Ultrasound modulated tomography for high sensitivity, high spatial resolution 3D imaging
Light offers an alternative to techniques such as X-rays, CT, PET or MRI for imaging body tissues. Optical techniques are useful because they provide rich functional information and are non-hazardous within the average and peak intensity limits. One of the main drawbacks of using light is that it is heavily scattered by tissue and so images often have poor spatial resolution. The outlined research aims to develop a new imaging technique that combines bioluminescence optical imaging (BLI) with ultrasound (US) that will offer significant improvement in spatial resolution over conventional BLI. This is important in pre-clinical imaging of small animals where more accurate imaging techniques will help to reduce the number of animals used in such experiments. The improvements in spatial resolution will be achieved in two ways; firstly, by modulating the bioluminescent light emitted within the tissue using focused US beam to produce a modulated light ‘beacon’ in the region of the US focus that reduces the effects of light scattering and improves the spatial resolution; and secondly by using the US image to inform a reconstruction algorithm. Based on our concept we anticipate that 3D image spatial resolution will be improved by at least a factor of 5 (500 μm compared with 2.5 mm) enabling more accurate and consistent clinical data to be obtained from a smaller set of animals. The potential of system to outperform current BLI system will be demonstrated through a number of exemplar pre-clinical studies, including tracking of mesenchymal stem cells in nude mice.
We have developed a novel small animal US mediated BLI platform (Fig. 1a) in which US (Verasonics scanner) is applied from the underside of the animal and modulated bioluminescence signals can be detected using a photomultiplier tube (PMT). In addition a CCD camera is also used to record the conventional optical image for the image registration. Scanning the US and recording the modulated bioluminescence signal at each point allows a 3D structural and bioluminescence image to be obtained.
In initial experiments, tissue like ‘phantoms’ of known optical and acoustic properties embedded with low illumination sources are used to mimic small animal experiments in order to optimize the system with respect to spatial resolution and signal to noise ratio (SNR). Higher spatial resolution can be achieved by applying pulsed US with time gating and increasing the US frequency. This will inevitably reduce the magnitude of the US modulated signal due to reduction in the US focal volume. Fig 1(b) shows the improvement in spatial resolution using US modulation (AC) compared to the conventional unmodulated light (DC), whereas fig. 1(c) shows a plot between SNR and modulation depth at different surface radiances.
This research project will contribute to a significant impact on the 3Rs. Replacement: better imaging will inform more accurate computational models; Reduction: imaging enables longitudinal studies on the same cohort, more accurate quantitative imaging allows fewer animals to be used in a study; Refinement: through the improved quality of research findings. This research has been funded by the National Centre for the Replacement, Refinement and Reduction of animals in research (NC3Rs).
An imaging system that combines ultrasound and optical techniques will be developed with the capability to significantly improve the spatial resolution and quantitative accuracy of bioluminescence imaging (BLI). Enhancement of the information obtainable from current BLI systems will be achieved by first modulating the bioluminescent light emitted in the tissue with a focused ultrasound beam. This produces an ultrasound modulated light 'beacon' within the tissue in the region of the ultrasound focus which can be used to reduce the effects of light scattering and improve the image spatial resolution. To provide an added layer of information conventional ultrasound imaging will be carried out enabling images of tissue structure to be co-registered. Both modulated light beacons and structural information will inform a reconstruction algorithm based on our widely used NIRFAST reconstruction code enabling the two datasets to be merged. The potential of the system to outperform current state of the art BLI will be demonstrated through a number of exemplar pre-clinical 3D imaging studies, including tracking of mesenchymal stem cells in nude mice. Based on our proof of concept data the 3D image spatial resolution will be improved by at least a factor of 5 (500 µm compared with 2.5 mm) enabling more accurate and consistent clinical data to be obtained from a smaller set of animals. This will contribute to a significant impact on the 3Rs. Replacement: better imaging will inform more accurate computational models; Reduction: imaging enables longitudinal studies on the same cohort, more accurate quantitative imaging allows fewer animals to be used in a study; Refinement: through the improved quality of research findings.
This project aims to reduce the number of animals needed in long time course studies by combining ultrasound with bioluminescence imaging to improve 3D spatial resolution and maximise the quantitative data that can be obtained from each animal. Background Traditional approaches for long term in vivo studies, for example, to track disease progression require groups of animals (up to ten) to be killed at defined time points and tissues removed for analysis (e.g. cell count, pathogen numbers). Longitudinal imaging offers an opportunity to reduce animal use in these studies, but due to optical scattering, current state of the art optical imaging systems lack quantitative accuracy and spatial resolution. This project will combine ultrasound with bioluminescence imaging (BLI) to create a new imaging platform to enable non-invasive, long term longitudinal imaging within the same group of animals. Combining these technologies will improve spatial resolution of the images, increasing the amount and quality of the data generated using this approach. A traditional six time point experiment would use 18-60 animals, but the current approach would require up to eight animals. Multiple imaging of the same animal throughout an experiment allows long term studies to be followed accurately with less variability.
Research details and methods
This project will build on initial efforts to combine ultrasound and BLI to develop a small animal ultrasound-mediated BLI platform capable of providing (i) ultrasound images of the tissue structure; (ii) 3D maps of bioluminescence from the ultrasound modulated signals; and (iii) unmodulated bioluminescence images. Professor Morgan will collaborate with Dr Hamid Dehghani at the University of Birmingham to modify the widely used NIRFAST software for reconstructing images to allow quantitative mapping of the signals from the new BLI platform. The development of the imaging system and the accuracy of the reconstruction algorithm will be tested in specially constructed tissue phantoms with known optical and acoustic properties. The validity of the system for in vivo imaging will be assessed in a variety of in vivo experiments already ongoing within the Morgan laboratory. This will include tracking mesenchymal stem cells in vivoand imaging bacterial infection/colonisation of indwelling devices, such as catheters.