Image Correlation Between Digitally Reconstructed Radiographs, C-arm Fluoroscopic Radiographs, and X-ray: A Phantom Study.

Wangler, Sebastian; Hofmann, Janic; Moser, Helen L; Kuenzler, Michael; Egli, Rainer J; Schaer, Michael (2024). Image Correlation Between Digitally Reconstructed Radiographs, C-arm Fluoroscopic Radiographs, and X-ray: A Phantom Study. Cureus, 16(1) 10.7759/cureus.51868

Full text not available from this repository. (Request a copy)

OBJECTIVE

Digitally reconstructed radiographs (DRRs) are planar two-dimensional (2D) X-rays derived from a three-dimensional (3D) computed tomography (CT) dataset. DRRs allow the simulation of radiographs of all desired views and facilitate preoperative planning. However, orthopedic surgeons rely on C-arm fluoroscopic imaging during surgery to verify fracture reduction and implant placement. Pincushion distortion represents a technical limitation of fluoroscopic imaging, resulting in a greater distance between points at the periphery of the image compared to the center. This project, therefore, aimed to assess the image correlation between digitally reconstructed radiographs (DRRs) and fluoroscopic imaging (C-arm) using conventional radiographs (X-ray) as a control.

METHODS

A 3D-printed cubic prototype and an anatomical humerus bone model were used. C-arm fluoroscopic radiographs and conventional X-ray images were taken in an anteroposterior (AP) view at 10-degree steps while rotating the objects from 0 to 90 degrees. CT scans were made and used to compute and export DRRs in AP view at 10-degree rotational steps from 0 to 90 degrees. The surface area (cm2) was measured and compared between the different modalities. For automated image analysis of the anatomical humerus model, matching (%) between modalities was calculated using the structural similarity index (SSIM).

RESULTS

The overall regression was statistically significant in all models, with an R2 >0.99 when comparing all three imaging modalities of the prototype. Surface correlation in the anatomical humerus model was R2 0.99 between X-ray and C-arm and R2 0.95 between C-arm and X-ray to DRRs, respectively. The SSIM was highest for comparing DRR and C-arm images (0.84±0.01%).

CONCLUSIONS

The study indicates a strong agreement between digitally reconstructed radiographs and X-ray/C-arm images. DRRs, therefore, represent a valuable tool for research and clinical application.

Item Type:

Journal Article (Original Article)

Division/Institute:

04 Faculty of Medicine > Department of Orthopaedic, Plastic and Hand Surgery (DOPH) > Clinic of Orthopaedic Surgery
04 Faculty of Medicine > Department of Radiology, Neuroradiology and Nuclear Medicine (DRNN) > Institute of Diagnostic, Interventional and Paediatric Radiology

UniBE Contributor:

Wangler, Sebastian, Moser, Helen Laura, Künzler, Michael, Egli, Rainer Josef (A), Schär, Michael

Subjects:

600 Technology > 610 Medicine & health

ISSN:

2168-8184

Language:

English

Submitter:

Pubmed Import

Date Deposited:

12 Feb 2024 16:52

Last Modified:

12 Feb 2024 16:52

Publisher DOI:

10.7759/cureus.51868

PubMed ID:

38327943

Uncontrolled Keywords:

c-arm ct digitally reconstructed radiographs drr phantom study

URI:

https://boris.unibe.ch/id/eprint/192701

Actions (login required)

Edit item Edit item
Provide Feedback