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

Machine Learning Based Multi-Parameter Modeling for Prediction of Post-Inflammatory Lung Changes

Objectives: Prediction of lung function deficits following pulmonary infection is challenging and suffers from inaccuracy. We sought to develop machine-learning models for prediction of post-inflammatory lung changes based on COVID-19 recovery data. Methods: In the prospective CovILD study (n = 420 longitudinal observations from n = 140 COVID-19 survivors), data on lung function testing (LFT), chest CT including severity scoring by a human radiologist and density measurement by artificial intelligence, demography, and persistent symptoms were collected. This information was used to develop models of numeric readouts and abnormalities of LFT with four machine learning algorithms (Random Forest, gradient boosted machines, neural network, and support vector machines). Results: Reduced DLCO (diffusion capacity for carbon monoxide <80% of reference) was found in 94 (22%) observations. Those observations were modeled with a cross-validated accuracy of 82-85%, AUC of 0.87-0.9, and Cohen's κ of 0.45-0.5. No reliable models could be established for FEV1 or FVC. For DLCO as a continuous variable, three machine learning algorithms yielded meaningful models with cross-validated mean absolute errors of 11.6-12.5% and R2 of 0.26-0.34. CT-derived features such as opacity, high opacity, and CT severity score were among the most influential predictors of DLCO impairment. Conclusions: Multi-parameter machine learning trained with demographic, clinical, and artificial intelligence chest CT data reliably and reproducibly predicts LFT deficits and outperforms single markers of lung pathology and human radiologist's assessment. It may improve diagnostic and foster personalized treatment.

Artificial intelligence-based quantification of pulmonary HRCT (AIqpHRCT) for the evaluation of interstitial lung disease in patients with inflammatory rheumatic diseases

High-resolution computed tomography (HRCT) is important for diagnosing interstitial lung disease (ILD) in inflammatory rheumatic disease (IRD) patients. However, visual ILD assessment via HRCT often has high inter-reader variability. Artificial intelligence (AI)-based techniques for quantitative image analysis promise more accurate diagnostic and prognostic information. This study evaluated the reliability of artificial intelligence-based quantification of pulmonary HRCT (AIqpHRCT) in IRD-ILD patients and verified IRD-ILD quantification using AIqpHRCT in the clinical setting. Reproducibility of AIqpHRCT was verified for each typical HRCT pattern (ground-glass opacity [GGO], non-specific interstitial pneumonia [NSIP], usual interstitial pneumonia [UIP], granuloma). Additional, 50 HRCT datasets from 50 IRD-ILD patients using AIqpHRCT were analysed and correlated with clinical data and pulmonary lung function parameters. AIqpHRCT presented 100% agreement (coefficient of variation = 0.00%, intraclass correlation coefficient = 1.000) regarding the detection of the different HRCT pattern. Furthermore, AIqpHRCT data showed an increase of ILD from 10.7 ± 28.3% (median = 1.3%) in GGO to 18.9 ± 12.4% (median = 18.0%) in UIP pattern. The extent of fibrosis negatively correlated with FVC (ρ=-0.501), TLC (ρ=-0.622), and DLCO (ρ=-0.693) (p < 0.001). GGO measured by AIqpHRCT also significant negatively correlated with DLCO (ρ=-0.699), TLC (ρ=-0.580) and FVC (ρ=-0.423). For the first time, the study demonstrates that AIpqHRCT provides a highly reliable method for quantifying lung parenchymal changes in HRCT images of IRD-ILD patients. Further, the AIqpHRCT method revealed significant correlations between the extent of ILD and lung function parameters. This highlights the potential of AIpqHRCT in enhancing the accuracy of ILD diagnosis and prognosis in clinical settings, ultimately improving patient management and outcomes.

Evolving and Novel Applications of Artificial Intelligence in Thoracic Imaging

The advent of artificial intelligence (AI) is revolutionizing medicine, particularly radiology. With the development of newer models, AI applications are demonstrating improved performance and versatile utility in the clinical setting. Thoracic imaging is an area of profound interest, given the prevalence of chest imaging and the significant health implications of thoracic diseases. This review aims to highlight the promising applications of AI within thoracic imaging. It examines the role of AI, including its contributions to improving diagnostic evaluation and interpretation, enhancing workflow, and aiding in invasive procedures. Next, it further highlights the current challenges and limitations faced by AI, such as the necessity of 'big data', ethical and legal considerations, and bias in representation. Lastly, it explores the potential directions for the application of AI in thoracic radiology.

A multimodal integration pipeline for accurate diagnosis, pathogen identification, and prognosis prediction of pulmonary infections

Pulmonary infections pose formidable challenges in clinical settings with high mortality rates across all age groups worldwide. Accurate diagnosis and early intervention are crucial to improve patient outcomes. Artificial intelligence (AI) has the capability to mine imaging features specific to different pathogens and fuse multimodal features to reach a synergistic diagnosis, enabling more precise investigation and individualized clinical management. In this study, we successfully developed a multimodal integration (MMI) pipeline to differentiate among bacterial, fungal, and viral pneumonia and pulmonary tuberculosis based on a real-world dataset of 24,107 patients. The area under the curve (AUC) of the MMI system comprising clinical text and computed tomography (CT) image scans yielded 0.910 (95% confidence interval [CI]: 0.904-0.916) and 0.887 (95% CI: 0.867-0.909) in the internal and external testing datasets respectively, which were comparable to those of experienced physicians. Furthermore, the MMI system was utilized to rapidly differentiate between viral subtypes with a mean AUC of 0.822 (95% CI: 0.805-0.837) and bacterial subtypes with a mean AUC of 0.803 (95% CI: 0.775-0.830). Here, the MMI system harbors the potential to guide tailored medication recommendations, thus mitigating the risk of antibiotic misuse. Additionally, the integration of multimodal factors in the AI-driven system also provided an evident advantage in predicting risks of developing critical illness, contributing to more informed clinical decision-making. To revolutionize medical care, embracing multimodal AI tools in pulmonary infections will pave the way to further facilitate early intervention and precise management in the foreseeable future.

Estimating lung function from computed tomography at the patient and lobe level using machine learning

BACKGROUND

Automated estimation of Pulmonary function test (PFT) results from Computed Tomography (CT) could advance the use of CT in screening, diagnosis, and staging of restrictive pulmonary diseases. Estimating lung function per lobe, which cannot be done with PFTs, would be helpful for risk assessment for pulmonary resection surgery and bronchoscopic lung volume reduction.

PURPOSE

To automatically estimate PFT results from CT and furthermore disentangle the individual contribution of pulmonary lobes to a patient's lung function.

METHODS

We propose I3Dr, a deep learning architecture for estimating global measures from an image that can also estimate the contributions of individual parts of the image to this global measure. We apply it to estimate the separate contributions of each pulmonary lobe to a patient's total lung function from CT, while requiring only CT scans and patient level lung function measurements for training. I3Dr consists of a lobe-level and a patient-level model. The lobe-level model extracts all anatomical pulmonary lobes from a CT scan and processes them in parallel to produce lobe level lung function estimates that sum up to a patient level estimate. The patient-level model directly estimates patient level lung function from a CT scan and is used to re-scale the output of the lobe-level model to increase performance. After demonstrating the viability of the proposed approach, the I3Dr model is trained and evaluated for PFT result estimation using a large data set of 8 433 CT volumes for training, 1 775 CT volumes for validation, and 1 873 CT volumes for testing.

RESULTS

First, we demonstrate the viability of our approach by showing that a model trained with a collection of digit images to estimate their sum implicitly learns to assign correct values to individual digits. Next, we show that our models can estimate lobe-level quantities, such as COVID-19 severity scores, pulmonary volume (PV), and functional pulmonary volume (FPV) from CT while only provided with patient-level quantities during training. Lastly, we train and evaluate models for producing spirometry and diffusion capacity of carbon mono-oxide (DLCO) estimates at the patient and lobe level. For producing Forced Expiratory Volume in one second (FEV1), Forced Vital Capacity (FVC), and DLCO estimates, I3Dr obtains mean absolute errors (MAE) of 0.377 L, 0.297 L, and 2.800 mL/min/mm Hg respectively. We release the resulting algorithms for lung function estimation to the research community at https://grand-challenge.org/algorithms/lobe-wise-lung-function-estimation/ CONCLUSIONS: I3Dr can estimate global measures from an image, as well as the contributions of individual parts of the image to this global measure. It offers a promising approach for estimating PFT results from CT scans and disentangling the individual contribution of pulmonary lobes to a patient's lung function. The findings presented in this work may advance the use of CT in screening, diagnosis, and staging of restrictive pulmonary diseases as well as in risk assessment for pulmonary resection surgery and bronchoscopic lung volume reduction.

