Automated Blood Volume Measurement on Head CT

Purpose and goal:

The aim of this study was to validate a novel automated software (qERNCCT by Qure.ai) for quantification of intracerebral bleedings (ICH) on CT. Manual segmentation was compared to qERNCCT in 1649 ICH patients selected from the Swedish Stroke Register. The qER-NCCT fully automated volume measurement had excellent agreement with manual segmentation.

Implementing such automated software for ICH volume measurement could improve diagnostics and prognostication considerably since exact volumetric measurement currently is an unmet medical need in routine healthcare. Expected results and effects:

The main publication for this project (Accuracy of automated intracerebral hemorrhage volume measurement on non-contrast computed tomography; a Swedish Stroke Register cohort study Amir Hillal, Gabriella Sultani, Birgitta Ramgren, Bo Norrving, Johan Wassélius and Teresa Ullberg) is currently in press for Neuroradiology. The project has also generated 3 additional manuscripts that are in preparation and one congress abstract at ECR 2022.

Approach and implementation:

The implementation of the algorithm in our hospital IT-environment was possible thanks to a close collaboration with the developer of the algorithm (Qure.ai), the PACS provider (Sectra) and the clinical IT-service (Henrik Granberg at the department of Radiology in Lund). The careful planning and the positive collaborative spirit brought by all parties to the project was key in setting up the system allowing for the validation study to be done within the time of this project.