizing the hidden information in massive medical imaging data, have been used in COVID-19 research of disease diagnosis and classification, treatment decision-making, efficacy evaluation, and prognosis prediction. This review article describes the extensive research of medical image-based ML and AI methods in preventing and controlling COVID-19, and summarizes their characteristics, differences, and significance in terms of application direction, image collection, and algorithm improvement, from the perspective of radiologists. The limitations and challenges faced by these systems and technologies, such as generalization and robustness, are discussed to indicate future research directions.

Detection and characterization of COVID-19 findings in chest CT: Feasibility and applicability of an AI-based software tool

ABSTRACT

The COVID-19 pandemic has challenged institutions' diagnostic processes worldwide. The aim of this study was to assess the feasibility of an artificial intelligence (AI)-based software tool that automatically evaluates chest computed tomography for findings of suspected COVID-19.Two groups were retrospectively evaluated for COVID-19-associated ground glass opacities of the lungs (group A: real-time polymerase chain reaction positive COVID patients, n = 108; group B: asymptomatic pre-operative group, n = 88). The performance of an AI-based software assessment tool for detection of COVID-associated abnormalities was compared with human evaluation based on COVID-19 reporting and data system (CO-RADS) scores performed by 3 readers.All evaluated variables of the AI-based assessment showed significant differences between the 2 groups (P < .01). The inter-reader reliability of CO-RADS scoring was 0.87. The CO-RADS scores were substantially higher in group A (mean 4.28) than group B (mean 1.50). The difference between CO-RADS scoring and AI assessment was statistically significant for all variables but showed good correlation with the clinical context of the CO-RADS score. AI allowed to predict COVID positive cases with an accuracy of 0.94.The evaluated AI-based algorithm detects COVID-19-associated findings with high sensitivity and may support radiologic workflows during the pandemic.

A practical artificial intelligence system to diagnose COVID-19 using computed tomography: A multinational external validation study

Computed tomography has gained an important role in the early diagnosis of COVID-19 pneumonia. However, the ever-increasing number of patients has overwhelmed radiology departments and has caused a reduction in quality of services. Artificial intelligence (AI) systems are the remedy to the current situation. However, the lack of application in real-world conditions has limited their consideration in clinical settings. This study validated a clinical AI system, COVIDiag, to aid radiologists in accurate and rapid evaluation of COVID-19 cases. 50 COVID-19 and 50 non-COVID-19 pneumonia cases were included from each of five centers: Argentina, Turkey, Iran, Netherlands, and Italy. The Dutch database included only 50 COVID-19 cases. The performance parameters namely sensitivity, specificity, accuracy, and area under the ROC curve (AUC) were computed for each database using COVIDiag model. The most common pattern of involvement among COVID-19 cases in all databases were bilateral involvement of upper and lower lobes with ground-glass opacities. The best sensitivity of 92.0% was recorded for the Italian database. The system achieved an AUC of 0.983, 0.914, 0.910, and 0.882 for Argentina, Turkey, Iran, and Italy, respectively. The model obtained a sensitivity of 86.0% for the Dutch database. COVIDiag model could diagnose COVID-19 pneumonia in all of cohorts with AUC of 0.921 (sensitivity, specificity, and accuracy of 88.8%, 87.0%, and 88.0%, respectively). Our study confirmed the accuracy of our proposed AI model (COVIDiag) in the diagnosis of COVID-19 cases. Furthermore, the system demonstrated consistent optimal diagnostic performance on multinational databases, which is critical to determine the generalizability and objectivity of the proposed COVIDiag model. Our results are significant as they provide real-world evidence regarding the applicability of AI systems in clinical medicine.

Automated Detection and Quantification of COVID-19 Airspace Disease on Chest Radiographs: A Novel Approach Achieving Expert Radiologist-Level Performance Using a Deep Convolutional Neural Network Trained on Digital Reconstructed Radiographs From Computed Tomography-Derived Ground Truth

OBJECTIVES

The aim of this study was to leverage volumetric quantification of airspace disease (AD) derived from a superior modality (computed tomography [CT]) serving as ground truth, projected onto digitally reconstructed radiographs (DRRs) to (1) train a convolutional neural network (CNN) to quantify AD on paired chest radiographs (CXRs) and CTs, and (2) compare the DRR-trained CNN to expert human readers in the CXR evaluation of patients with confirmed COVID-19.

MATERIALS AND METHODS

We retrospectively selected a cohort of 86 COVID-19 patients (with positive reverse transcriptase-polymerase chain reaction test results) from March to May 2020 at a tertiary hospital in the northeastern United States, who underwent chest CT and CXR within 48 hours. The ground-truth volumetric percentage of COVID-19-related AD (POv) was established by manual AD segmentation on CT. The resulting 3-dimensional masks were projected into 2-dimensional anterior-posterior DRR to compute area-based AD percentage (POa). A CNN was trained with DRR images generated from a larger-scale CT dataset of COVID-19 and non-COVID-19 patients, automatically segmenting lungs, AD, and quantifying POa on CXR. The CNN POa results were compared with POa quantified on CXR by 2 expert readers and to the POv ground truth, by computing correlations and mean absolute errors.