CoSev: Data-Driven Optimizations for COVID-19 Severity Assessment in Low-Sample Regimes

Given the pronounced impact COVID-19 continues to have on society-infecting 700 million reported individuals and causing 6.96 million deaths-many deep learning works have recently focused on the virus's diagnosis. However, assessing severity has remained an open and challenging problem due to a lack of large datasets, the large dimensionality of images for which to find weights, and the compute limitations of modern graphics processing units (GPUs). In this paper, a new, iterative application of transfer learning is demonstrated on the understudied field of 3D CT scans for COVID-19 severity analysis. This methodology allows for enhanced performance on the MosMed Dataset, which is a small and challenging dataset containing 1130 images of patients for five levels of COVID-19 severity (Zero, Mild, Moderate, Severe, and Critical). Specifically, given the large dimensionality of the input images, we create several custom shallow convolutional neural network (CNN) architectures and iteratively refine and optimize them, paying attention to learning rates, layer types, normalization types, filter sizes, dropout values, and more. After a preliminary architecture design, the models are systematically trained on a simplified version of the dataset-building models for two-class, then three-class, then four-class, and finally five-class classification. The simplified problem structure allows the model to start learning preliminary features, which can then be further modified for more difficult classification tasks. Our final model CoSev boosts classification accuracies from below 60% at first to 81.57% with the optimizations, reaching similar performance to the state-of-the-art on the dataset, with much simpler setup procedures. In addition to COVID-19 severity diagnosis, the explored methodology can be applied to general image-based disease detection. Overall, this work highlights innovative methodologies that advance current computer vision practices for high-dimension, low-sample data as well as the practicality of data-driven machine learning and the importance of feature design for training, which can then be implemented for improvements in clinical practices.

Inter-observer Variability in the Analysis of CO-RADS Classification for COVID-19 Patients

During the early stages of the pandemic, computed tomography (CT) of the chest, along with serological and clinical data, was frequently utilized in diagnosing COVID-19, particularly in regions facing challenges such as shortages of PCR kits. In these circumstances, CT scans played a crucial role in diagnosing COVID-19 and guiding patient management. The COVID-19 Reporting and Data System (CO-RADS) was established as a standardized reporting system for cases of COVID-19 pneumonia. Its implementation necessitates a high level of agreement among observers to prevent any potential confusion. This study aimed to assess the inter-observer agreement between physicians from different specialties with variable levels of experience in their CO-RADS scoring of CT chests for confirmed COVID-19 patients, and to assess the feasibility of applying this reporting system to those having little experience with it. All chest CT images of patients with positive RT-PCR tests for COVID-19 were retrospectively reviewed by seven observers. The observers were divided into three groups according to their type of specialty (three radiologists, three house officers, and one pulmonologist). The observers assessed each image and categorized the patients into five CO-RADS groups. A total of 630 participants were included in this study. The inter-observer agreement was almost perfect among the radiologists, substantial among a pulmonologist and the house officers, and moderate-to-substantial among the radiologists, the pulmonologist, and the house officers. There was substantial to almost perfect inter-observer agreement when reporting using the CO-RADS among observers with different experience levels. Although the inter-observer variability among the radiologists was high, it decreased compared to the pulmonologist and house officers. Radiologists, house officers, and pulmonologists applying the CO-RADS can accurately and promptly identify typical CT imaging features of lung involvement in COVID-19.

Segmentation of lung lobes and lesions in chest CT for the classification of COVID-19 severity

To precisely determine the severity of COVID-19-related pneumonia, computed tomography (CT) is an imaging modality beneficial for patient monitoring and therapy planning. Thus, we aimed to develop a deep learning-based image segmentation model to automatically assess lung lesions related to COVID-19 infection and calculate the total severity score (TSS). The entire dataset consisted of 124 COVID-19 patients acquired from Chulabhorn Hospital, divided into 28 cases without lung lesions and 96 cases with lung lesions categorized severity by radiologists regarding TSS. The model used a 3D-UNet along with DenseNet and ResNet models that had already been trained to separate the lobes of the lungs and figure out the percentage of lung involvement due to COVID-19 infection. It also used the Dice similarity coefficient (DSC) to measure TSS. Our final model, consisting of 3D-UNet integrated with DenseNet169, achieved segmentation of lung lobes and lesions with the Dice similarity coefficients of 91.52% and 76.89%, respectively. The calculated TSS values were similar to those evaluated by radiologists, with an R2 of 0.842. The correlation between the ground-truth TSS and model prediction was greater than that of the radiologist, which was 0.890 and 0.709, respectively.

COVID-19 screening in low resource settings using artificial intelligence for chest radiographs and point-of-care blood tests

Artificial intelligence (AI) systems for detection of COVID-19 using chest X-Ray (CXR) imaging and point-of-care blood tests were applied to data from four low resource African settings. The performance of these systems to detect COVID-19 using various input data was analysed and compared with antigen-based rapid diagnostic tests. Participants were tested using the gold standard of RT-PCR test (nasopharyngeal swab) to determine whether they were infected with SARS-CoV-2. A total of 3737 (260 RT-PCR positive) participants were included. In our cohort, AI for CXR images was a poor predictor of COVID-19 (AUC = 0.60), since the majority of positive cases had mild symptoms and no visible pneumonia in the lungs. AI systems using differential white blood cell counts (WBC), or a combination of WBC and C-Reactive Protein (CRP) both achieved an AUC of 0.74 with a suggested optimal cut-off point at 83% sensitivity and 63% specificity. The antigen-RDT tests in this trial obtained 65% sensitivity at 98% specificity. This study is the first to validate AI tools for COVID-19 detection in an African setting. It demonstrates that screening for COVID-19 using AI with point-of-care blood tests is feasible and can operate at a higher sensitivity level than antigen testing.

A Systematic Review on Deep Structured Learning for COVID-19 Screening Using Chest CT from 2020 to 2022

The emergence of the COVID-19 pandemic in Wuhan in 2019 led to the discovery of a novel coronavirus. The World Health Organization (WHO) designated it as a global pandemic on 11 March 2020 due to its rapid and widespread transmission. Its impact has had profound implications, particularly in the realm of public health. Extensive scientific endeavors have been directed towards devising effective treatment strategies and vaccines. Within the healthcare and medical imaging domain, the application of artificial intelligence (AI) has brought significant advantages. This study delves into peer-reviewed research articles spanning the years 2020 to 2022, focusing on AI-driven methodologies for the analysis and screening of COVID-19 through chest CT scan data. We assess the efficacy of deep learning algorithms in facilitating decision making processes. Our exploration encompasses various facets, including data collection, systematic contributions, emerging techniques, and encountered challenges. However, the comparison of outcomes between 2020 and 2022 proves intricate due to shifts in dataset magnitudes over time. The initiatives aimed at developing AI-powered tools for the detection, localization, and segmentation of COVID-19 cases are primarily centered on educational and training contexts. We deliberate on their merits and constraints, particularly in the context of necessitating cross-population train/test models. Our analysis encompassed a review of 231 research publications, bolstered by a meta-analysis employing search keywords (COVID-19 OR Coronavirus) AND chest CT AND (deep learning OR artificial intelligence OR medical imaging) on both the PubMed Central Repository and Web of Science platforms.

Development and validation of a hybrid deep learning-machine learning approach for severity assessment of COVID-19 and other pneumonias

The Coronavirus Disease 2019 (COVID-19) is transitioning into the endemic phase. Nonetheless, it is crucial to remain mindful that pandemics related to infectious respiratory diseases (IRDs) can emerge unpredictably. Therefore, we aimed to develop and validate a severity assessment model for IRDs, including COVID-19, influenza, and novel influenza, using CT images on a multi-centre data set. Of the 805 COVID-19 patients collected from a single centre, 649 were used for training and 156 were used for internal validation (D1). Additionally, three external validation sets were obtained from 7 cohorts: 1138 patients with COVID-19 (D2), and 233 patients with influenza and novel influenza (D3). A hybrid model, referred to as Hybrid-DDM, was constructed by combining two deep learning models and a machine learning model. Across datasets D1, D2, and D3, the Hybrid-DDM exhibited significantly improved performance compared to the baseline model. The areas under the receiver operating curves (AUCs) were 0.830 versus 0.767 (p = 0.036) in D1, 0.801 versus 0.753 (p < 0.001) in D2, and 0.774 versus 0.668 (p < 0.001) in D3. This study indicates that the Hybrid-DDM model, trained using COVID-19 patient data, is effective and can also be applicable to patients with other types of viral pneumonia.

Artificial intelligence and machine learning responses to COVID-19 related inquiries

Researchers and scientists can use computational-based models to turn linked data into useful information, aiding in disease diagnosis, examination, and viral containment due to recent artificial intelligence and machine learning breakthroughs. In this paper, we extensively study the role of artificial intelligence and machine learning in delivering efficient responses to the COVID-19 pandemic almost four years after its start. In this regard, we examine a large number of critical studies conducted by various academic and research communities from multiple disciplines, as well as practical implementations of artificial intelligence algorithms that suggest potential solutions in investigating different COVID-19 decision-making scenarios. We identify numerous areas where artificial intelligence and machine learning can impact this context, including diagnosis (using chest X-ray imaging and CT imaging), severity, tracking, treatment, and the drug industry. Furthermore, we analyse the dilemma's limits, restrictions, and hazards.

