X-ray computed tomography (CT) is a powerful technique for imaging three-dimensional objects revealing the interior of a sample. X-ray CT is routinely used in medical imaging and industrial inspection, for example. Currently, contrast in X-ray CT images is largely based upon how strongly X-rays are attenuated by the sample. For example, bones in the human body strongly attenuate X-rays incident upon them, which is why bones, which are highly attenuating, appear with relatively high contrast in X-ray CT images. There is much interest in extending X-ray CT to be able to identify particular materials in a sample. For example, personalised diagnosis of complex diseases increasingly involves the use of imaging agents. Our method would provide the ability to pinpoint multiple agents simultaneously in a single scan, allowing for more effective diagnostic tests to be performed in a simple and speedy manner At present, material identification using X-ray CT can only be performed using two separate CT scans which therefore takes twice the amount of time of a single CT, thus delivering double the radiation does to the sample. Furthermore, this approach only works for monochromatic X-ray source, such as are available at synchrotrons, which are expensive to operate and have very limited capacity. Even if synchrotron access is possible, in many applications, such as medical imaging, performing two scans is not feasible due to the dose they can safely receive. More recently, methods which make use of expensive detectors can achieve this using a single scan, however severely limiting the sample size. We propose a method of performing material identification which uses a single phase-sensitive CT scan and is based on the edge illumination x-ray phase imaging technique which employs standard, therefore cost effective, x-ray imaging equipment. Our technique works by recognizing that three-dimensional images of electron density and effective atomic number, reconstructed from a single edge-illumination CT data set will only have sharp interfaces between two materials when both of these materials have been correctly identified. Thus, the unknown material can be determined by varying a reconstruction parameter and inspecting an interface between the known and unknown materials. The parameter which leads to this interface becoming sharp essentially reveals the unknown material. This phenomenon can be used sequentially to identify all materials in a sample, resulting in complete material identification using a single CT scan.

Oesophageal cancer has been identified by CRUK and other institutions as a "cancer of unmet need"; survival rates have not improved significantly for decades. In England and Wales, these are of only 12% for 10 or more years. Pitfalls can be identified at all stages of surveillance/staging/treatment (referred to in the following as "streams" 1-3). We believe that highly sensitive, in-room, real-time imaging with microscopic resolution could address pitfalls in all streams, consequently improving treatment outcomes. The objective of our project is to increase life expectancy and survival rates of patients with oesophageal cancer by developing novel imaging tools for in-room analysis that can guide treatment effectively. Stream 1 refers to biopsies collected at endoscopy, which are sent to the histopathology lab for expert evaluation using a tissue staining method that has not changed for 100 years. The turnaround time is up to one month, leading to delays in scheduling therapy during which pre-cancerous lesions can develop into cancer. Real time analysis as they are collected will allow for immediate in-room therapy, significantly improving patient management. Stream 2 refers to Endoscopic Mucosal Resections (EMRs) which may be sufficient treatment for complete cure of early stage cancers. The crucial questions are whether the entire tumour has been removed, or whether the tumour has extended deep into the wall of the oesophagus. This would change the treatment pathway. Real-time analysis would allow for immediate repeat EMR if needed, minimising the number of patient having to return later (which makes repeat EMR harder to perform); it would also reduce the number of patients requiring major surgery to remove the entire oesophagus (oesophagectomy). Stream 3 refers to oseophagectomy, undergone by patients with locally advanced cancer that has not spread beyond the local area that can be removed surgically. Success depends on achieving clear margins: for this, frozen sections are collected and analysed through a procedure that can take over an hour, while our technology would reveal this in real time. Another need is identification of the number, position and infiltration state of surrounding lymph nodes. An insufficient number of lymph nodes is sometimes collected, which is not found out until later - again with implications for treatment pathways (e.g. need for chemotherapy after surgery). Our team has developed a new approach to x-ray imaging called x-ray phase contrast imaging (XPCI). It uses a different physical principle (refraction and interference) to generate image contrast, instead of x-ray attenuation which is what every system in existence has been using since Roentgen. Thanks to this, XPCI can reveal features considered invisible to conventional x-rays, notably faint structural changes in soft biological tissue. We have already proven that XPCI, unlike conventional x-rays, has sufficient sensitivity to distinguish between layers of the wall of the oesophagus, which is very relevant to this project. We have also demonstrated that XPCI can perform full 3D ("computed tomography") scans in minutes, and reach resolution of 1 micron while using conventional x-ray sources. We believe that targeted implementations of XPCI can fulfil the needs of real-time analysis for all above streams (albeit possibly through two separate instruments with different field-of-view and resolution), and we have assembled a team of engineers, physicists, clinicians and industrialists to tackle this problem. Engineers and physicist will design and build the imaging systems using input from the clinicians; the systems will be used to image a sufficient number of specimens from all streams to allow drawing significant conclusions on the clinical benefits. The industrialists will oversee the process to ensure compatibility with industrial processes and regulatory compliance, and ultimately take the research into clinical exploitation.