RESULTS

Bootstrap mean absolute error and correlations between POa and POv were 11.98% (11.05%-12.47%) and 0.77 (0.70-0.82) for average of expert readers and 9.56% to 9.78% (8.83%-10.22%) and 0.78 to 0.81 (0.73-0.85) for the CNN, respectively.

CONCLUSIONS

Our CNN trained with DRR using CT-derived airspace quantification achieved expert radiologist level of accuracy in the quantification of AD on CXR in patients with positive reverse transcriptase-polymerase chain reaction test results for COVID-19.

A deep-learning-based framework for severity assessment of COVID-19 with CT images

Millions of positive COVID-19 patients are suffering from the pandemic around the world, a critical step in the management and treatment is severity assessment, which is quite challenging with the limited medical resources. Currently, several artificial intelligence systems have been developed for the severity assessment. However, imprecise severity assessment and insufficient data are still obstacles. To address these issues, we proposed a novel deep-learning-based framework for the fine-grained severity assessment using 3D CT scans, by jointly performing lung segmentation and lesion segmentation. The main innovations in the proposed framework include: 1) decomposing 3D CT scan into multi-view slices for reducing the complexity of 3D model, 2) integrating prior knowledge (dual-Siamese channels and clinical metadata) into our model for improving the model performance. We evaluated the proposed method on 1301 CT scans of 449 COVID-19 cases collected by us, our method achieved an accuracy of 86.7% for four-way classification, with the sensitivities of 92%, 78%, 95%, 89% for four stages. Moreover, ablation study demonstrated the effectiveness of the major components in our model. This indicates that our method may contribute a potential solution to severity assessment of COVID-19 patients using CT images and clinical metadata.

The Applications of Artificial Intelligence in Chest Imaging of COVID-19 Patients: A Literature Review

Diagnostic imaging is regarded as fundamental in the clinical work-up of patients with a suspected or confirmed COVID-19 infection. Recent progress has been made in diagnostic imaging with the integration of artificial intelligence (AI) and machine learning (ML) algorisms leading to an increase in the accuracy of exam interpretation and to the extraction of prognostic information useful in the decision-making process. Considering the ever expanding imaging data generated amid this pandemic, COVID-19 has catalyzed the rapid expansion in the application of AI to combat disease. In this context, many recent studies have explored the role of AI in each of the presumed applications for COVID-19 infection chest imaging, suggesting that implementing AI applications for chest imaging can be a great asset for fast and precise disease screening, identification and characterization. However, various biases should be overcome in the development of further ML-based algorithms to give them sufficient robustness and reproducibility for their integration into clinical practice. As a result, in this literature review, we will focus on the application of AI in chest imaging, in particular, deep learning, radiomics and advanced imaging as quantitative CT.

BS-Net: Learning COVID-19 pneumonia severity on a large chest X-ray dataset

In this work we design an end-to-end deep learning architecture for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients. Such semi-quantitative scoring system, namely Brixia score, is applied in serial monitoring of such patients, showing significant prognostic value, in one of the hospitals that experienced one of the highest pandemic peaks in Italy. To solve such a challenging visual task, we adopt a weakly supervised learning strategy structured to handle different tasks (segmentation, spatial alignment, and score estimation) trained with a "from-the-part-to-the-whole" procedure involving different datasets. In particular, we exploit a clinical dataset of almost 5,000 CXR annotated images collected in the same hospital. Our BS-Net demonstrates self-attentive behavior and a high degree of accuracy in all processing stages. Through inter-rater agreement tests and a gold standard comparison, we show that our solution outperforms single human annotators in rating accuracy and consistency, thus supporting the possibility of using this tool in contexts of computer-assisted monitoring. Highly resolved (super-pixel level) explainability maps are also generated, with an original technique, to visually help the understanding of the network activity on the lung areas. We also consider other scores proposed in literature and provide a comparison with a recently proposed non-specific approach. We eventually test the performance robustness of our model on an assorted public COVID-19 dataset, for which we also provide Brixia score annotations, observing good direct generalization and fine-tuning capabilities that highlight the portability of BS-Net in other clinical settings. The CXR dataset along with the source code and the trained model are publicly released for research purposes.

Automatic lung segmentation in COVID-19 patients: Impact on quantitative computed tomography analysis

Purpose

To assess the impact of lung segmentation accuracy in an automatic pipeline for quantitative analysis of CT images.

Methods

Four different platforms for automatic lung segmentation based on convolutional neural network (CNN), region-growing technique and atlas-based algorithm were considered. The platforms were tested using CT images of 55 COVID-19 patients with severe lung impairment. Four radiologists assessed the segmentations using a 5-point qualitative score (QS). For each CT series, a manually revised reference segmentation (RS) was obtained. Histogram-based quantitative metrics (QM) were calculated from CT histogram using lung segmentationsfrom all platforms and RS. Dice index (DI) and differences of QMs (ΔQMs) were calculated between RS and other segmentations.