Gallium-68 labelled RGD PET/CT imaging of endothelial activation in COVID-19 patients

In coronavirus disease 2019 (COVID-19), endothelial cells play a central role and an inadequate response is associated with vascular complications. PET imaging with gallium-68 labelled RGD-peptide (⁶⁸Ga-RGD) targets α_vβ₃ integrin expression which allows quantification of endothelial activation. In this single-center, prospective observational study, we included ten hospitalized patients with COVID-19 between October 2020 and January 2021. Patients underwent ⁶⁸Ga-RGD PET/CT followed by iodine mapping of lung parenchyma. CT-based segmentation of lung parenchyma, carotid arteries and myocardium was used to quantify tracer uptake by calculating standardized uptake values (SUV). Five non-COVID-19 patients were used as reference. The study population was 68.5 (IQR 52.0-74.5) years old, with median oxygen need of 3 l/min (IQR 0.9-4.0). ⁶⁸Ga-RGD uptake quantified as SUV ± SD was increased in lungs (0.99 ± 0.32 vs. 0.45 ± 0.18, p < 0.01) and myocardium (3.44 ± 1.59 vs. 0.65 ± 0.22, p < 0.01) of COVID-19 patients compared to reference but not in the carotid arteries. Iodine maps showed local variations in parenchymal perfusion but no correlation with SUV. In conclusion, using ⁶⁸Ga-RGD PET/CT in COVID-19 patients admitted with respiratory symptoms, we demonstrated increased endothelial activation in the lung parenchyma and myocardium. Our findings indicate the involvement of increased and localized endothelial cell activation in the cardiopulmonary system in COVID-19 patients.Trail registration: NCT04596943.

Artificial intelligence for reducing the radiation burden of medical imaging for the diagnosis of coronavirus disease

Medical imaging has been intensively employed in screening, diagnosis and monitoring during the COVID-19 pandemic. With the improvement of RT-PCR and rapid inspection technologies, the diagnostic references have shifted. Current recommendations tend to limit the application of medical imaging in the acute setting. Nevertheless, efficient and complementary values of medical imaging have been recognized at the beginning of the pandemic when facing unknown infectious diseases and a lack of sufficient diagnostic tools. Optimizing medical imaging for pandemics may still have encouraging implications for future public health, especially for long-lasting post-COVID-19 syndrome theranostics. A critical concern for the application of medical imaging is the increased radiation burden, particularly when medical imaging is used for screening and rapid containment purposes. Emerging artificial intelligence (AI) technology provides the opportunity to reduce the radiation burden while maintaining diagnostic quality. This review summarizes the current AI research on dose reduction for medical imaging, and the retrospective identification of their potential in COVID-19 may still have positive implications for future public health.

Dense regression activation maps for lesion segmentation in CT scans of COVID-19 patients

Automatic lesion segmentation on thoracic CT enables rapid quantitative analysis of lung involvement in COVID-19 infections. However, obtaining a large amount of voxel-level annotations for training segmentation networks is prohibitively expensive. Therefore, we propose a weakly-supervised segmentation method based on dense regression activation maps (dRAMs). Most weakly-supervised segmentation approaches exploit class activation maps (CAMs) to localize objects. However, because CAMs were trained for classification, they do not align precisely with the object segmentations. Instead, we produce high-resolution activation maps using dense features from a segmentation network that was trained to estimate a per-lobe lesion percentage. In this way, the network can exploit knowledge regarding the required lesion volume. In addition, we propose an attention neural network module to refine dRAMs, optimized together with the main regression task. We evaluated our algorithm on 90 subjects. Results show our method achieved 70.2% Dice coefficient, substantially outperforming the CAM-based baseline at 48.6%. We published our source code at https://github.com/DIAGNijmegen/bodyct-dram.

Quantification of pulmonary involvement in COVID-19 pneumonia: an upgrade of the LungQuant software for lung CT segmentation

Computed tomography (CT) scans are used to evaluate the severity of lung involvement in patients affected by COVID-19 pneumonia. Here, we present an improved version of the LungQuant automatic segmentation software (LungQuant v2), which implements a cascade of three deep neural networks (DNNs) to segment the lungs and the lung lesions associated with COVID-19 pneumonia. The first network (BB-net) defines a bounding box enclosing the lungs, the second one (U-net 1 ) outputs the mask of the lungs, and the final one (U-net 2 ) generates the mask of the COVID-19 lesions. With respect to the previous version (LungQuant v1), three main improvements are introduced: the BB-net, a new term in the loss function in the U-net for lesion segmentation and a post-processing procedure to separate the right and left lungs. The three DNNs were optimized, trained and tested on publicly available CT scans. We evaluated the system segmentation capability on an independent test set consisting of ten fully annotated CT scans, the COVID-19-CT-Seg benchmark dataset. The test performances are reported by means of the volumetric dice similarity coefficient (vDSC) and the surface dice similarity coefficient (sDSC) between the reference and the segmented objects. LungQuant v2 achieves a vDSC (sDSC) equal to 0.96 ± 0.01 (0.97 ± 0.01) and 0.69 ± 0.08 (0.83 ± 0.07) for the lung and lesion segmentations, respectively. The output of the segmentation software was then used to assess the percentage of infected lungs, obtaining a Mean Absolute Error (MAE) equal to 2%.

Imaging-derived biomarkers in Asthma: Current status and future perspectives

Asthma is a common disorder affecting around 315 million individuals worldwide. The heterogeneity of asthma is becoming increasingly important in the era of personalized treatment and response assessment. Several radiological imaging modalities are available in asthma including chest x-ray, computed tomography (CT) and magnetic resonance imaging (MRI) scanning. In addition to qualitative imaging, quantitative imaging could play an important role in asthma imaging to identify phenotypes with distinct disease course and response to therapy, including biologics. MRI in asthma is mainly performed in research settings given cost, technical challenges, and there is a need for standardization. Imaging analysis applications of artificial intelligence (AI) to subclassify asthma using image analysis have demonstrated initial feasibility, though additional work is necessary to inform the role of AI in clinical practice.

Computed tomography severity score as a predictor of disease severity and mortality in COVID-19 patients: A systematic review and meta-analysis

BACKGROUND

Prediction of outcomes in severe COVID-19 patients using chest computed tomography severity score (CTSS) may enable more effective clinical management and early, timely ICU admission. We conducted a systematic review and meta-analysis to determine the predictive accuracy of the CTSS for disease severity and mortality in severe COVID-19 subjects.

METHODS

The electronic databases PubMed, Google Scholar, Web of Science, and the Cochrane Library were searched to find eligible studies that investigated the impact of CTSS on disease severity and mortality in COVID-19 patients between 7 January 2020 and 15 June 2021. Two independent authors looked into the risk of bias using the Quality in Prognosis Studies (QUIPS) tool.

RESULTS

Seventeen studies involving 2788 patients reported the predictive value of CTSS for disease severity. The pooled sensitivity, specificity, and summary area under the curve (sAUC) of CTSS were 0.85 (95% CI 0.78-0.90, I² =83), 0.86 (95% CI 0.76-0.92, I² =96) and 0.91 (95% CI 0.89-0.94), respectively. Six studies involving 1403 patients reported the predictive values of CTSS for COVID-19 mortality. The pooled sensitivity, specificity, and sAUC of CTSS were 0.77 (95% CI 0.69-0.83, I² = 41), 0.79 (95% CI 0.72-0.85, I² = 88), and 0.84 (95% CI 0.81-0.87), respectively.

DISCUSSION

Early prediction of prognosis is needed to deliver the better care to patients and stratify them as soon as possible. Because different CTSS thresholds have been reported in various studies, clinicians are still determining whether CTSS thresholds should be used to define disease severity and predict prognosis.

CONCLUSION

Early prediction of prognosis is needed to deliver optimal care and timely stratification of patients.  CTSS has strong discriminating power for the prediction of disease severity and mortality in patients with COVID-19.

Is the generalizability of a developed artificial intelligence algorithm for COVID-19 on chest CT sufficient for clinical use? Results from the International Consortium for COVID-19 Imaging AI (ICOVAI)

OBJECTIVES

Only few published artificial intelligence (AI) studies for COVID-19 imaging have been externally validated. Assessing the generalizability of developed models is essential, especially when considering clinical implementation. We report the development of the International Consortium for COVID-19 Imaging AI (ICOVAI) model and perform independent external validation.

METHODS

The ICOVAI model was developed using multicenter data (n = 1286 CT scans) to quantify disease extent and assess COVID-19 likelihood using the COVID-19 Reporting and Data System (CO-RADS). A ResUNet model was modified to automatically delineate lung contours and infectious lung opacities on CT scans, after which a random forest predicted the CO-RADS score. After internal testing, the model was externally validated on a multicenter dataset (n = 400) by independent researchers. CO-RADS classification performance was calculated using linearly weighted Cohen's kappa and segmentation performance using Dice Similarity Coefficient (DSC).

RESULTS

Regarding internal versus external testing, segmentation performance of lung contours was equally excellent (DSC = 0.97 vs. DSC = 0.97, p = 0.97). Lung opacities segmentation performance was adequate internally (DSC = 0.76), but significantly worse on external validation (DSC = 0.59, p < 0.0001). For CO-RADS classification, agreement with radiologists on the internal set was substantial (kappa = 0.78), but significantly lower on the external set (kappa = 0.62, p < 0.0001).

CONCLUSION

In this multicenter study, a model developed for CO-RADS score prediction and quantification of COVID-19 disease extent was found to have a significant reduction in performance on independent external validation versus internal testing. The limited reproducibility of the model restricted its potential for clinical use. The study demonstrates the importance of independent external validation of AI models.

