Automatic Assessment of the Quality of Patient Positioning and Field of View of Head CT Scans

摘要

The purpose of this paper is the investigation of automatic evaluation of the quality of patient positioning andField of View (FoV) in head CT scans. Studies have shown elevated risk of radiation-induced cataract in patientsundergoing head CT examinations. The American Association of Physicists in Medicine (AAPM) published aprotocol for head CT scans including requirements linking the optimal scan angle to anatomic landmarks in theskull. To help sensitizing sta for the need of correct patient positioning, a software-based tool detecting non-optimal patient positioning was developed. Our experiments were conducted on 209 head CT exams acquiredat the University Medical Center Hamburg Eppendorf (UKE). All of these examinations were done on the samePhilips iCT scanner. Each exam contains a 3D volume with an in-plane voxel spacing of 0.44mm x 0.44mm and aslice distance of 1mm. As ground truth anatomic landmarks on the skull were annotated independently by threedierent readers. We applied an atlas registration technique to map CT scans to a probabilistic anatomical atlas.For a new CT scan, previously dened model landmarks were mapped back to the CT volume when registeringit to the atlas thus labelling new head CT scans. From the location of the detected landmarks we derive thedeviation of the actual head angulation and scan length from the optimal values. Furthermore, the presenceof the eye-lenses in the FoV is predicted. The median error of the estimated landmark positions measured asdistance to the plane generated from the ground truth landmark positions is below 1mm and comparable tothe interobserver variability. A classier for the prediction of the presence of the eye-lenses in the FoV fromthe estimated landmark locations achieves a value of 0:74. Furthermore there is moderate agreement of theestimated deviations of optimal head tilt and scan length with an expert's rating.