Results

Highest QS and lower ΔQMs values were associated to the CNN algorithm. However, only 45% CNN segmentations were judged to need no or only minimal corrections, and in only 17 cases (31%), automatic segmentations provided RS without manual corrections. Median values of the DI for the four algorithms ranged from 0.993 to 0.904. Significant differences for all QMs calculated between automatic segmentations and RS were found both when data were pooled together and stratified according to QS, indicating a relationship between qualitative and quantitative measurements. The most unstable QM was the histogram 90th percentile, with median ΔQMs values ranging from 10HU and 158HU between different algorithms.

Conclusions

None of tested algorithms provided fully reliable segmentation. Segmentation accuracy impacts differently on different quantitative metrics, and each of them should be individually evaluated according to the purpose of subsequent analyses.

"Standalone Ayurvedic treatment of high-risk COVID-19 patients with multiple comorbidities: A case series."

We report a case-series of Ayurvedic treatment in seven COVID-19 positive patients with multiple co-morbidities, categorized as high-risk for poor outcome from SARS-CoV-2 infection. All of them recovered completely from their illness with resolution of symptoms following Ayurvedic treatment. The data was collected from patients treated during the early months of the COVID-19 pandemic (June 2020 to September 2020) at an out-patient Ayurvedic Clinic, Chennai, India.

This is a retrospective case series from among the initial 247 COVID-19 patients out of whom 39% were found to be suffering from co-morbidities. We have chosen seven of these patients who fulfilled the criteria for high-risk category, represented by multiple co-morbidities that included cancer, chronic kidney disease (CKD), coronary artery disease (CAD), chronic obstructive pulmonary disease (COPD), diabetes mellitus (DM), hypertension, and an elderly person over the age of 90 years.

Classical Ayurvedic formulations for COVID -19 were chosen so as to avoid complicating co-morbid conditions and patients were maintained on a modified diet. All these high-risk patients were treated at an out-patient setting. The patients were under home quarantine and self-monitored their progress with daily follow-up over the phone by the treating Ayurvedic physician.

The main outcome measure included resolution of symptoms and complete recovery from COVID-19 disease in all patients.

This case series demonstrates the scope of Ayurvedic interventions in the management of high-risk COVID-19 patients with severe co-morbidities with successful outcome in an out-patient setting.

Integrating Digital Technologies and Public Health to Fight Covid-19 Pandemic: Key Technologies, Applications, Challenges and Outlook of Digital Healthcare

Integration of digital technologies and public health (or digital healthcare) helps us to fight the Coronavirus Disease 2019 (COVID-19) pandemic, which is the biggest public health crisis humanity has faced since the 1918 Influenza Pandemic. In order to better understand the digital healthcare, this work conducted a systematic and comprehensive review of digital healthcare, with the purpose of helping us combat the COVID-19 pandemic. This paper covers the background information and research overview of digital healthcare, summarizes its applications and challenges in the COVID-19 pandemic, and finally puts forward the prospects of digital healthcare. First, main concepts, key development processes, and common application scenarios of integrating digital technologies and digital healthcare were offered in the part of background information. Second, the bibliometric techniques were used to analyze the research output, geographic distribution, discipline distribution, collaboration network, and hot topics of digital healthcare before and after COVID-19 pandemic. We found that the COVID-19 pandemic has greatly accelerated research on the integration of digital technologies and healthcare. Third, application cases of China, EU and U.S using digital technologies to fight the COVID-19 pandemic were collected and analyzed. Among these digital technologies, big data, artificial intelligence, cloud computing, 5G are most effective weapons to combat the COVID-19 pandemic. Applications cases show that these technologies play an irreplaceable role in controlling the spread of the COVID-19. By comparing the application cases in these three regions, we contend that the key to China's success in avoiding the second wave of COVID-19 pandemic is to integrate digital technologies and public health on a large scale without hesitation. Fourth, the application challenges of digital technologies in the public health field are summarized. These challenges mainly come from four aspects: data delays, data fragmentation, privacy security, and data security vulnerabilities. Finally, this study provides the future application prospects of digital healthcare. In addition, we also provide policy recommendations for other countries that use digital technology to combat COVID-19.