KEY POINTS

• The ICOVAI model for prediction of CO-RADS and quantification of disease extent on chest CT of COVID-19 patients was developed using a large sample of multicenter data. • There was substantial performance on internal testing; however, performance was significantly reduced on external validation, performed by independent researchers. The limited generalizability of the model restricts its potential for clinical use. • Results of AI models for COVID-19 imaging on internal tests may not generalize well to external data, demonstrating the importance of independent external validation.

Additional value of chest CT AI-based quantification of lung involvement in predicting death and ICU admission for COVID-19 patients

Objectives

We evaluated the contribution of lung lesion quantification on chest CT using a clinical Artificial Intelligence (AI) software in predicting death and intensive care units (ICU) admission for COVID-19 patients.

Methods

For 349 patients with positive COVID-19-PCR test that underwent a chest CT scan at admittance or during hospitalization, we applied the AI for lung and lung lesion segmentation to obtain lesion volume (LV), and LV/Total Lung Volume (TLV) ratio. ROC analysis was used to extract the best CT criterion in predicting death and ICU admission. Two prognostic models using multivariate logistic regressions were constructed to predict each outcome and were compared using AUC values. The first model ("Clinical") was based on patients' characteristics and clinical symptoms only. The second model ("Clinical+LV/TLV") included also the best CT criterion.

Results

LV/TLV ratio demonstrated best performance for both outcomes; AUC of 67.8% (95% CI: 59.5 - 76.1) and 81.1% (95% CI: 75.7 - 86.5) respectively. Regarding death prediction, AUC values were 76.2% (95% CI: 69.9 - 82.6) and 79.9% (95%IC: 74.4 - 85.5) for the "Clinical" and the "Clinical+LV/TLV" models respectively, showing significant performance increase (+ 3.7%; p-value<0.001) when adding LV/TLV ratio. Similarly, for ICU admission prediction, AUC values were 74.9% (IC 95%: 69.2 - 80.6) and 84.8% (IC 95%: 80.4 - 89.2) respectively corresponding to significant performance increase (+ 10%: p-value<0.001).

Conclusions

Using a clinical AI software to quantify the COVID-19 lung involvement on chest CT, combined with clinical variables, allows better prediction of death and ICU admission.

Cov-TransNet: Dual branch fusion network with transformer for COVID-19 infection segmentation

Segmentation of COVID-19 infection is a challenging task due to the blurred boundaries and low contrast between the infected and the non-infected areas in COVID-19 CT images, especially for small infection regions. COV-TransNet is presented to achieve high-precision segmentation of COVID-19 infection regions in this paper. The proposed segmentation network is composed of the auxiliary branch and the backbone branch. The auxiliary branch network adopts transformer to provide global information, helping the convolution layers in backbone branch to learn specific local features better. A multi-scale feature attention module is introduced to capture contextual information and adaptively enhance feature representations. Specially, a high internal resolution is maintained during the attention calculation process. Moreover, feature activation module can effectively reduce the loss of valid information during sampling. The proposed network can take full advantage of different depth and multi-scale features to achieve high sensitivity for identifying lesions of varied sizes and locations. We experiment on several datasets of the COVID-19 lesion segmentation task, including COVID-19-CT-Seg, UESTC-COVID-19, MosMedData and COVID-19-MedSeg. Comprehensive results demonstrate that COV-TransNet outperforms the existing state-of-the-art segmentation methods and achieves better segmentation performance for multi-scale lesions.

An Intelligent Sensor Based Decision Support System for Diagnosing Pulmonary Ailment through Standardized Chest X-ray Scans

Academics and the health community are paying much attention to developing smart remote patient monitoring, sensors, and healthcare technology. For the analysis of medical scans, various studies integrate sophisticated deep learning strategies. A smart monitoring system is needed as a proactive diagnostic solution that may be employed in an epidemiological scenario such as COVID-19. Consequently, this work offers an intelligent medicare system that is an IoT-empowered, deep learning-based decision support system (DSS) for the automated detection and categorization of infectious diseases (COVID-19 and pneumothorax). The proposed DSS system was evaluated using three independent standard-based chest X-ray scans. The suggested DSS predictor has been used to identify and classify areas on whole X-ray scans with abnormalities thought to be attributable to COVID-19, reaching an identification and classification accuracy rate of 89.58% for normal images and 89.13% for COVID-19 and pneumothorax. With the suggested DSS system, a judgment depending on individual chest X-ray scans may be made in approximately 0.01 s. As a result, the DSS system described in this study can forecast at a pace of 95 frames per second (FPS) for both models, which is near to real-time.

Discrepancy between biomarkers of lung injury and 1-year mortality in COVID-19

BACKGROUND

COVID-19 global pandemic started in late 2019 with the first wave. In this cross-sectional and observational study, we evaluated the associations between the biomarkers, COVID-19 pneumonia severity and 1-year mortality.

METHODS

A sample of 276 polymerase chain reaction (PCR)-positive patients for SARS-CoV-2 was included. Computerized tomography severity score (CT-SS) was used to assess the severity of COVID-19 pneumonia in 222 cases. Multivariate analyses were performed to find the predictors of CT-SS, severe CT-SS (≥20) and 1-year mortality. Biomarkers of ferritin, high-sensitive C-reactive protein (CRP), lactate dehydrogenase (LDH), cardiac troponin (cTn), neutrophil-to-lymphocyte ratio (NLR), uric acid (UA) and d-dimer were routinely measured.

RESULTS

Severe CT-SS (>20) was observed in 86 (31.2%) cases. Mortality was observed in 75 (27.2%) patients at 1 year. LDH displayed the highest predictive accuracy for severe CT-SS (AUC 0.741, sensitivity = 81% and specificity = 68%, cut-off value: 360 mg/dl). Linear regression analysis displayed that LDH predicted CT-SS [B = 11 (95% CI for B = 5-17, p < .001)]. Age was the most significant parameter that was associated with severe CT-SS (OR 0.96, 95% CI 0.92-0.99, p = .015). d-dimer was the only biomarker that predicted with 1-year mortality (OR 1.62, 95% CI 1.08-2.42, p = .020).

CONCLUSION

LDH is a sensitive and specific biomarker to determine patients with severe lung injury in COVID-19. d-dimer is the only biomarker that predicts 1-year mortality. Neither LDH nor CT-SS is associated with 1-year mortality.

Automated Lung Segmentation from Computed Tomography Images of Normal and COVID-19 Pneumonia Patients

Background

Automated image segmentation is an essential step in quantitative image analysis. This study assesses the performance of a deep learning-based model for lung segmentation from computed tomography (CT) images of normal and COVID-19 patients.

Methods

A descriptive-analytical study was conducted from December 2020 to April 2021 on the CT images of patients from various educational hospitals affiliated with Mashhad University of Medical Sciences (Mashhad, Iran). Of the selected images and corresponding lung masks of 1,200 confirmed COVID-19 patients, 1,080 were used to train a residual neural network. The performance of the residual network (ResNet) model was evaluated on two distinct external test datasets, namely the remaining 120 COVID-19 and 120 normal patients. Different evaluation metrics such as Dice similarity coefficient (DSC), mean absolute error (MAE), relative mean Hounsfield unit (HU) difference, and relative volume difference were calculated to assess the accuracy of the predicted lung masks. The Mann-Whitney U test was used to assess the difference between the corresponding values in the normal and COVID-19 patients. P<0.05 was considered statistically significant.

Results

The ResNet model achieved a DSC of 0.980 and 0.971 and a relative mean HU difference of -2.679% and -4.403% for the normal and COVID-19 patients, respectively. Comparable performance in lung segmentation of normal and COVID-19 patients indicated the model's accuracy for identifying lung tissue in the presence of COVID-19-associated infections. Although a slightly better performance was observed in normal patients.

Conclusion

The ResNet model provides an accurate and reliable automated lung segmentation of COVID-19 infected lung tissue.A preprint version of this article was published on arXiv before formal peer review (https://arxiv.org/abs/2104.02042).

Automatized lung disease quantification in patients with COVID-19 as a predictive tool to assess hospitalization severity

The pandemic of COVID-19 led to a dramatic situation in hospitals, where staff had to deal with a huge number of patients in respiratory distress. To alleviate the workload of radiologists, we implemented an artificial intelligence (AI) - based analysis named CACOVID-CT, to automatically assess disease severity on chest CT scans obtained from those patients. We retrospectively studied CT scans obtained from 476 patients admitted at the University Hospital of Liege with a COVID-19 disease. We quantified the percentage of COVID-19 affected lung area (% AA) and the CT severity score (total CT-SS). These quantitative measurements were used to investigate the overall prognosis and patient outcome: hospital length of stay (LOS), ICU admission, ICU LOS, mechanical ventilation, and in-hospital death. Both CT-SS and % AA were highly correlated with the hospital LOS, the risk of ICU admission, the risk of mechanical ventilation and the risk of in-hospital death. Thus, CAD4COVID-CT analysis proved to be a useful tool in detecting patients with higher hospitalization severity risk. It will help for management of the patients flow. The software measured the extent of lung damage with great efficiency, thus relieving the workload of radiologists.

