Mobile benchtop in-vivo micro-CT imaging

2019 
532 Introduction: The current generation dedicated in-vivo micro-CT systems allows for fast and high-resolution imaging, but are still prohibitively large and heavy. We have developed a truly mobile bench-top in-vivo micro-CT system allowing for fast in vivo scanning with small form factor and low weight, retaining excellent image contrast, high image resolution and low dose capabilities. Our CT system was designed around four key components: a self-shielded X-ray source, a state-of-the-art CMOS x-ray detector with reduced axial length, a slipring for continuous rotation, and GPU based iterative image reconstruction. The small axial detector allows for extra x-ray beam collimation in the axial length, reducing the scatter-to-primary ratio. This leads to improved image contrast at the low x-ray energies commonly used in small animal imaging. At the same time, the irradiated field-of-view is kept to a minimum, which keeps the imaging dose low for focused scans during dedicated cardiac, lung or brain experiments. Because the small detector size limits the single-view FOV, a slipring mechanism is used to enable fully continuous rotations and helical scanning, allowing scans of whole body mice and rats without the need for image stitching. Iterative reconstruction with noise regularization is included by default, allowing further dose reductions by decreasing the tube current to levels impossible to attain with analytical reconstruction. Our small animal beds have fully integrated monitoring for ECG, respiratory rate and animal heating. A low-latency in-house developed monitoring system allows us to use prospectively triggered as well as data-based cardiac and respiratory gating. An automated quality control procedure is included to guarantee image quality day-to-day. It is based on a phantom specifically designed to determine image resolution, image uniformity, and low contrast detectability. All system functionality can be controlled over WiFi through an intuitive user interface available from workstation or tablets, as all electronics and processing power are embedded inside the same system with footprint of 56x56 cm and a weight just below 105 kg. Voxel sizes down to 10 um are possible. The best nominal resolution has been measured as 50 µm using a Fourier-based analytical approach based on [1] which is visually confirmed using the QRM bar phantom. Low contrast detectability is automatically scored from 9 low-contrast spheres with an NPWE model observer based on [2] (Fig 1). Thanks to regularized reconstruction, full body mouse scans can be executed in under 2.5 mGy radiation dose and under 15 seconds scanning time (Fig. 2). In conclusion, we have shown that it is possible to develop a truly mobile bench-top in-vivo micro-CT system with no compromise in image quality. Many users worldwide have used this CT system successfully in applications such as dynamic vascular contrast imaging, in vivo trabecular bone analysis, and gated imaging. The use of GPU-based regularized iterative reconstruction enables very low imaging doses, previously unfeasible with analytical reconstruction. [1] S. N. Friedman, G. S. K. Fung, J. H. Siewerdsen, and B. M. W. Tsui, “A simple approach to measure computed tomography (CT) modulation transfer function (MTF) and noise-power spectrum (NPS) using the American College of Radiology (ACR) accreditation phantom,” Med. Phys., vol. 40, no. 5, p. 051907, 2013. [2] I. Hernandez-Giron, J. Geleijns, A. Calzado, and W. J. H. Veldkamp, “Automated assessment of low contrast sensitivity for CT systems using a model observer,” Med Phys, vol. 38, no. 1, p. S25, 2011.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []