Back

A globally available COVID-19 - Template for clinical imaging studies

Salg, G. A.; Ganten, M. K.; Baumhauer, M.; Heussel, C. P.; Kleesiek, J.

2020-04-07 radiology and imaging
10.1101/2020.04.02.20048793 medRxiv
Show abstract

BackgroundThe pandemic spread of COVID-19 has caused worldwide implications on societies and economies. Chest computed tomography (CT) has been found to support both, current diagnostic and disease monitoring. A joint approach to collect, analyze and share clinical and imaging information about COVID-19 in the highest quality possible is urgently needed. MethodsAn evidence-based reporting template was developed for assessing COVID-19 pneumonia using an FDA-approved medical software. The annotation of qualitative and quantitative findings including radiomics features is performed directly on primary imaging data. For data collection, secondary information from the patient history and clinical data such as symptoms and comorbidities are queried. ResultsLicense-royalty free, cloud-based web platform and on-premise deployments are offered. Hospitals can upload, assess, report and if pseudonymized share their COVID-19 cases. The aggregation of radiomics in correlation with rt-PCR, patient history, clinical and radiological findings, systematically documented in a single database, will lead to optimized diagnosis, risk stratification and response evaluation. A customizable analytics dashboard allows the explorative real-time data analysis of imaging features and clinical information. ConclusionsThe COVID-19-Template is based on a systematic, computer-assisted and context-guided approach to collect, analyze and share data. Epidemiological and clinical studies for therapies and vaccine candidates can be implemented in compliance with high data quality, integrity and traceability. An additional explanation video of the COVID-19-Template video is provided via:http://cloud1.mint-medical.de/downloads/player/index.html?v=Covid19StandardizedAssessmentWeb HighlightsO_LIDynamic evidence-based electronic case report form (eCRF) for COVID-19 including documentation of primary imaging data, secondary clinical data and patient history including radiomics features C_LIO_LIComputer-assisted, context-guided reporting approach based on FDA approved medical product software package available free of charge C_LIO_LIData quality, traceability, integrity in open-access web platform C_LIO_LICustomizable analytics dashboard for explorative real-time data analysis of imaging features and clinical information C_LIO_LIHuman and machine-readable data export for clinical trials C_LI

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.