Related Experiment Video
Updated: Jul 8, 2026
Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
Diagnostic quality of multiplanar reformations obtained with a newly developed cone beam device for maxillofacial
R A Mischkowski1, P Scherer, L Ritter
1Department of Craniomaxillofacial and Plastic Surgery, University of Cologne, Kerpener Str. 62, 50937 Köln, Germany. r.mischkowski@uni-koeln.de
Objectives:
To compare the diagnostic quality of multiplanar reformations generated by a newly developed cone beam device with the diagnostic quality of multiplanar reformations obtained using a state of the art multidetector row CT scanner.
Methods:
Digital volume tomograms (DVTs) were obtained with the prototype of GALILEOS (Sirona Dental Systems Inc., Bensheim, Germany), a newly developed compact-sized cone beam machine with a scan volume of 15 x 15 x 15 cm. CT scans were performed for comparison with two 16-detector row CT scanners. The study included 30 image pairs obtained from the same patients. The multiplanar reformations in axial, coronal and sagittal planes were subjectively assessed by 3 observers using a 1(excellent)-5(poor) scale on 35 weighted criteria including findings detection, image quality and visualization of several anatomical structures.
Results:
There was no statistically significant difference between DVT and CT in cumulative values expressing the diagnostic quality of the reformations (P = 0.071). The interobserver agreement was better with CT than with DVT. Findings detection and the visualization of the majority of anatomical structures within the mandibulomaxillary area were ranked at the same level for DVT as with CT.
Conclusions:
The diagnostic quality of multiplanar reformations can be considered the same in both devices. Although sharpness, noise level and contrast resolution do not reach the level of CT, DVT images proved statistically not inferior for the detection of lesions which may be adequately depicted by both imaging modes.
