Virtual Imaging Trials for Novel Coronavirus Disease (COVID-19)
Virtual imaging trial is a unique framework that can greatly facilitate the assessment and optimization of imaging, by emulating the experiments using representative models of patients and scanners. This study aimed to impalement and demonstrate a virtual imaging trial platform for COVID-19, enabling effective assessment and optimization of CT and X-ray radiography acquisitions for reliable imaging and management of COVID-19.
METHOD AND MATERIALS
An existing VIT platform was adapted to incorporate COVID-19 features. Using an IRB-approved protocol, confirmed COVID-19 patient cases were manually segmented for all pulmonary features related to the disease. The segmentations were verified by a cardiothoracic radiologist. The segmented features (e.g., ground glass opacity, consolidation, crazy paving) were made into 4D extended cardiac-torso (XCAT) phantoms. Within a given disease area, the texture and material of the lung parenchyma in the XCAT were modified to match the physical properties observed in the clinical images. To demonstrate the utility, the COVID-19 phantoms were virtually imaged using a scanner-specific CT and radiography simulator (DukeSim). CT and Radiography simulation included varied models of COVID-19 as a function of applied radiation dose.
We developed the first computational models of COVID-19 patients and demonstrated how their combination with imaging simulators for imaging studies. Qualitatively, the simulated abnormalities had realistic shapes and texture. This platform gives us the ability to image the same patients with both modalities under various parameters. For example, results showed that in the abnormal regions the contrast to noise ratio, a metric relevant to abnormality detection, were 1.6, 3.0, and 3.6 for 5, 25, and 50 mAs images, respectively.
The developed toolsets in this study provides the foundation for use of virtual imaging trials in effective assessment and optimization of CT and X-ray radiography acquisitions and analysis tools to manage COVID-19 pandemic.
We establish a platform in which the potential utility of CT and X-ray imaging for COVID-19 management can be assessed using models of patients and scanner with a priori knowledge of ground truth.
Abadi, E; Segars, W; Sharma, S; Ria, F; Chalian, H; Samei, E
Radiological Society of North America 106th Scientific Assembly and Annual Meeting
Conference Start Date
Conference End Date