Tissue engineering - aimed at developing "lab-grown" organs and tissue by combining appropriate scaffolds and cells - could solve one of the biggest medical problems of our times, the shortage of donor organs. While the pool of scaffold materials is large (e.g. natural/synthetic biomaterials), there is consensus that the extracellular matrix (ECM) of the target tissue is an excellent choice as it possesses native structural and biomechanical properties. ECMs can be derived from cadaver tissue (e.g. from animals) through a process called decellularization, by which the tissue undergoes several cycles of flushing with detergents and enzymes. A successfully decellularised tissue is characterised by the absence of cellular material and the presence of an intact ECM. Imaging, for assessing the ECM, is an extremely important tool for the development of decellularisation methods that are simultaneously gentle and effective. This project is about developing a new imaging tool for characterising decellularised tissue based on x-ray micro computed tomography (CT). Since micro-CT is a non-destructive technique, the inspected samples can be used further in longitudinal studies or be implanted into animals to test their performance in vivo. In comparison, the current gold standard techniques for inspecting ECMs (histology, electron microscopy) require that samples are sliced, sectioned and/or stained in preparation for being imaged, prohibiting using them in any further studies. A number of substantial developments will be needed before micro-CT can become a valuable tool for validating decellularisation techniques and other methodologies in tissue engineering. Currently, micro-CT fails to meet the complex imaging needs of this field, which often requires multi-scale and multi-contrast approaches. First, a micro-CT machine with zooming in capabilities would be required to inspect the multi-level structure of ECMs. Second, decellularised tissue generally exhibits weak x-ray attenuation; hence, the micro-CT machine should provide access to phase contrast alongside attenuation contrast, which is known to provide a much better visualisation of tissue scaffolds than the latter. The micro-CT machine proposed here will have both these functionalities. It will exploit an innovative imaging mechanism that is underpinned by the idea to structure the x-ray beam into an array of narrow (micrometric) beamlets via a mask placed immediately upstream of the sample. This provides flexibility in terms of spatial resolution, as this metric - unlike in conventional micro-CT scanners - is not defined by the blur of the source and detector. Instead, resolution is driven by the beamlet width, which can be made smaller than the intrinsic system blur, bearing unique potential for fast resolution switching and multi-scale imaging. Second, it provides access to complementary contrast channels (phase, ultra-small angle x-ray scattering). These channels result from small x-ray photon deviations which occur alongside attenuation when x-rays interact with matter. While most conventional micro-CT scanners are blind to these effects, the machine proposed here will enable their detection, allowing to reconstruct three sets of complementary tomographic images for each sample. While the phase channel can provide a much higher contrast-to-noise ratio than the attenuation channel, the ultra-small angle x-ray scattering channel encodes the presence of sub-resolution features in a sample. The latter bears unique potential for image-guided zooming in. The project will culminate in the design, construction and test of an experimental prototype for image-guided multi-scale and multi-contrast imaging with a field of view of up to 10 cm by 10 cm, which may be expanded to larger dimensions in the future. A broad range of decellularised tissues will be scanned, and the results benchmarked against the current gold standard (histology or electron microscopy).

