Evaluation of Ethos HyperSight imaging performance compared to standard CBCT and FBCT
PD-0662
Abstract
Evaluation of Ethos HyperSight imaging performance compared to standard CBCT and FBCT
Authors: Clara Fallone1,2, Lee MacDonald1,3,2, Amanda Cherpak1,2,3, Krista Chytyk-Praznik1,2,3, James Robar1,3,2
1Nova Scotia Health, Department of Medical Physics, Halifax, Canada; 2Dalhousie University, Department of Radiation Oncology, Halifax, Canada; 3Dalhousie University, Department of Physics & Atmospheric Science, Halifax, Canada
Show Affiliations
Hide Affiliations
Purpose or Objective
To compare image quality of the novel Varian Ethos 4.0 kV-CBCT HyperSight imaging with Varian Truebeam (TB) kV-CBCT and a GE Siemens kV-FBCT CT scanner.
Material and Methods
The Catphan 600 phantom (Phantom Library Inc.) was imaged using the novel Varian HyperSight imaging platform with standard pelvis, thorax, and breast protocols. Images were also acquired on the Varian TB and GE simulator. Imaging protocols on the TB and GE were selected such that CTDI (CT dose index) and beam quality were as equivalent as possible to HyperSight protocols. Image quality metrics, including spatial resolution, contrast, noise, and integral non-uniformity from the Catphan phantom were assessed for each protocol on each modality. A porcine head was also scanned on each modality using the pelvis protocol to assess soft tissue visualization; images were qualitatively compared.
Results
Image quality metrics and associated CTDI and energy parameters are summarized in Table I. Figure 1 illustrates the images for the Catphan inserts of high contrast spatial resolution, low contrast resolution, and uniformity modules for the pelvis protocol for all three systems. Spatial resolution using HyperSight was superior to that of TB for two of three protocols and superior to that of GE for one of three protocols. In cases where HyperSight resolution results were not superior to TB or GE, they were equivalent. Low contrast resolution visibility was superior with HyperSight compared to TB or GE for the pelvis and thorax protocols; in all three modalities no objects were visible using the lower dose breast protocol. The measured noise in HyperSight images was noticeably lower than that in TB or GE images for the pelvis protocol, and comparable or lower for the other two protocols. Uniformity using HyperSight was superior to that using TB for the pelvis mode, and comparable to TB and GE for the other modes. It should be noted that CTDI for the TB breast protocol was about double that of HyperSight given minimum mAs limitations on the TB. Figure 2 demonstrates visual image quality differences between the porcine head images acquired from HyperSight, TB, or GE. The porcine images reveal increased streaking artifact (indicated with the red arrow) in GE and TB images compared to HyperSight. Non-uniformities and distortion artifacts are also seen in the TB image (circled in blue), which are improved using HyperSight.
Table 1
| Modality | Protocol | Energy (kVp)
| CTDI (mGy)
| Spatial resolution
(lp/cm)
| Contrast resolution (objects visible at 1% level)
| Noise
(HU)
| Integral non-uniformity
(HU)
|
| Pelvis |
| 7.4 | 7 | 8 | 5.8 | 0.5 |
| HyperSight CBCT | Thorax | 125 | 2.7 | 6 | 7 | 11.4 | 0.9 |
| Breast |
| 0.5 | 5 | 0 | 24.9 | 2.2 |
| Pelvis |
| 7.7 | 6 | 6 | 19.5 | 5.4 |
| TB CBCT | Thorax | 125 | 3.3 | 5 | 5 | 11.8 | 1.5 |
| Breast |
| 1.3 | 5 | 0 | 28.8 | 2.7 |
| Pelvis |
| 7.0 | 7 | 7 | 8.5 | 0.8 |
| GE FBCT | Thorax | 120 | 3.0 | 5 | 2 | 15.2 | 1.2 |
| Breast |
| 0.7 | 5 | 0 | 33 | 1.8 |
Figure 1
Figure 2
Conclusion
HyperSight imaging yields several superior image quality metrics compared to Varian TB kV-CBCT and GE kV-FBCT for imaging protocols of similar energy and CTDI. Image quality improvements can also be visualized in porcine anatomy.