Database and AI Diagnostic Tools Improve Understanding of Lung Damage, Correlation of Pulmonary Disease and Brain Damage in COVID-19

The COVID-19 pandemic caused a sharp increase in the interest in artificial intelligence (AI) as a tool supporting the work of doctors in difficult conditions and providing early detection of the implications of the disease. Recent studies have shown that AI has been successfully applied in the healthcare sector. The objective of this paper is to perform a systematic review to summarize the electroencephalogram (EEG) findings in patients with coronavirus disease (COVID-19) and databases and tools used in artificial intelligence algorithms, supporting the diagnosis and correlation between lung disease and brain damage, and lung damage. Available search tools containing scientific publications, such as PubMed and Google Scholar, were comprehensively evaluated and searched with open databases and tools used in AI algorithms. This work aimed to collect papers from the period of January 2019-May 2022 including in their resources the database from which data necessary for further development of algorithms supporting the diagnosis of the respiratory system can be downloaded and the correlation between lung disease and brain damage can be evaluated. The 10 articles which show the most interesting AI algorithms, trained by using open databases and associated with lung diseases, were included for review with 12 articles related to EEGs, which have/or may be related with lung diseases.

AI-Based Chest CT Analysis for Rapid COVID-19 Diagnosis and Prognosis: A Practical Tool to Flag High-Risk Patients and Lower Healthcare Costs

Early diagnosis of COVID-19 is required to provide the best treatment to our patients, to prevent the epidemic from spreading in the community, and to reduce costs associated with the aggravation of the disease. We developed a decision tree model to evaluate the impact of using an artificial intelligence-based chest computed tomography (CT) analysis software (icolung, icometrix) to analyze CT scans for the detection and prognosis of COVID-19 cases. The model compared routine practice where patients receiving a chest CT scan were not screened for COVID-19, with a scenario where icolung was introduced to enable COVID-19 diagnosis. The primary outcome was to evaluate the impact of icolung on the transmission of COVID-19 infection, and the secondary outcome was the in-hospital length of stay. Using EUR 20000 as a willingness-to-pay threshold, icolung is cost-effective in reducing the risk of transmission, with a low prevalence of COVID-19 infections. Concerning the hospitalization cost, icolung is cost-effective at a higher value of COVID-19 prevalence and risk of hospitalization. This model provides a framework for the evaluation of AI-based tools for the early detection of COVID-19 cases. It allows for making decisions regarding their implementation in routine practice, considering both costs and effects.

Atlas-Based Lung Segmentation Combined With Automatic Densitometry Characterization In COVID-19 Patients: Training, Validation And First Application In A Longitudinal Study

Purpose

To develop and validate an automated segmentation tool for COVID-19 lung CTs. To combine it with densitometry information in identifying Aerated, Intermediate and Consolidated Volumes in admission (CT1) and follow up CT (CT3).

Materials and Methods

An Atlas was trained on manually segmented CT1 of 250 patients and validated on 10 CT1 of the training group, 10 new CT1 and 10 CT3, by comparing DICE index between automatic (AUTO), automatic-corrected (AUTOMAN) and manual (MAN) contours. A previously developed automatic method was applied on HU lung density histograms to quantify Aerated, Intermediate and Consolidated Volumes. Volumes of subregions in validation CT1 and CT3 were quantified for each method.

Results

In validation CT1/CT3, manual correction of automatic contours was not necessary in 40% of cases. Mean DICE values for both lungs were 0.94 for AUTOVsMAN and 0.96 for AUTOMANVsMAN. Differences between Aerated and Intermediate Volumes quantified with AUTOVsMAN contours were always < 6%. Consolidated Volumes showed larger differences (mean: −95 ± 72 cc). If considering AUTOMANVsMAN volumes, differences got further smaller for Aerated and Intermediate, and were drastically reduced for consolidated Volumes (mean: −36 ± 25 cc). The average time for manual correction of automatic lungs contours on CT1 was 5 ± 2 min.

Conclusions

An Atlas for automatic segmentation of lungs in COVID-19 patients was developed and validated. Combined with a previously developed method for lung densitometry characterization, it provides a fast, operator-independent way to extract relevant quantitative parameters with minimal manual intervention.

Radiology Implementation Considerations for Artificial Intelligence (AI) Applied to COVID-19, From the AJR Special Series on AI Applications

Hundreds of imaging-based artificial intelligence (AI) models have been developed in response to the COVID-19 pandemic. AI systems that incorporate imaging have shown promise in primary detection, severity grading, and prognostication of outcomes in COVID-19, and have enabled integration of imaging with a broad range of additional clinical and epidemiologic data. However, systematic reviews of AI models applied to COVID-19 medical imaging have highlighted problems in the field, including methodologic issues and problems in real-world deployment. Clinical use of such models should be informed by both the promise and potential pitfalls of implementation. How does a practicing radiologist make sense of this complex topic, and what factors should be considered in the implementation of AI tools for imaging of COVID-19? This critical review aims to help the radiologist understand the nuances that impact the clinical deployment of AI for imaging of COVID-19. We review imaging use cases for AI models in COVID-19 (e.g., diagnosis, severity assessment, and prognostication) and explore considerations for AI model development and testing, deployment infrastructure, clinical user interfaces, quality control, and institutional review board and regulatory approvals, with a practical focus on what a radiologist should consider when implementing an AI tool for COVID-19.

Healthcare Innovations to Address the Challenges of the COVID-19 Pandemic

We have been faced with an unprecedented challenge in combating the COVID-19/SARS-CoV2 outbreak that is threatening the fabric of our civilization, causing catastrophic human losses and a tremendous economic burden globally. During this difficult time, there has been an urgent need for biomedical engineers, clinicians, and healthcare industry leaders to work together to develop novel diagnostics and treatments to fight the pandemic including the development of portable, rapidly deployable, and affordable diagnostic testing kits, personal protective equipment, mechanical ventilators, vaccines, and data analysis and modeling tools. In this position paper, we address the urgent need to bring these inventions into clinical practices. This paper highlights and summarizes the discussions and new technologies in COVID-19 healthcare, screening, tracing, and treatment-related presentations made at the IEEE EMBS Public Forum on COVID-19. The paper also provides recent studies, statistics and data and new perspectives on ongoing and future challenges pertaining to the COVID-19 pandemic.

The Role of Imaging in the Detection of Non-COVID-19 Pathologies during the Massive Screening of the First Pandemic Wave

During the COVID-19 pandemic induced by the SARS-CoV-2, numerous chest scans were carried out in order to establish the diagnosis, quantify the extension of lesions but also identify the occurrence of potential pulmonary embolisms. In this perspective, the performed chest scans provided a varied database for a retrospective analysis of non-COVID-19 chest pathologies discovered de novo. The fortuitous discovery of de novo non-COVID-19 lesions was generally not detected by the automated systems for COVID-19 pneumonia developed in parallel during the pandemic and was thus identified on chest CT by the radiologist. The objective is to use the study of the occurrence of non-COVID-19-related chest abnormalities (known and unknown) in a large cohort of patients having suffered from confirmed COVID-19 infection and statistically correlate the clinical data and the occurrence of these abnormalities in order to assess the potential of increased early detection of lesions/alterations. This study was performed on a group of 362 COVID-19-positive patients who were prescribed a CT scan in order to diagnose and predict COVID-19-associated lung disease. Statistical analysis using mean, standard deviation (SD) or median and interquartile range (IQR), logistic regression models and linear regression models were used for data analysis. Results were considered significant at the 5% critical level (p < 0.05). These de novo non-COVID-19 thoracic lesions detected on chest CT showed a significant prevalence in cardiovascular pathologies, with calcifying atheromatous anomalies approaching nearly 35.4% in patients over 65 years of age. The detection of non-COVID-19 pathologies was mostly already known, except for suspicious nodule, thyroid goiter and the ascending thoracic aortic aneurysm. The presence of vertebral compression or signs of pulmonary fibrosis has shown a significant impact on inpatient length of stay. The characteristics of the patients in this sample, both from a demographic and a tomodensitometric point of view on non-COVID-19 pathologies, influenced the length of hospital stay as well as the risk of intra-hospital death. This retrospective study showed that the potential importance of the detection of these non-COVID-19 lesions by the radiologist was essential in the management and the intra-hospital course of the patients.

Correlation between chest CT severity scores and clinical and biochemical parameters of COVID-19 pneumonia

BACKGROUND

The COVID-19 pandemic, which first appeared in Wuhan, China, in December 2019 and spread rapidly around the globe, continues to be a serious threat today. Rapid and accurate diagnostic methods are needed to identify, isolate and treat patients as soon as possible because of the rapid contagion of COVID-19. In the present study, the relation of the semi-quantitative scoring method with computed tomography in the diagnosis of COVID-19 in determining the severity of the disease with clinical and laboratory parameters and survival of the patients were investigated along with its value in prognostic prediction.

MATERIAL AND METHOD

A total of 277 adult patients who were followed up in the chest diseases clinic because of COVID-19 pneumonia between 11.03.2020 and 31.05.2020 were evaluated retrospectively in the present study. Both lungs were divided into five regions in line with their anatomical structures, and semiquantitative radiological scoring was made between 0 and 25 points according to the distribution of lesions in each region. The relations between semiquantitative radiological score and age, gender, comorbidity, and clinical and laboratory parameters were examined.