X-Ray Imaging (XRI) has a fundamental role in medicine and security, and is instrumental in the automotive, aerospace, pharmaceutical industries and in manufacturing in general. Cultural heritage relies on XRI, as do materials science, biology, and many other scientific fields. Through our established collaboration between Nikon X-Tek Systems (NXTS, Nikon's UK based x-ray division) and UCL, we are targeting the next paradigm shift in XRI. Our vision is that this will involve the incorporation of phase effects in the image formation process ("Phase-based" XRI) coupled with energy-resolved ("colour") XRI and new data reconstruction and interpretation algorithms. "Colour" XRI could be seen as the x-ray equivalent of the transition from black and white to colour photography, meaning a much wider spectrum of information can be obtained from the imaged sample. Phase-based XRI enables contrast increases of up to two orders of magnitude, thus allowing the detection of features classically considered "x-ray invisible". Our vision is to marry UCL's world-class research and expertise on phase-based XRI, inverse problems and nanofabrication with NXTS's innovation on scatter analysis, image reconstruction and colour x-ray imaging in order to achieve the next step change in XRI technology, with the UK industrial and academic communities firmly at the centre. This will deliver transformative solutions that are practicable in an industrial context and beneficial to a wide user base, while also enabling new science. Our ambition is to replace conventional attenuation based XRI with energy-resolved, phase-based technology combined with scatter retrieval and novel algorithms in most application areas. At synchrotron facilities, UCL researchers have used phase-based XRI to image rocks, metals, tissues, animals, humans, cells, foams, fabrics, batteries, manufacturing processes, food, and heritage artefacts. They have done this statically and dynamically, in situ and in operando, in vivo and ex vivo, invariably detecting key features that were invisible to other methods. Making this available through standard, lab-size machines would be nothing short of a revolution, leading to economic and societal impact through the multi-disciplinary applications, making NXTS the commercial leader in the field, and cementing UK's leading research status. In our vision this will be strengthen even further by its combination with "colour" imaging, and with new ways of handling scattered radiation such that the "structured" scatter signal leading to additional information is exploited, while the uniform background that limits image contrast and therefore detail visibility is rejected. We will pursue this vision through a combination of modelling and experimental work. Using experimentally validated simulation software developed jointly by the UCL and NXTS teams, we will model experiments before they are carried out, compare simulated and experimental results, refine models and setups until all discrepancies are clarified, and only then proceed to the next step. This will enable us to develop systems where i) we keep all parameters under control and have full understanding of their effects and implications, and ii) we can steer the design towards effective solutions to specific problems. Cutting-edge nanofabrication methods (available at UCL's Photonic Innovations Lab and London Centre for Nanotechnology) will enable the development of beam modulators that allow the exploitation of phase effects with the conventional x-ray sources routinely used by NXTS. We will apply the novel technologies to a range of high-impact applications, including non-destructive testing of composite materials and additive manufacturing processes and products, biomaterials and tissue-engineered organs, digital histology, improved detection of concealed explosives and forensics.

X-ray imaging technology is increasingly used for specialised non-destructive testing and measurement applications, especially in high value and safety/performance critical industries, such as defence or aerospace. Current tomographic methods, which can image detailed internal geometries, require x-ray images to be taken from angles evenly spaced around the object under investigation. However, problems arise because 1) x-ray penetration depth is limited and 2) different materials absorb x-rays in varying amounts. Thus, large objects such as larger panels or assemblies with materials with widely varying x-ray absorption profiles can thus often not be imaged with these techniques, even if x-ray penetration is possible in some directions and through some materials. The inspection of many composite components encountered in aerospace structures is not currently feasible with traditional x-ray imaging techniques, either due to their high aspect ratio or due to the presence of highly absorbing metal structures. Yet owing to their safety critical function, such components often need to be inspected regularly for internal defects using non-destructive testing methods. Current non-destructive inspection approaches, however, are often not applicable. For example, ultrasound testing can often not be applied to many advanced composite materials whilst thermography has limited depth capability. For such composite materials, x-ray based techniques remain one of the few alternatives if these materials are to be used to their full potential in future aerospace design. Building on a current project in which we are developing x-ray imaging hardware that allows us to scan flat panel carbon fibre structures with unconventional scan trajectories, the current proposal aims at the development of new computational methods to recover the three dimensional x-ray absorption profile from the x-ray measurements made with our new system. Current experiments using traditional methods have shown that available x-ray tomography reconstruction algorithms perform poorly in scan settings where a component cannot be scanned from all directions. This is due to a fundamental physical limitation of the system, which is just not able to acquire enough information to allow good reconstructions. This fundamental limit can only be overcome if we can introduce additional information into the reconstruction. For nearly all modern manufactured components, such information is luckily available in the form of detailed CAD drawings and it is this information we propose to use to overcome the x-ray tomographic reconstruction problem. In particular, we will address the following problems. 1) Design a reconstruction algorithm that uses CAD drawings as prior information in order to directly estimate object boundaries. 2) Allow the estimated surface to be close to, but not necessarily identical to the surfaces predicted by the CAD model. 3) Only once object boundaries have been estimated, use a second reconstruction step to estimate the spatial distribution of x-ray absorption. 4) Use a multigrid approach to increase computational efficiency of the method To design lighter, cheaper and more efficient airplanes, many components need to become lighter and cheaper whilst still satisfying strict performance standards. This can often be achieved using modern composite materials. As the use of these components is safety critical, reliable inspection methods are required if these components are to be used. Our developments will enable this.