COVID-19 in the Age of Artificial Intelligence: A Comprehensive Review

The recent COVID-19 pandemic, which broke at the end of the year 2019 in Wuhan, China, has infected more than 98.52 million people by today (January 23, 2021) with over 2.11 million deaths across the globe. To combat the growing pandemic on urgent basis, there is need to design effective solutions using new techniques that could exploit recent technology, such as machine learning, deep learning, big data, artificial intelligence, Internet of Things, for identification and tracking of COVID-19 cases in near real time. These technologies have offered inexpensive and rapid solution for proper screening, analyzing, prediction and tracking of COVID-19 positive cases. In this paper, a detailed review of the role of AI as a decisive tool for prognosis, analyze, and tracking the COVID-19 cases is performed. We searched various databases including Google Scholar, IEEE Library, Scopus and Web of Science using a combination of different keywords consisting of COVID-19 and AI. We have identified various applications, where AI can help healthcare practitioners in the process of identification and monitoring of COVID-19 cases. A compact summary of the corona virus cases are first highlighted, followed by the application of AI. Finally, we conclude the paper by highlighting new research directions and discuss the research challenges. Even though scientists and researchers have gathered and exchanged sufficient knowledge over last couple of months, but this structured review also examined technological perspectives while encompassing the medical aspect to help the healthcare practitioners, policymakers, decision makers, policymakers, AI scientists and virologists to quell this infectious COVID-19 pandemic outbreak.

Deep Learning for Malignancy Risk Estimation of Pulmonary Nodules Detected at Low-Dose Screening CT

Background Accurate estimation of the malignancy risk of pulmonary nodules at chest CT is crucial for optimizing management in lung cancer screening. Purpose To develop and validate a deep learning (DL) algorithm for malignancy risk estimation of pulmonary nodules detected at screening CT. Materials and Methods In this retrospective study, the DL algorithm was developed with 16 077 nodules (1249 malignant) collected -between 2002 and 2004 from the National Lung Screening Trial. External validation was performed in the following three -cohorts -collected between 2004 and 2010 from the Danish Lung Cancer Screening Trial: a full cohort containing all 883 nodules (65 -malignant) and two cancer-enriched cohorts with size matching (175 nodules, 59 malignant) and without size matching (177 -nodules, 59 malignant) of benign nodules selected at random. Algorithm performance was measured by using the area under the receiver operating characteristic curve (AUC) and compared with that of the Pan-Canadian Early Detection of Lung Cancer (PanCan) model in the full cohort and a group of 11 clinicians composed of four thoracic radiologists, five radiology residents, and two pulmonologists in the cancer-enriched cohorts. Results The DL algorithm significantly outperformed the PanCan model in the full cohort (AUC, 0.93 [95% CI: 0.89, 0.96] vs 0.90 [95% CI: 0.86, 0.93]; P = .046). The algorithm performed comparably to thoracic radiologists in cancer-enriched cohorts with both random benign nodules (AUC, 0.96 [95% CI: 0.93, 0.99] vs 0.90 [95% CI: 0.81, 0.98]; P = .11) and size-matched benign nodules (AUC, 0.86 [95% CI: 0.80, 0.91] vs 0.82 [95% CI: 0.74, 0.89]; P = .26). Conclusion The deep learning algorithm showed excellent performance, comparable to thoracic radiologists, for malignancy risk estimation of pulmonary nodules detected at screening CT. This algorithm has the potential to provide reliable and reproducible malignancy risk scores for clinicians, which may help optimize management in lung cancer screening. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Tammemägi in this issue.

Artificial intelligence-based approaches for COVID-19 patient management

During the highly infectious pandemic of coronavirus disease 2019 (COVID-19), artificial intelligence (AI) has provided support in addressing challenges and accelerating achievements in controlling this public health crisis. It has been applied in fields varying from outbreak forecasting to patient management and drug/vaccine development. In this paper, we specifically review the current status of AI-based approaches for patient management. Limitations and challenges still exist, and further needs are highlighted.

Comparison of the CO-RADS and the RSNA chest CT classification system concerning sensitivity and reliability for the diagnosis of COVID-19 pneumonia

Background

The Radiological Society of North America (RSNA) recently published a chest CT classification system and Dutch Association for Radiology has announced Coronavirus disease 2019 (COVID-19) reporting and data system (CO-RADS) to provide guidelines to radiologists who interpret chest CT images of patients with suspected COVID-19 pneumonia. This study aimed to compare CO-RADS and RSNA classification with respect to their sensitivity and reliability for diagnosis of COVID-19 pneumonia.

Results

A retrospective study assessed consecutive CT chest imaging of 359 COVID-19-positive patients. Three experienced radiologists who were aware of the final diagnosis of all patients, independently categorized each patient according to CO-RADS and RSNA classification. RT-PCR test performed within one week of chest CT scan was used as a reference standard for calculating sensitivity of each system. Kappa statistics and intraclass correlation coefficient were used to assess reliability of each system. The study group included 359 patients (180 men, 179 women; mean age, 45 ± 16.9 years). Considering combination of CO-RADS 3, 4 and 5 and combination of typical and indeterminate RSNA categories as positive predictors for COVID-19 diagnosis, the overall sensitivity was the same for both classification systems (72.7%). Applying both systems in moderate and severe/critically ill patients resulted in a significant increase in sensitivity (94.7% and 97.8%, respectively). The overall inter-reviewer agreement was excellent for CO-RADS (κ = 0.801), and good for RSNA classification (κ = 0.781).

Conclusion

CO-RADS and RSNA chest CT classification systems are comparable in diagnosis of COVID-19 pneumonia with similar sensitivity and reliability.

