COVID-19 diagnosis and disease severity prediction assessment through an innovative AI-based model

Introduction: CT imaging has been widely used during the COVID-19 pandemic to diagnose and assess disease severity. Its use for diagnosis is not indicated apart from specific settings such as triage of patients for referral to RTPCR testing or severity assessment. Nowadays, the added value of AI-based models is still unknown and has to be addressed. Method(s): We evaluated the added value of an automated lung involvement assessment tool, named icolung. Since software version 7.0, icolung automatically extracts the Severity Score proposed by Pan F. et al., (2020, Radiology), to help radiologists assess the severity of lung involvement in COVID-19 infected patients. We evaluate retrospectively a group of 785 COVID-19 positive patients compared to a group of 1049 COVID-19 negative patients. We used the severity score (SS) in order to predict the positivity of COVID-19 PCR testing and evaluated the potential impact in the prediction of patients' outcome. Result(s): The icolung SS allows to identify infected (PCR-proven) COVID-19 patients with a sensitivity of 83% and a specificity of 77% (AUC of 0.86, 95% CI 0.85-0.88) for patients with a SS of more than 1.5 on a scale of 0 to 25. An SS of > 7.5 identifies patients at risk of ICU admission with a sensitivity of 70% and specificity of 65% (AUC of 0.74, p<0.0001). Conclusion(s): The severity score as estimated via icolung allows to identify positive PCR-tested COVID-19 patients and helps to predict ICU admission. This automated evaluation tool can support clinicians with the in-hospital management of patients (suspected to be) infected with COVID-19.

(Early) Economic Evaluation of the AI-based software 'icolung' for the detection and prognosis of COVID cases from CT scans

Early detection of COVID-19 infection, followed by appropriate patient management, has the potential to reduce costs related to developing severe forms of the disease, as well as spreading of the disease, if left undetected. Our objective was to evaluate the impact of an AI-based chest CT analysis software (icolung, icometrix) for the detection and prognosis of COVID-19 cases in patients receiving a CT scan in a hospital setting in Belgium. We developed a decision analytic model comparing routine practice scenario where patients receiving a CT scan in the hospital are not screened for COVID-19 with a scenario where icolung is used to analyze CT scans for the detection and prognosis of COVID-19 cases. We evaluated the impact of the technology in preventing the further spreading of the infection in the community and in reducing the length of hospitalization of COVID-19 patients. In the base case using a relatively low COVID-19 prevalence of 0.36%, icolung is cost-effective in preventing COVID19 transmission in the community, costing 8.221 to prevent one infection. At low prevalence of the disease and low risk of hospitalization, the technology is not cost-effective in reducing the length of hospitalization. However, icolung may be cost-effective in situations with high disease prevalence (>30%) or high risk of hospitalization (>6%) such as patients suffering from chronic oncological diseases and benefiting from recurring thoracic imaging. This model provides initial evidence of cost-effectiveness of AI-based chest CT analysis software and may help to provide guidance regarding further health care research and policy.

Late Breaking Abstract - CNN-RNN network to classify COVID, Influenza and non-infectious cases on CT imaging

COVID-19 associated lung diseases can mimic radiological characteristics of other viral lung diseases such as influenza which may lead to misdiagnosis. In this study, we proposed an Artificial Intelligence framework based on a combination of a Convolutional Neural network architecture and a Recurrent Neural Network architecture to classify CT volumes with COVID-19, Influenza, and no-infection. The model was trained on a dataset of 300 patients (100 patients in each class). Each set of 15 consecutive axial slices with the associated label of the corresponding CT volume was input as a 3 channel input at 5 time points to the CNN-RNN network. Benchmarked against RT-PCR confirmed cases of COVID-19 and Influenza, our model, when evaluated on an independent validation set of 400 CT patients, can accurately classify CT volumes of patients with COVID-19, Influenza, or no-infection with a sensitivity of 96% (COVID-19) and 95% (Influenza) (Tablel). Figurel shows the percentage of correctly classified and misclassified cases in each class. Our model provides rapid accurate diagnosis in patients suspected of COVID-19 infection, facilitating the timely implementation of isolation procedures and early intervention.

Late Breaking Abstract - Development and validation of an automated radiomic CT signature for detecting? COVID-i9

The coronavirus disease 2019 (COVID-19) outbreak has reached pandemic status and pushed healthcare systems beyond the limits. We aim to develop a fully automatic framework to detect COVID-19 by applying artificial intelligence (Al). A fully automated Al framework was developed to extract radiomics features from chest CT scans to detect COVID-19 patients. We curated and analysed the data from a total of 1381 patients. A cohort of 181 RT-PCR confirmed COVID-19 patients and 1200 control patients was included for model development. An independent dataset of 697 patients was used to validate the model. The datasets were collected from CHU Liège, Belgium. Model performance was assessed by the area under the receiver operating characteristic curve (AUC). Assuming 15% disease prevalence, a comprehensive analysis of classification performance in terms of accuracy, sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) was performed for all possible decision thresholds. The final curated dataset used for model development and testing consisted of chest CT scans of 1224 patients and 641 patients, respectively. The model had an AUC of 0.882 (95% CI: 0.851-0.913) in the independent test dataset. Assuming the cost of false negatives is twice as high as the cost of false positives, the optimal decision threshold resulted in an accuracy of 85.18%, a sensitivity of 69.52, a specificity of 91.63%, an NPV of 94.46% and a PPV of 59.44%. Our Al framework can accurately detect COVID-19. Thus, providing rapid accurate diagnosis in patients suspected of COVID-19 infection, facilitating the implementation of isolation procedures and early intervention.

[Chest radiological lesions in COVID-19 : from classical imaging to artificial intelligence]

In the course of the pandemic induced by the appearance of a new coronavirus (SARS-CoV-2;COVID-19) causing acute respiratory distress syndrome (ARDS), we had to rethink the diagnostic approach for patients suffering from respiratory symptoms. Indeed, although the use of RT-PCR remains the keystone of the diagnosis, the delay in diagnosis as well as the overload of the microbiological platforms have led us to make almost systematic the use of thoracic imaging for taking in charge of patients. In this context, thoracic imaging has shown a major interest in diagnostic aid in order to better guide the management of patients admitted to hospital. The most common signs encountered are particularly well described in thoracic computed tomography. Typical imaging combines bilateral, predominantly peripheral and posterior, multi-lobar, ground glass opacities. Of note, it is common to identify significant lesions in asymptomatic patients, with imaging sometimes preceding the onset of symptoms. Beyond conventional chest imaging, many teams have developed new artificial intelligence tools to better help clinicians in decision-making.