RESULTS

A significant correlation was detected between advanced age, lymphopenia, low oxygen saturation, high ferritin, D-dimer, and radiological score in the univariate analysis performed in the present study. The cut-off value of the semiquantitative radiology score was found to be 15 (AUC: 0.615, 95% CI: 0.554-0.617, p = 0.106) in ROC analysis. The survival was found to be better in cases with a radiology score below 15, in Kaplan-Meier analysis (HR: 4.71, 95% CI: 1.43-15.46, p < 0.01). In the radiological score and nonparametric correlation analyses, positive correlations were detected between CRP, D-dimer, AST, LDH, ferritin, and pro-BNP, and a negative correlation was found between partial oxygen pressure and oxygen saturation (p = 0.01, r = 0.321/0.313/0.362/0.343/0.313/0.333/-0.235/-0.231, respectively) CONCLUSION: It was found that the scoring system that was calculated quantitatively in thorax HRCTs in Covid-19 patients is a predictive actor in determining the severity and prognosis of the disease in correlation with clinical and laboratory parameters. Considering patients who have a score of 15 and above with semiquantitative scoring risky in terms of poor prognosis and short survival and close follow-up and early treatment may be effective to reduce mortality rates.

Deep learning-based lesion subtyping and prediction of clinical outcomes in COVID-19 pneumonia using chest CT

The main objective of this work is to develop and evaluate an artificial intelligence system based on deep learning capable of automatically identifying, quantifying, and characterizing COVID-19 pneumonia patterns in order to assess disease severity and predict clinical outcomes, and to compare the prediction performance with respect to human reader severity assessment and whole lung radiomics. We propose a deep learning based scheme to automatically segment the different lesion subtypes in nonenhanced CT scans. The automatic lesion quantification was used to predict clinical outcomes. The proposed technique has been independently tested in a multicentric cohort of 103 patients, retrospectively collected between March and July of 2020. Segmentation of lesion subtypes was evaluated using both overlapping (Dice) and distance-based (Hausdorff and average surface) metrics, while the proposed system to predict clinically relevant outcomes was assessed using the area under the curve (AUC). Additionally, other metrics including sensitivity, specificity, positive predictive value and negative predictive value were estimated. 95% confidence intervals were properly calculated. The agreement between the automatic estimate of parenchymal damage (%) and the radiologists' severity scoring was strong, with a Spearman correlation coefficient (R) of 0.83. The automatic quantification of lesion subtypes was able to predict patient mortality, admission to the Intensive Care Units (ICU) and need for mechanical ventilation with an AUC of 0.87, 0.73 and 0.68 respectively. The proposed artificial intelligence system enabled a better prediction of those clinically relevant outcomes when compared to the radiologists' interpretation and to whole lung radiomics. In conclusion, deep learning lesion subtyping in COVID-19 pneumonia from noncontrast chest CT enables quantitative assessment of disease severity and better prediction of clinical outcomes with respect to whole lung radiomics or radiologists' severity score.

External Validation of Deep Learning Algorithms for Radiologic Diagnosis: A Systematic Review

Purpose

To assess generalizability of published deep learning (DL) algorithms for radiologic diagnosis.

Materials and Methods

In this systematic review, the PubMed database was searched for peer-reviewed studies of DL algorithms for image-based radiologic diagnosis that included external validation, published from January 1, 2015, through April 1, 2021. Studies using nonimaging features or incorporating non-DL methods for feature extraction or classification were excluded. Two reviewers independently evaluated studies for inclusion, and any discrepancies were resolved by consensus. Internal and external performance measures and pertinent study characteristics were extracted, and relationships among these data were examined using nonparametric statistics.

Results

Eighty-three studies reporting 86 algorithms were included. The vast majority (70 of 86, 81%) reported at least some decrease in external performance compared with internal performance, with nearly half (42 of 86, 49%) reporting at least a modest decrease (≥0.05 on the unit scale) and nearly a quarter (21 of 86, 24%) reporting a substantial decrease (≥0.10 on the unit scale). No study characteristics were found to be associated with the difference between internal and external performance.

Conclusion

Among published external validation studies of DL algorithms for image-based radiologic diagnosis, the vast majority demonstrated diminished algorithm performance on the external dataset, with some reporting a substantial performance decrease.Keywords: Meta-Analysis, Computer Applications-Detection/Diagnosis, Neural Networks, Computer Applications-General (Informatics), Epidemiology, Technology Assessment, Diagnosis, Informatics Supplemental material is available for this article. © RSNA, 2022.

CT-based severity assessment for COVID-19 using weakly supervised non-local CNN

Evaluating patient criticality is the foremost step in administering appropriate COVID-19 treatment protocols. Learning an Artificial Intelligence (AI) model from clinical data for automatic risk-stratification enables accelerated response to patients displaying critical indicators. Chest CT manifestations including ground-glass opacities and consolidations are a reliable indicator for prognostic studies and show variability with patient condition. To this end, we propose a novel attention framework to estimate COVID-19 severity as a regression score from a weakly annotated CT scan dataset. It takes a non-locality approach that correlates features across different parts and spatial scales of the 3D scan. An explicit guidance mechanism from limited infection labeling drives attention refinement and feature modulation. The resulting encoded representation is further enriched through cross-channel attention. The attention model also infuses global contextual awareness into the deep voxel features by querying the base CT scan to mine relevant features. Consequently, it learns to effectively localize its focus region and chisel out the infection precisely. Experimental validation on the MosMed dataset shows that the proposed architecture has significant potential in augmenting existing methods as it achieved a 0.84 R-squared score and 0.133 mean absolute difference.

The Accuracy of Machine Learning Approaches Using Non-image Data for the Prediction of COVID-19: A Meta-Analysis

Objective

COVID-19 is a novel, severely contagious disease with enormous negative impact on humanity as well as the world economy. An expeditious, feasible tool for detecting COVID-19 remains yet elusive. Recently, there has been a surge of interest in applying machine learning techniques to predict COVID-19 using non-image data. We have therefore undertaken a meta-analysis to quantify the diagnostic performance of machine learning models facilitating the prediction of COVID-19.

Materials and methods

A comprehensive electronic database search for the period between January 1st, 2021 and December 3rd, 2021 was undertaken in order to identify eligible studies relevant to this meta-analysis. Summary sensitivity, specificity, and the area under receiver operating characteristic curves were used to assess potential diagnostic accuracy. Risk of bias was assessed by means of a revised Quality Assessment of Diagnostic Studies.

Results

A total of 30 studies, including 34 models, met all of the inclusion criteria. Summary sensitivity, specificity, and area under receiver operating characteristic curves were 0.86, 0.86, and 0.91, respectively. The purpose of machine learning models, class imbalance, and feature selection are significant covariates useful in explaining the between-study heterogeneity, in terms of both sensitivity and specificity.

Conclusions

Our study findings show that non-image data can be used to predict COVID-19 with an acceptable performance. Further, class imbalance and feature selection are suggested to be incorporated whenever building models for the prediction of COVID-19, thus improving further diagnostic performance.

Imaging related to underlying immunological and pathological processes in COVID-19

The introduction of coronavirus disease-2019 (COVID-19) as a global pandemic has contributed to overall morbidity and mortality. With a focus on understanding the immunology and pathophysiology of the disease, these features can be linked with the respective findings of imaging studies. Thus, the constellation between clinical presentation, histological, laboratory, immunological, and imaging results is crucial for the proper management of patients. The purpose of this article is to examine the role of imaging during the particular stages of severe acute respiratory syndrome coronavirus 2 infection – asymptomatic stage, typical and atypical COVID-19 pneumonia, acute respiratory distress syndrome, multiorgan failure, and thrombosis. The use of imaging methods to assess the severity and duration of changes is crucial in patients with COVID-19. Radiography and computed tomography are among the methods that allow accurate characterization of changes.

An Interpretable Chest CT Deep Learning Algorithm for Quantification of COVID-19 Lung Disease and Prediction of Inpatient Morbidity and Mortality

Rationale and Objectives

The burden of coronavirus disease 2019 (COVID-19) airspace opacities is time consuming and challenging to quantify on computed tomography. The purpose of this study was to evaluate the ability of a deep convolutional neural network (dCNN) to predict inpatient outcomes associated with COVID-19 pneumonia.

Materials and Methods

A previously trained dCNN was tested on an external validation cohort of 241 patients who presented to the emergency department and received a chest computed tomography scan, 93 with COVID-19 and 168 without. Airspace opacity scoring systems were defined by the extent of airspace opacity in each lobe, totaled across the entire lungs. Expert and dCNN scores were concurrently evaluated for interobserver agreement, while both dCNN identified airspace opacity scoring and raw opacity values were used in the prediction of COVID-19 diagnosis and inpatient outcomes.

Results

Interobserver agreement for airspace opacity scoring was 0.892 (95% CI 0.834-0.930). Probability of each outcome behaved as a logistic function of the opacity scoring (25% intensive care unit admission at score of 13/25, 25% intubation at 17/25, and 25% mortality at 20/25). Length of hospitalization, intensive care unit stay, and intubation were associated with larger airspace opacity score (p = 0.032, 0.039, 0.036, respectively).

Conclusion

The tested dCNN was highly predictive of inpatient outcomes, performs at a near expert level, and provides added value for clinicians in terms of prognostication and disease severity.

