Image quality and dose characteristics for an O-arm intraoperative imaging system with model-based image reconstruction

Abstract

Purpose: To assess the imaging performance and radiation dose characteristics of the O-arm CBCT imaging system (Medtronic Inc., Littleton MA) and demonstrate the potential for improved image quality and reduced dose via model-based image reconstruction (MBIR).

Methods: Two main studies were performed to investigate previously unreported characteristics of the O-arm system. First is an investigation of dose and 3D image quality achieved with filtered back-projection (FBP) - including enhancements in geometric calibration, handling of lateral truncation and detector saturation, and incorporation of an isotropic apodization filter. Second is implementation of an MBIR algorithm based on Huber-penalized likelihood estimation (PLH) and investigation of image quality improvement at reduced dose. Each study involved measurements in quantitative phantoms as a basis for analysis of contrast-to-noise ratio and spatial resolution as well as imaging of a human cadaver to test the findings under realistic imaging conditions.

Results: View-dependent calibration of system geometry improved the accuracy of reconstruction as quantified by the full-width at half maximum of the point-spread function - from 0.80 to 0.65 mm - and yielded subtle but perceptible improvement in high-contrast detail of bone (e.g., temporal bone). Standard technique protocols for the head and body imparted absorbed dose of 16 and 18 mGy, respectively. For low-to-medium contrast (<100 HU) imaging at fixed spatial resolution (1.3 mm edge-spread function) and fixed dose (6.7 mGy), PLH improved CNR over FBP by +48% in the head and +35% in the body. Evaluation at different dose levels demonstrated 30% increase in CNR at 62% of the dose in the head and 90% increase in CNR at 50% dose in the body.

Conclusions: A variety of improvements in FBP implementation (geometric calibration, truncation and saturation effects, and isotropic apodization) offer the potential for improved image quality and reduced radiation dose on the O-arm system. Further gains are possible with MBIR, including improved soft-tissue visualization, low-dose imaging protocols, and extension to methods that naturally incorporate prior information of patient anatomy and/or surgical instrumentation.

Keywords: cone-beam CT; image-guided surgery; model-based image reconstruction; radiation dose; surgical navigation.

Figure 1

(a) Laboratory setup of the O‐arm and StealthStation S7 (Medtronic, Louisville CO) surgical navigation systems. An anthropomorphic phantom is shown in prone position. (b) Two‐circle BB phantom for geometric calibration and annotated coordinates for 2D projections and 3D reconstructions. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2

Analysis of spatial resolution from the ESF of a spherical soft‐tissue‐simulating insert. (a) Example PLQ reconstructions of a low‐contrast (95 HU) sphere in the liver for varying values of β. (b) Example measurement of the ESF sampled about the periphery of the sphere along with the resulting erf fit. (c) Illustration of erf fits for regularization strength log β ranging 0 to 3.25 at 0.25 increments, showing parameters for spatial resolution (σ _ESF) and contrast (c).

Figure 3

Anthropomorphic head and body phantoms. Axial images from a diagnostic CT scanner (Siemens Somatom Definition Flash, Forcheim Germany) show the low‐to‐medium contrast inserts (~85–95 HU) simulating soft‐tissue (ranging −60 HU to −15 HU marked as foreground, F) against uniform background (B) region. The reconstruction FOV for the O‐arm is marked by dotted circles, and an example elliptic fit for mitigation of truncation effects in the body phantom is marked with a dashed ellipse. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4

Geometric calibration and effect on image quality. (a–e) System geometry parameters over a 360° orbit. (a) Variations in SDD compared to the nominal value (1168 mm) assumed in an ideal circular orbit. (b) Displacement of the piercing point relative to (u, v) coordinates of the detector. (c–e) Excursions in 3D x‐ray source position. (f) FWHM of the PSF before and after calibration for the Ideal Profile (circular orbit), GeoCal (from a BB phantom), and Average GeoCal (average over 10 BB phantom calibrations). (g) Isosignal contours (at FWHM) of 2D Gaussian fit to axial images of the tungsten wire, showing a symmetric profile with reduced width for the Average GeoCal. (h–i) FBP₂ reconstructions of the temporal bone (f _c = 1.00).

Figure 5

Noise‐resolution characteristics evaluated in the head phantom. Each curve is formed by sweeping the f _c or β parameter for FBP or PL, respectively. The PLH method with a modest value of δ = 8 HU yields a superior noise‐resolution tradeoff without introducing unnaturally piecewise constant patchiness in the image. The inset in each axial image shows a zoomed region about a sphere of −80 HU contrast to background.

Figure 6

Noise‐resolution characteristics evaluated in the anthropomorphic body phantom. As in Fig. 5, PLH with a modest value of δ = 8 HU improved noise‐resolution tradeoffs without causing unnaturally piecewise constant patchiness in the image. The inset in each axial image shows a zoomed region about a sphere of −95 HU contrast to background.

Figure 7

Low‐dose imaging performance of FBP and PLH algorithms evaluated in the anthropomorphic head phantom. (a) CNR measured as a function of dose (double horizontal axis) for each algorithm at fixed spatial resolution (σ _ESF = 1.3 mm), where PLH was observed to achieve +30% CNR at 4.2 mGy over FBP₂ at 6.7 mGy. (b) Axial images of the head phantom for each algorithm at various low‐dose protocols.

Figure 8

Low‐dose imaging performance of FBP and PLH algorithms evaluated in the anthropomorphic body phantom. (a) The gains were more pronounced in the higher attenuating body phantom, where PLH achieved +90% CNR at 3.4 mGy over FBP₂ at 6.9 mGy. (b) Similar to Fig. 7, axial images demonstrate the improvement in the region of the liver for each algorithm at various low‐dose protocols.

Figure 9

Image quality evaluation of reconstruction methods for various anatomical sites in cadaver. Line patterns in the top row mark soft‐tissue fascia for which edge‐spread width (σ _ESF) was matched for each algorithm. [Color figure can be viewed at wileyonlinelibrary.com]