: As cone-beam CT(CBCT) systems dedicated to various imaging specialties proliferate, technical assessment grounded in imaging physics is important to ensuring that image quality and radiationdose are quantified, understood, and justified. This paper involves technical assessment of a new CBCTscanner (CS 9300, Carestream Health, Rochester, NY) dedicated to imaging of the ear and sinuses for applications in otolaryngology-head and neck surgery (OHNS). The results guided evaluation of technique protocols to minimize radiationdose in a manner sufficient for OHNS imaging tasks.Methods
: The technical assessment focused on the imaging performance and radiationdose for each of seven technique protocols recommended by the manufacturer: three sinus protocols and four ear (temporal bone) protocols. Absolute dose was measured using techniques adapted from AAPM Task Group Report No. 111, involving three stacked 16 cm diameter acrylic cylinders (CTDI phantoms) and a 0.6 cm3 Farmer ionization chamber to measure central and peripheral dose. The central dose (D o ) was also measured as a function of longitudinal position (z) within and beyond the primary radiation field to assess, for example, out-of-field dose to the neck. Signal-difference-to-noise ratio (SDNR) and Hounsfield unit (HU) accuracy were assessed in a commercially available quality assurance phantom (CATPHAN module CTP404, The Phantom Laboratory, Greenwich, NY) and a custom phantom with soft-tissue-simulating plastic inserts (Gammex RMI, Madison, WI). Spatial resolution was assessed both qualitatively (a line-pair pattern, CATPHAN module CTP528) and quantitatively (modulation transfer function,MTF, measured with a wire phantom). Imaging performance pertinent to various OHNS imaging tasks was qualitatively assessed using an anthropomorphic phantom as evaluated by two experienced OHNS specialists.Results:
The technical assessment motivated a variety of modifications to the manufacturer-specified protocols to provide reduced radiationdose without compromising pertinent task-based imaging performance. The revised protocols yieldedD o ranging 2.9–5.7 mGy, representing a ∼30% reduction in dose from the original technique chart. Out-of-field dose was ∼10% of D o at a distance of ∼8 cm from the field edge. Soft-tissue contrast resolution was fairly limited (water-brain SDNR ∼0.4–0.7) while high-contrast performance was reasonably good (SDNR ∼2–4 for a polystyrene insert in the CATPHAN). The scanner does not demonstrate (or claim to provide) accurate HU and exhibits a systematic error in CT number that could potentially be addressed by further calibration. The spatial resolution is ∼10–16 lp/cm as assessed in a line-pair phantom, with MTF exceeding 10% out to ∼20 lp/cm. Qualitative assessment by expert readers suggested limited soft-tissue visibility but excellent high-contrast (bone) visualization with isotropic spatial resolution suitable to a broad spectrum of pertinent sinus and temporal bone imaging tasks.Conclusions:
The CBCTscanner provided spatial and contrast resolution suitable to visualization of high-contrast morphology in sinus, maxillofacial, and otologic imaging applications. Rigorous technical assessment guided revision of technique protocols to reduce radiationdose while maintaining image quality sufficient for pertinent imaging tasks. The scanner appears well suited to high-contrast sinus and temporal bone imaging at doses comparable to or less than that reported for conventional diagnostic CT of the head.
The authors extend particular gratitude to Dr. Jonathan Lewin and Dr. John A. Carrino (Russell H. Morgan Department of Radiology) for logistical support and valuable conversations on image quality and deployment of dedicated imaging systems in nonradiology medical specialties. Dr. Krishnamoorthy Subramanyan (Carestream Health, Rochester, NY) is gratefully acknowledged for expertise, technical assistance, and valuable discussion regarding operation and performance of the scanner. The work was supported in part by academic-industry collaboration with Carestream Health (Rochester, NY) preliminary to a clinical trial of the scanner in the Department of Otolaryngology-Head and Neck Surgery at Johns Hopkins University.
II. METHODS AND MATERIALS
II.A. The CS 9300 and default protocols
II.B. Dose measurement: Experimental setup
II.C. Imaging performance
II.C.1. High-contrast SDNR
II.C.2. Low-contrast (Soft-tissue) SDNR
II.C.3. HU accuracy
II.C.4. Spatial resolution
II.C.5. Image quality in an anthropomorphic head phantom
III.B. Signal difference to noise ratio and CT number accuracy
III.C. Spatial resolution
III.D. Image quality assessed in an anthropomorphic Rando phantom
IV. DISCUSSION AND CONCLUSIONS
- Medical imaging
- Medical image quality
- Spatial resolution
- Medical X-ray imaging
Data & Media loading...
Article metrics loading...
Full text loading...