Automated COVID-19 Grading With Convolutional Neural Networks in Computed Tomography Scans: A Systematic Comparison

Amidst the ongoing pandemic, the assessment of computed tomography (CT) images for COVID-19 presence can exceed the workload capacity of radiologists. Several studies addressed this issue by automating COVID-19 classification and grading from CT images with convolutional neural networks (CNNs). Many of these studies reported initial results of algorithms that were assembled from commonly used components. However, the choice of the components of these algorithms was often pragmatic rather than systematic and systems were not compared to each other across papers in a fair manner. We systematically investigated the effectiveness of using 3-D CNNs instead of 2-D CNNs for seven commonly used architectures, including DenseNet, Inception, and ResNet variants. For the architecture that performed best, we furthermore investigated the effect of initializing the network with pretrained weights, providing automatically computed lesion maps as additional network input, and predicting a continuous instead of a categorical output. A 3-D DenseNet-201 with these components achieved an area under the receiver operating characteristic curve of 0.930 on our test set of 105 CT scans and an AUC of 0.919 on a publicly available set of 742 CT scans, a substantial improvement in comparison with a previously published 2-D CNN. This article provides insights into the performance benefits of various components for COVID-19 classification and grading systems. We have created a challenge on grand-challenge.org to allow for a fair comparison between the results of this and future research.

Facilitating standardized COVID-19 suspicion prediction based on computed tomography radiomics in a multi-demographic setting

Objective

To develop an automatic COVID-19 Reporting and Data System (CO-RADS)–based classification in a multi-demographic setting.

Methods

This multi-institutional review boards–approved retrospective study included 2720 chest CT scans (mean age, 58 years [range 18–100 years]) from Italian and Russian patients. Three board-certified radiologists from three countries assessed randomly selected subcohorts from each population and provided CO-RADS–based annotations. CT radiomic features were extracted from the selected subcohorts after preprocessing steps like lung lobe segmentation and automatic noise reduction. We compared three machine learning models, logistic regression (LR), multilayer perceptron (MLP), and random forest (RF) for the automated CO-RADS classification. Model evaluation was carried out in two scenarios, first, training on a mixed multi-demographic subcohort and testing on an independent hold-out dataset. In the second scenario, training was done on a single demography and externally validated on the other demography.

Results

The overall inter-observer agreement for the CO-RADS scoring between the radiologists was substantial (k = 0.80). Irrespective of the type of validation test scenario, suspected COVID-19 CT scans were identified with an accuracy of 84%. SHapley Additive exPlanations (SHAP) interpretation showed that the “wavelet_(LH)_GLCM_Imc1” feature had a positive impact on COVID prediction both with and without noise reduction. The application of noise reduction improved the overall performance between the classifiers for all types.

Conclusion

Using an automated model based on the COVID-19 Reporting and Data System (CO-RADS), we achieved clinically acceptable performance in a multi-demographic setting. This approach can serve as a standardized tool for automated COVID-19 assessment.

Keypoints

• Automatic CO-RADS scoring of large-scale multi-demographic chest CTs with mean AUC of 0.93 ± 0.04.

• Validation procedure resembles TRIPOD 2b and 3 categories, enhancing the quality of experimental design to test the cross-dataset domain shift between institutions aiding clinical integration.

• Identification of COVID-19 pneumonia in the presence of community-acquired pneumonia and other comorbidities with an AUC of 0.92.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-022-08730-6.

Computed Tomography Severity Scoring on High-Resolution Computed Tomography Thorax and Inflammatory Markers With COVID-19 Related Mortality in a Designated COVID Hospital

Introduction Radiological Society of the Netherlands introduced the coronavirus disease 2019 (COVID-19) Reporting and Data System (CO-RADS) and the corresponding CT severity score (CTSS) to diagnose COVID-19 severity. However, data regarding the same is very limited. Objectives The objective of this study was to correlate the computed tomography severity scoring (CTSS) on high-resolution computed tomography (HRCT) thorax and inflammatory markers with COVID-19 related mortality. Methods A retrospective observational study was conducted in a tertiary center between June 2020 to May 2021 among 2343 adult patients at the department of radio-diagnosis with suspected and confirmed COVID-19 cases who received an HRCT thorax. Data was collected retrospectively from the records regarding age, sex, and information regarding inflammatory markers such as C-reactive protein (CRP), ferritin, lactate dehydrogenase (LDH), erythrocyte sedimentation rate (ESR), D-dimer, and neutrophil-to-lymphocyte ratio. Information on HRCT thorax of patients was reviewed for radiological suspicion of COVID-19 related lung changes using CO-RADS scoring and severity of lung involvement using CT-severity scoring. Data was analyzed using SPSS version 22 (IBM Inc., Armonk, New York). Results The mean age was 51.69 ± 16.02 years, and most of the study participants were male (1592, 67.95%). The majority (999, 42.64%) had diabetes as a comorbidity. The reverse transcription polymerase chain reaction (RT-PCR) test was positive in 1571 (67.05%) participants. The majority (1571, 67.05%) had a CO-RADS score of six, and only 150 (6.40%) had CO-RADS score of four. The CT severity score was normal in 147 (6.27%), mild in 724 (30.90%), moderate in 903 (38.54%), and severe in 569 (24.29%) participants. The CRP levels were moderate in 1200 (51.22%) and severe in 428 (18.27%) participants. The mean ferritin, D-dimer and interleukin-6 (IL-6) levels were 321.83 ± 266.42 ng/ml, 1.51 ± 0.85mg/l, and 323.05 ± 95.52pg/ml, respectively. The mean length of hospital stay was 10.25 ± 6.52 days. Most of them (1926 out of 2343, 82.20%) survived, and nearly 417 out of 2343 (17.80%) died. Out of 2343, 569 participants had severe CT severity scores, out of which 205 (36.03%) died, and 364 (63.97%) participants survived. Conclusion A positive correlation was found between CT severity scoring on HRCT thorax and inflammatory markers with COVID-19 related mortality and can be used in early diagnosis and timely management of COVID-19 positive patients.

Prognostic findings for ICU admission in patients with COVID-19 pneumonia: baseline and follow-up chest CT and the added value of artificial intelligence

Infection with SARS-CoV-2 has dominated discussion and caused global healthcare and economic crisis over the past 18 months. Coronavirus disease 19 (COVID-19) causes mild-to-moderate symptoms in most individuals. However, rapid deterioration to severe disease with or without acute respiratory distress syndrome (ARDS) can occur within 1-2 weeks from the onset of symptoms in a proportion of patients. Early identification by risk stratifying such patients who are at risk of severe complications of COVID-19 is of great clinical importance. Computed tomography (CT) is widely available and offers the potential for fast triage, robust, rapid, and minimally invasive diagnosis: Ground glass opacities (GGO), crazy-paving pattern (GGO with superimposed septal thickening), and consolidation are the most common chest CT findings in COVID pneumonia. There is growing interest in the prognostic value of baseline chest CT since an early risk stratification of patients with COVID-19 would allow for better resource allocation and could help improve outcomes. Recent studies have demonstrated the utility of baseline chest CT to predict intensive care unit (ICU) admission in patients with COVID-19. Furthermore, developments and progress integrating artificial intelligence (AI) with computer-aided design (CAD) software for diagnostic imaging allow for objective, unbiased, and rapid assessment of CT images.

Diagnostic performance of CO-RADS for COVID-19: a systematic review and meta-analysis

Objectives

To investigate the diagnostic performance of the coronavirus disease 2019 (COVID-19) Reporting and Data System (CO-RADS) for detecting COVID-19.

Methods

We searched PubMed, EMBASE, MEDLINE, Web of Science, Cochrane Library, and Scopus database until September 21, 2021. Statistical analysis included data pooling, forest plot construction, heterogeneity testing, meta-regression, and subgroup analyses.

Results

We included 24 studies with 8382 patients. The pooled sensitivity and specificity and the area under the curve (AUC) of CO-RADS ≥ 3 for detecting COVID-19 were 0.89 (95% confidence interval (CI) 0.85–0.93), 0.68 (95% CI 0.60–0.75), and 0.87 (95% CI 0.84–0.90), respectively. The pooled sensitivity and specificity and AUC of CO-RADS ≥ 4 were 0.83 (95% CI 0.79–0.87), 0.84 (95% CI 0.78–0.88), and 0.90 (95% CI 0.87–0.92), respectively. Cochran’s Q test (p < 0.01) and Higgins I² heterogeneity index revealed considerable heterogeneity. Studies with both symptomatic and asymptomatic patients had higher specificity than those with only symptomatic patients using CO-RADS ≥ 3 and CO-RADS ≥ 4. Using CO-RADS ≥ 4, studies with participants aged < 60 years had higher sensitivity (0.88 vs. 0.80, p = 0.02) and lower specificity (0.77 vs. 0.87, p = 0.01) than studies with participants aged > 60 years.

Conclusions

CO-RADS has favorable performance in detecting COVID-19. CO-RADS ≥ 3/4 might be applied as cutoff values given their high sensitivity and specificity. However, there is a need for more well-designed studies on CO-RADS.

Key Points

• CO-RADS shows a favorable performance in detecting COVID-19.