Practical clinical and radiological models to diagnose COVID-19 based on a multicentric teleradiological emergency chest CT cohort

Our aim was to develop practical models built with simple clinical and radiological features to help diagnosing Coronavirus disease 2019 [COVID-19] in a real-life emergency cohort. To do so, 513 consecutive adult patients suspected of having COVID-19 from 15 emergency departments from 2020-03-13 to 2020-04-14 were included as long as chest CT-scans and real-time polymerase chain reaction (RT-PCR) results were available (244 [47.6%] with a positive RT-PCR). Immediately after their acquisition, the chest CTs were prospectively interpreted by on-call teleradiologists (OCTRs) and systematically reviewed within one week by another senior teleradiologist. Each OCTR reading was concluded using a 5-point scale: normal, non-infectious, infectious non-COVID-19, indeterminate and highly suspicious of COVID-19. The senior reading reported the lesions’ semiology, distribution, extent and differential diagnoses. After pre-filtering clinical and radiological features through univariate Chi-2, Fisher or Student t-tests (as appropriate), multivariate stepwise logistic regression (Step-LR) and classification tree (CART) models to predict a positive RT-PCR were trained on 412 patients, validated on an independent cohort of 101 patients and compared with the OCTR performances (295 and 71 with available clinical data, respectively) through area under the receiver operating characteristics curves (AUC). Regarding models elaborated on radiological variables alone, best performances were reached with the CART model (i.e., AUC = 0.92 [versus 0.88 for OCTR], sensitivity = 0.77, specificity = 0.94) while step-LR provided the highest AUC with clinical-radiological variables (AUC = 0.93 [versus 0.86 for OCTR], sensitivity = 0.82, specificity = 0.91). Hence, these two simple models, depending on the availability of clinical data, provided high performances to diagnose positive RT-PCR and could be used by any radiologist to support, modulate and communicate their conclusion in case of COVID-19 suspicion. Practically, using clinical and radiological variables (GGO, fever, presence of fibrotic bands, presence of diffuse lesions, predominant peripheral distribution) can accurately predict RT-PCR status.

Computed tomography findings of COVID-19 pneumonia in Intensive Care Unit-patients

BACKGROUND

In December 2019, a cluster of unknown etiology pneumonia cases occurred in Wuhan, China leading to identification of the responsible pathogen as SARS-coV-2. Since then, the coronavirus disease 2019 (COVID-19) has spread to the entire world. Computed Tomography (CT) is frequently used to assess severity and complications of COVID-19 pneumonia. The purpose of this study is to compare the CT patterns and clinical characteristics in intensive care unit (ICU) and non-ICU patients with COVID-19 pneumonia.

DESIGN AND METHODS

This retrospective study included 218 consecutive patients (136 males; 82 females; mean age 63±15 years) with laboratory-confirmed SARS-coV-2. Patients were categorized in two different groups: (a) ICU patients and (b) non-ICU inpatients. We assessed the type and extent of pulmonary opacities on chest CT exams and recorded the information on comorbidities and laboratory values for all patients.

RESULTS

Of the 218 patients, 23 (20 males: 3 females; mean age 60 years) required ICU admission, 195 (118 males: 77 females, mean age 64 years) were admitted to a clinical ward. Compared with non-ICU patients, ICU patients were predominantly males (60% versus 83% p=0.03), had more comorbidities, a positive CRP (p=0.04) and higher LDH values (p=0.008). ICU patients' chest CT demonstrated higher incidence of consolidation (p=0.03), mixed lesions (p=0.01), bilateral opacities (p<0.01) and overall greater lung involvement by consolidation (p=0.02) and GGO (p=0.001).

CONCLUSIONS

CT imaging features of ICU patients affected by COVID-19 are significantly different compared with non-ICU patients. Identification of CT features could assist in a stratification of the disease severity and supportive treatment.

Pulmonary Hypertension in Association with Lung Disease: Quantitative CT and Artificial Intelligence to the Rescue? State-of-the-Art Review

Accurate phenotyping of patients with pulmonary hypertension (PH) is an integral part of informing disease classification, treatment, and prognosis. The impact of lung disease on PH outcomes and response to treatment remains a challenging area with limited progress. Imaging with computed tomography (CT) plays an important role in patients with suspected PH when assessing for parenchymal lung disease, however, current assessments are limited by their semi-qualitative nature. Quantitative chest-CT (QCT) allows numerical quantification of lung parenchymal disease beyond subjective visual assessment. This has facilitated advances in radiological assessment and clinical correlation of a range of lung diseases including emphysema, interstitial lung disease, and coronavirus disease 2019 (COVID-19). Artificial Intelligence approaches have the potential to facilitate rapid quantitative assessments. Benefits of cross-sectional imaging include ease and speed of scan acquisition, repeatability and the potential for novel insights beyond visual assessment alone. Potential clinical benefits include improved phenotyping and prediction of treatment response and survival. Artificial intelligence approaches also have the potential to aid more focused study of pulmonary arterial hypertension (PAH) therapies by identifying more homogeneous subgroups of patients with lung disease. This state-of-the-art review summarizes recent QCT developments and potential applications in patients with PH with a focus on lung disease.

Comparing Visual Scoring of Lung Injury with a Quantifying AI-Based Scoring in Patients with COVID-19

OBJECTIVES

Fast diagnosis of Coronavirus Disease 2019 (COVID-19), and the detection of high-risk patients are crucial but challenging in the pandemic outbreak. The aim of this study was to evaluate if deep learning-based software correlates well with the generally accepted visual-based scoring for quantification of the lung injury to help radiologist in triage and monitoring of COVID-19 patients.

MATERIALS AND METHODS

In this retrospective study, the lobar analysis of lung opacities (% opacities) by means of a prototype deep learning artificial intelligence (AI)-based software was compared to visual scoring. The visual scoring system used five categories (0: 0%, 1: 0-5%, 2: 5-25%, 3: 25-50%, 4: 50-75% and 5: >75% involvement). The total visual lung injury was obtained by the sum of the estimated grade of involvement of each lobe and divided by five.

RESULTS

The dataset consisted of 182 consecutive confirmed COVID-19 positive patients with a median age of 65 ± 16 years, including 110 (60%) men and 72 (40%) women. There was a correlation coefficient of 0.89 (p < 0.001) between the visual and the AI-based estimates of the severity of lung injury.

CONCLUSION

The study indicates a very good correlation between the visual scoring and AI-based estimates of lung injury in COVID-19.

Application of Machine Learning in Diagnosis of COVID-19 Through X-Ray and CT Images: A Scoping Review

Coronavirus disease, first detected in late 2019 (COVID-19), has spread fast throughout the world, leading to high mortality. This condition can be diagnosed using RT-PCR technique on nasopharyngeal and throat swabs with sensitivity values ranging from 30 to 70%. However, chest CT scans and X-ray images have been reported to have sensitivity values of 98 and 69%, respectively. The application of machine learning methods on CT and X-ray images has facilitated the accurate diagnosis of COVID-19. In this study, we reviewed studies which used machine and deep learning methods on chest X-ray images and CT scans for COVID-19 diagnosis and compared their performance. The accuracy of these methods ranged from 76% to more than 99%, indicating the applicability of machine and deep learning methods in the clinical diagnosis of COVID-19.

Volume-of-Interest Aware Deep Neural Networks for Rapid Chest CT-Based COVID-19 Patient Risk Assessment

Since December 2019, the world has been devastated by the Coronavirus Disease 2019 (COVID-19) pandemic. Emergency Departments have been experiencing situations of urgency where clinical experts, without long experience and mature means in the fight against COVID-19, have to rapidly decide the most proper patient treatment. In this context, we introduce an artificially intelligent tool for effective and efficient Computed Tomography (CT)-based risk assessment to improve treatment and patient care. In this paper, we introduce a data-driven approach built on top of volume-of-interest aware deep neural networks for automatic COVID-19 patient risk assessment (discharged, hospitalized, intensive care unit) based on lung infection quantization through segmentation and, subsequently, CT classification. We tackle the high and varying dimensionality of the CT input by detecting and analyzing only a sub-volume of the CT, the Volume-of-Interest (VoI). Differently from recent strategies that consider infected CT slices without requiring any spatial coherency between them, or use the whole lung volume by applying abrupt and lossy volume down-sampling, we assess only the "most infected volume" composed of slices at its original spatial resolution. To achieve the above, we create, present and publish a new labeled and annotated CT dataset with 626 CT samples from COVID-19 patients. The comparison against such strategies proves the effectiveness of our VoI-based approach. We achieve remarkable performance on patient risk assessment evaluated on balanced data by reaching 88.88%, 89.77%, 94.73% and 88.88% accuracy, sensitivity, specificity and F1-score, respectively.

Artificial Intelligence Clinicians Can Use Chest Computed Tomography Technology to Automatically Diagnose Coronavirus Disease 2019 (COVID-19) Pneumonia and Enhance Low-Quality Images

Purpose

Nowadays, the number of patients with COVID-19 pneumonia worldwide is still increasing. The clinical diagnosis of COVID-19 pneumonia faces challenges, such as the difficulty to perform RT-PCR tests in real time, the lack of experienced radiologists, clinical low-quality images, and the similarity of imaging features of community-acquired pneumonia and COVID-19. Therefore, we proposed an artificial intelligence model GARCD that uses chest CT images to assist in the diagnosis of COVID-19 in real time. It can show better diagnostic performance even facing low-quality CT images.

Methods

We used 14,129 CT images from 104 patients. A total of 12,929 samples were used to build artificial intelligence models, and 1200 samples were used to test its performance. The image quality improvement module is based on the generative adversarial structure. It improves the quality of the input image under the joint drive of feature loss and content loss. The enhanced image is sent to the disease diagnosis model based on residual convolutional network. It automatically extracts the semantic features of the image and then infers the probability that the sample belongs to COVID-19. The ROC curve is used to evaluate the performance of the model.