• CO-RADS ≥ 3 had a high sensitivity 0.89 (95% CI 0.85–0.93), and it may prove advantageous in screening the potentially infected people to prevent the spread of COVID-19.

• CO-RADS ≥ 4 had high specificity 0.84 (95% CI 0.78–0.88) and may be more suitable for definite diagnosis of COVID-19.

A Novel Threshold-Based Segmentation Method for Quantification of COVID-19 Lung Abnormalities

Since December 2019, the novel coronavirus disease 2019 (COVID-19) has claimed the lives of more than 3.75 million people worldwide. Consequently, methods for accurate COVID-19 diagnosis and classification are necessary to facilitate rapid patient care and terminate viral spread. Lung infection segmentations are useful to identify unique infection patterns that may support rapid diagnosis, severity assessment, and patient prognosis prediction, but manual segmentations are time-consuming and depend on radiologic expertise. Deep learning-based methods have been explored to reduce the burdens of segmentation; however, their accuracies are limited due to the lack of large, publicly available annotated datasets that are required to establish ground truths. For these reasons, we propose a semi-automatic, threshold-based segmentation method to generate region of interest (ROI) segmentations of infection visible on lung computed tomography (CT) scans. Infection masks are then used to calculate the percentage of lung abnormality (PLA) to determine COVID-19 severity and to analyze the disease progression in follow-up CTs. Compared with other COVID-19 ROI segmentation methods, on average, the proposed method achieved improved precision ( 47.49 % ) and specificity ( 98.40 % ) scores. Furthermore, the proposed method generated PLAs with a difference of ± 3.89 % from the ground-truth PLAs. The improved ROI segmentation results suggest that the proposed method has potential to assist radiologists in assessing infection severity and analyzing disease progression in follow-up CTs.

Comparison of CO-RADS Scores Based on Visual and Artificial Intelligence Assessments in a Non-Endemic Area

In this study, we first developed an artificial intelligence (AI)-based algorithm for classifying chest computed tomography (CT) images using the coronavirus disease 2019 Reporting and Data System (CO-RADS). Subsequently, we evaluated its accuracy by comparing the calculated scores with those assigned by radiologists with varying levels of experience. This study included patients with suspected SARS-CoV-2 infection who underwent chest CT imaging between February and October 2020 in Japan, a non-endemic area. For each chest CT, the CO-RADS scores, determined by consensus among three experienced chest radiologists, were used as the gold standard. Images from 412 patients were used to train the model, whereas images from 83 patients were tested to obtain AI-based CO-RADS scores for each image. Six independent raters (one medical student, two residents, and three board-certified radiologists) evaluated the test images. Intraclass correlation coefficients (ICC) and weighted kappa values were calculated to determine the inter-rater agreement with the gold standard. The mean ICC and weighted kappa were 0.754 and 0.752 for the medical student and residents (taken together), 0.851 and 0.850 for the diagnostic radiologists, and 0.913 and 0.912 for AI, respectively. The CO-RADS scores calculated using our AI-based algorithm were comparable to those assigned by radiologists, indicating the accuracy and high reproducibility of our model. Our study findings would enable accurate reading, particularly in areas where radiologists are unavailable, and contribute to improvements in patient management and workflow.

Automated prediction of low ferritin concentrations using a machine learning algorithm

OBJECTIVES

Computational algorithms for the interpretation of laboratory test results can support physicians and specialists in laboratory medicine. The aim of this study was to develop, implement and evaluate a machine learning algorithm that automatically assesses the risk of low body iron storage, reflected by low ferritin plasma levels, in anemic primary care patients using a minimal set of basic laboratory tests, namely complete blood count and C-reactive protein (CRP).

METHODS

Laboratory measurements of anemic primary care patients were used to develop and validate a machine learning algorithm. The performance of the algorithm was compared to twelve specialists in laboratory medicine from three large teaching hospitals, who predicted if patients with anemia have low ferritin levels based on laboratory test reports (complete blood count and CRP). In a second round of assessments the algorithm outcome was provided to the specialists in laboratory medicine as a decision support tool.

RESULTS

Two separate algorithms to predict low ferritin concentrations were developed based on two different chemistry analyzers, with an area under the curve of the ROC of 0.92 (Siemens) and 0.90 (Roche). The specialists in laboratory medicine were less accurate in predicting low ferritin concentrations compared to the algorithms, even when knowing the output of the algorithms as support tool. Implementation of the algorithm in the laboratory system resulted in one new iron deficiency diagnosis on average per day.

CONCLUSIONS

Low ferritin levels in anemic patients can be accurately predicted using a machine learning algorithm based on routine laboratory test results. Moreover, implementation of the algorithm in the laboratory system reduces the number of otherwise unrecognized iron deficiencies.

Evaluation of AI-Based Segmentation Tools for COVID-19 Lung Lesions on Conventional and Ultra-low Dose CT Scans

A reliable diagnosis and accurate monitoring are pivotal steps for treatment and prevention of COVID-19. Chest computed tomography (CT) has been considered a crucial diagnostic imaging technique for the injury assessment of the viral pneumonia. Furthermore, the automatization of the segmentation methods for lung alterations helps to speed up the diagnosis and lighten radiologists' workload. Considering the assiduous pathology monitoring, ultra-low dose (ULD) chest CT protocols have been implemented to drastically reduce the radiation burden. Unfortunately, the available AI technologies have not been trained on ULD-CT data and validated and their applicability deserves careful evaluation. Therefore, this work aims to compare the results of available AI tools (BCUnet, CORADS AI, NVIDIA CLARA Train SDK and CT Pneumonia Analysis) on a dataset of 73 CT examinations acquired both with conventional dose (CD) and ULD protocols. COVID-19 volume percentage, resulting from each tool, was statistically compared. This study demonstrated high comparability of the results on CD-CT and ULD-CT data among the four AI tools, with high correlation between the results obtained on both protocols (R > .68, P < .001, for all AI tools).

Quantitative Analysis of Pleural Line and B-Lines in Lung Ultrasound Images for Severity Assessment of COVID-19 Pneumonia

Specific patterns of lung ultrasound (LUS) images are used to assess the severity of coronavirus disease 2019 (COVID-19) pneumonia, while such assessment is mainly based on clinicians' qualitative and subjective observations. In this study, we quantitatively analyze the LUS images to assess the severity of COVID-19 pneumonia by characterizing the patterns related to the pleural line (PL) and B-lines (BLs). Twenty-seven patients with COVID-19 pneumonia, including 13 moderate cases, seven severe cases, and seven critical cases, are enrolled. Features related to the PL, including the thickness (TPL) and roughness of the PL (RPL), and the mean (MPLI) and standard deviation (SDPLI) of the PL intensities are extracted from the LUS images. Features related to the BLs, including the number (NBL), accumulated width (AWBL), attenuation coefficient (ACBL), and accumulated intensity (AIBL) of BLs, are also extracted. The correlations of these features with the disease severity are evaluated. The performances of the binary severe/non-severe classification are assessed for each feature and support vector machine (SVM) classifiers with various combinations of features as input. Several features, including the RPL, NBL, AWBL, and AIBL, show significant correlations with disease severity (all ). The classification performance is optimal using the SVM classifier using all the features as input (area under the receiver operating characteristic (ROC) curve = 0.96, sensitivity = 0.93, and specificity = 1). These findings demonstrate that the proposed method may be a promising tool for automatic grading diagnosis and follow-up of patients with COVID-19 pneumonia.

COLI-Net: Deep learning-assisted fully automated COVID-19 lung and infection pneumonia lesion detection and segmentation from chest computed tomography images

We present a deep learning (DL)-based automated whole lung and COVID-19 pneumonia infectious lesions (COLI-Net) detection and segmentation from chest computed tomography (CT) images. This multicenter/multiscanner study involved 2368 (347'259 2D slices) and 190 (17 341 2D slices) volumetric CT exams along with their corresponding manual segmentation of lungs and lesions, respectively. All images were cropped, resized, and the intensity values clipped and normalized. A residual network with non-square Dice loss function built upon TensorFlow was employed. The accuracy of lung and COVID-19 lesions segmentation was evaluated on an external reverse transcription-polymerase chain reaction positive COVID-19 dataset (7'333 2D slices) collected at five different centers. To evaluate the segmentation performance, we calculated different quantitative metrics, including radiomic features. The mean Dice coefficients were 0.98 ± 0.011 (95% CI, 0.98-0.99) and 0.91 ± 0.038 (95% CI, 0.90-0.91) for lung and lesions segmentation, respectively. The mean relative Hounsfield unit differences were 0.03 ± 0.84% (95% CI, -0.12 to 0.18) and -0.18 ± 3.4% (95% CI, -0.8 to 0.44) for the lung and lesions, respectively. The relative volume difference for lung and lesions were 0.38 ± 1.2% (95% CI, 0.16-0.59) and 0.81 ± 6.6% (95% CI, -0.39 to 2), respectively. Most radiomic features had a mean relative error less than 5% with the highest mean relative error achieved for the lung for the range first-order feature (-6.95%) and least axis length shape feature (8.68%) for lesions. We developed an automated DL-guided three-dimensional whole lung and infected regions segmentation in COVID-19 patients to provide fast, consistent, robust, and human error immune framework for lung and pneumonia lesion detection and quantification.