Results

This model can effectively enhance the low-quality image and make the image that is difficult to be recognized become recognizable. The model proposed in this paper reached 97.8% AUC, 96.97% sensitivity and 91.16% specificity in an independent test set. ResNet, GADCD, CNN, and DenseNet achieved 80.9%, 97.3%, 70.7% and 85.7% AUC in the same test set, respectively. By comparing the performance with related works, it is proved that the model proposed has stronger clinical usability.

Conclusion

The method proposed can effectively assist doctors in real-time detection of suspected cases of COVID-19 pneumonia even faces unclear image. It can quickly isolate patients in a targeted manner, which is of positive significance for preventing the further spread of COVID-19 pneumonia.

Use of the COVID-19 Reporting and Data System (CO-RADS) classification and chest computed tomography involvement score (CT-IS) in COVID-19 pneumonia

Purpose

The increasing tendency of chest CT usage throughout the COVID-19 epidemic requires new tools and a systematic scheme for diagnosing and assessing the lung involvement in Coronavirus Disease 2019 (COVID-19). To investigate the use of the COVID-19 Reporting and Data System (CO-RADS) classification and chest CT Involvement Score (CT-IS) in COVID-19 pneumonia.

Material and methods

This retrospective study enrolled 280 hospitalized patients diagnosed with COVID-19 pneumonia in a tertiary hospital in Turkey. All patients underwent non-contrast CT chest imaging. Two radiologists interpreted all CT images according to CO-RADS classification without knowing the clinical features, laboratory findings. We used CT involvement score (CT-IS) for assessing chest CT images of COVID-19 patients. Also, we examined the relationship between CT-IS and clinical outcomes in COVID-19 patients.

Results

Of the patients, 111(39.6%) had positive real-time reverse transcriptase-polymerase chain reaction (RT-PCR) results. CO-RADS 5 group patients had statistically significant positive RT-PCR results than the other groups (P < 0.001). All of the CO-RADS 2 group patients (30) had negative RT-PCR results. The mean total CT-IS in CO-RADS 2 group was 3.4 ± 2.8. The mean total CT-IS in CO-RADS 5 group was 8.2 ± 4.7. Total CT-IS was statistically significantly different among CO-RADS groups (P < 0.001). The mean total CT-IS was statistically significantly different between survivors and patients died of COVID-19 pneumonia (P < 0.001).

Conclusions

CO-RADS is useful in detecting COVID-19 disease, even if RT-PCR testing is negative. CT-IS is also helpful as an imaging tool for evaluation of the severity and extent of COVID-19 pneumonia.

Machine learning models for predicting critical illness risk in hospitalized patients with COVID-19 pneumonia

Background

To develop machine learning classifiers at admission for predicting which patients with coronavirus disease 2019 (COVID-19) who will progress to critical illness.

Methods

A total of 158 patients with laboratory-confirmed COVID-19 admitted to three designated hospitals between December 31, 2019 and March 31, 2020 were retrospectively collected. 27 clinical and laboratory variables of COVID-19 patients were collected from the medical records. A total of 201 quantitative CT features of COVID-19 pneumonia were extracted by using an artificial intelligence software. The critically ill cases were defined according to the COVID-19 guidelines. The least absolute shrinkage and selection operator (LASSO) logistic regression was used to select the predictors of critical illness from clinical and radiological features, respectively. Accordingly, we developed clinical and radiological models using the following machine learning classifiers, including naive bayes (NB), linear regression (LR), random forest (RF), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), K-nearest neighbor (KNN), kernel support vector machine (k-SVM), and back propagation neural networks (BPNN). The combined model incorporating the selected clinical and radiological factors was also developed using the eight above-mentioned classifiers. The predictive efficiency of the models is validated using a 5-fold cross-validation method. The performance of the models was compared by the area under the receiver operating characteristic curve (AUC).

Results

The mean age of all patients was 58.9±13.9 years and 89 (56.3%) were males. 35 (22.2%) patients deteriorated to critical illness. After LASSO analysis, four clinical features including lymphocyte percentage, lactic dehydrogenase, neutrophil count, and D-dimer and four quantitative CT features were selected. The XGBoost-based clinical model yielded the highest AUC of 0.960 [95% confidence interval (CI): 0.913–1.000)]. The XGBoost-based radiological model achieved an AUC of 0.890 (95% CI: 0.757–1.000). However, the predictive efficacy of XGBoost-based combined model was very close to that of the XGBoost-based clinical model, with an AUC of 0.955 (95% CI: 0.906–1.000).

Conclusions

A XGBoost-based based clinical model on admission might be used as an effective tool to identify patients at high risk of critical illness.

Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT

The Coronavirus disease 2019 (COVID-19) presents open questions in how we clinically diagnose and assess disease course. Recently, chest computed tomography (CT) has shown utility for COVID-19 diagnosis. In this study, we developed Deep COVID DeteCT (DCD), a deep learning convolutional neural network (CNN) that uses the entire chest CT volume to automatically predict COVID-19 (COVID+) from non-COVID-19 (COVID-) pneumonia and normal controls. We discuss training strategies and differences in performance across 13 international institutions and 8 countries. The inclusion of non-China sites in training significantly improved classification performance with area under the curve (AUCs) and accuracies above 0.8 on most test sites. Furthermore, using available follow-up scans, we investigate methods to track patient disease course and predict prognosis.

An Automatic Approach for Individual HU-Based Characterization of Lungs in COVID-19 Patients

The ongoing COVID-19 pandemic currently involves millions of people worldwide. Radiology plays an important role in the diagnosis and management of patients, and chest computed tomography (CT) is the most widely used imaging modality. An automatic method to characterize the lungs of COVID-19 patients based on individually optimized Hounsfield unit (HU) thresholds was developed and implemented. Lungs were considered as composed of three components—aerated, intermediate, and consolidated. Three methods based on analytic fit (Gaussian) and maximum gradient search (using polynomial and original data fits) were implemented. The methods were applied to a population of 166 patients scanned during the first wave of the pandemic. Preliminarily, the impact of the inter-scanner variability of the HU-density calibration curve was investigated. Results showed that inter-scanner variability was negligible. The median values of individual thresholds th1 (between aerated and intermediate components) were −768, −780, and −798 HU for the three methods, respectively. A significantly lower median value for th2 (between intermediate and consolidated components) was found for the maximum gradient on the data (−34 HU) compared to the other two methods (−114 and −87 HU). The maximum gradient on the data method was applied to quantify the three components in our population—the aerated, intermediate, and consolidation components showed median values of 793 ± 499 cc, 914 ± 291 cc, and 126 ± 111 cc, respectively, while the median value of the first peak was −853 ± 56 HU.

Correlation between Chest CT Severity Scores and the Clinical Parameters of Adult Patients with COVID-19 Pneumonia

PURPOSE

Our aim is to correlate the clinical condition of patients with COVID-19 infection with the 25-point CT severity score by Chang et al. (devised for assessment of ARDS in patients with SARS in 2005).

MATERIALS AND METHODS

Data of consecutive symptomatic patients who were suspected to have COVID-19 infection and presented to our hospital were collected from March to April 2020. All patients underwent two consecutive RT-PCR tests and had a noncontrast HRCT scan done at presentation. From the original cohort of 1062 patients, 160 patients were excluded leaving a total number of 902 patients.

RESULTS

The mean age was 44.2 ± 11.9 years (85.3% males, 14.7% females). CT severity score was found to be positively correlated with lymphopenia, increased serum CRP, d-dimer, and ferritin levels (p < 0.0001). The oxygen requirements and length of hospital stay were increasing with the increase in scan severity.

CONCLUSION

The 25-point CT severity score correlates well with the COVID-19 clinical severity. Our data suggest that chest CT scoring system can aid in predicting COVID-19 disease outcome and significantly correlates with lab tests and oxygen requirements.

Lessons learned in transitioning to AI in the medical imaging of COVID-19

The coronavirus disease 2019 (COVID-19) pandemic has wreaked havoc across the world. It also created a need for the urgent development of efficacious predictive diagnostics, specifically, artificial intelligence (AI) methods applied to medical imaging. This has led to the convergence of experts from multiple disciplines to solve this global pandemic including clinicians, medical physicists, imaging scientists, computer scientists, and informatics experts to bring to bear the best of these fields for solving the challenges of the COVID-19 pandemic. However, such a convergence over a very brief period of time has had unintended consequences and created its own challenges. As part of Medical Imaging Data and Resource Center initiative, we discuss the lessons learned from career transitions across the three involved disciplines (radiology, medical imaging physics, and computer science) and draw recommendations based on these experiences by analyzing the challenges associated with each of the three associated transition types: (1) AI of non-imaging data to AI of medical imaging data, (2) medical imaging clinician to AI of medical imaging, and (3) AI of medical imaging to AI of COVID-19 imaging. The lessons learned from these career transitions and the diffusion of knowledge among them could be accomplished more effectively by recognizing their associated intricacies. These lessons learned in the transitioning to AI in the medical imaging of COVID-19 can inform and enhance future AI applications, making the whole of the transitions more than the sum of each discipline, for confronting an emergency like the COVID-19 pandemic or solving emerging problems in biomedicine.

What Can COVID-19 Teach Us about Using AI in Pandemics?

The COVID-19 pandemic put significant strain on societies and their resources, with the healthcare system and workers being particularly affected. Artificial Intelligence (AI) offers the unique possibility of improving the response to a pandemic as it emerges and evolves. Here, we utilize the WHO framework of a pandemic evolution to analyze the various AI applications. Specifically, we analyzed AI from the perspective of all five domains of the WHO pandemic response. To effectively review the current scattered literature, we organized a sample of relevant literature from various professional and popular resources. The article concludes with a consideration of AI’s weaknesses as key factors affecting AI in future pandemic preparedness and response.