I joined after finishing my MSci Physics degree at Imperial College London in 2013. I am working in the Photoacoustic group under the supervision of Dr Ben Cox and Prof Paul Beard. My research is focused on quantitative spectroscopic photoacoustic imaging, which involves acquiring images at multiple wavelengths and using spectroscopic decomposition techniques to quantify the concentrations of specific tissue chromophores by exploiting their spectral characteristics. Thus the known differences in the absorption spectra of oxy- and deoxyhaemoglobin can be used to obtain a measure of blood oxygen saturation – an important physiological parameter related to a variety of pathophysiological processes such as tumour growth. As part of the MRes course I will be taking modules in programming, optics in medicine and cancer biology, which I think will be great preparation for my research.
Angela M. d'Esposito
I have completed both my BSc in Clinical Engineering and MSc in Biomedical Engineering in my home town at Sapienza, University of Rome. During my academic career I have had the chance to study at TU/e, Technical University of Eindhoven (NL), and later on to work on my final Master project at Berkeley, University of California. Learning much from these experiences from both a personal and an academic point of view, I have decided to continue walking in that direction by starting a PhD research at UCL, CABI (Centre for Advanced Biomedical Imaging). Here I am working on the development of an Optical Projection Tomography (OPT) imaging system to study fundamental biological processes in the heart and brain using light emitted from inside the organ, via optical fluorescence. OPT is a novel technology and represents the next generation of optical microscopy. Developments in this research will fill a technological gap, allowing 3-D imaging of cellular process in unprecedented detail.
Matthias Joachim Ehrhardt
My background is in applied mathematics in which I completed my Diploma at the University of Bremen, Germany, in 2011. Alongside my major subject mathematics the studies included industrial engineering and computer science as minor subjects. In my Diploma thesis I applied sparsity to decomposition tasks in geosciences. Since 2012 I am a PhD student at CMIC under the supervision of Professor Simon Arridge. My research interests comprise inverse problems, convex optimization, sparsity, and signal and image processing – in particular application of these techniques to medical imaging. In my research I work on simultaneous reconstruction of positron emission tomography (PET) and magnetic resonance imaging (MRI). As PET scans show cancer but lack good spatial resolution modern scanners obtain anatomical information from a built in MRI system which can be used to enhance the spatial resolution.
As a specialization of my education at Ecole Polytechnique engineering school in France with a major in applied mathematics I obtained a Master in Biomedical Engineering from ETH Zurich in March 2013. Strongly attracted by the translational motivation of CMIC and fascinated by the power of medical imaging, I spent the last 6 months of the programme at CMIC on a project that I will pursue during my PhD in collaboration with the DRC (Dementia Research Center). My research focuses on the study of the lesions of the white matter in the brain that can be observed in neurodegenerative diseases such as Alzheimer’s disease. Delineating, characterizing, quantifying and modeling these lesions in different images acquired with magnetic resonance imaging might indeed provide a deeper understanding of the processes of these diseases, link it to other assessments of dementia and avoid bias in other brain measurements.
I obtained an Electronic Engineering degree from a French graduate school in electrical engineering, computer science and telecommunications and an MSc in Biomedical Engineering from Imperial College London, in 2012. I am now a PhD student at the Centre for Medical Image Computing (CMIC). My research concentrates on the improvement of estimation of radio-pharmaceutical uptake in MR-PET imaging. While Magnetic Resonance Imaging (MRI) provides high-resolution anatomical information, Positron Emission Tomography (PET) provides functional information. The combined images have applications in oncology, neurology, or cardiology. Several factors limit the PET resolution such as photon count statistics and motion of the patient. My work during my PhD will be to improve an existing reconstruction algorithm, adding a new method of MR-based attenuation correction and a motion compensation algorithm.
I obtained my Bachelor’s degree in Electronic and Communication Engineering, and Master’s degree in Ophthalmology, both in the University of Hong Kong. During that time, I started to be fascinated by the prospective future of using medical imaging methods to investigate and tackle neuroscience problems. I then joined the DTP in 2011, and my research project involves joint collaboration between two centres in UCL: Centre of Medical Image Computing and Centre of Advanced Biomedical Imaging. My research project is to develop automatic segmentation method on mouse brain images and to identify fine structural differences or changes. The project is also part of the mouse phenotyping group project, which is to use high-resolution microscopic MRI to investigate microstructural changes in the mouse brain with the corresponding genomic variations or defects.
I am a neuroscientist because I am fascinated by the brain how little we know about how it works! I began studying for a BSc in neuroscience at Bristol University. Then I moved to UCL for a MSc in neuroscience. I was attracted to UCL as it is internationally acclaimed in neuroscience and has an outstanding community of scientists. I joined CABI (Centre for Advance Biomedical Imaging) with Mark Lythgoe to begin an MRI and optogenetics collaborative PhD project. Supervised by Alex Gourine I have developed in vivo physiology skills which enable me to transfect animals with light sensitive actuator proteins. When these proteins are stimulated with light, physiological effects can be monitored with functional MRI.
I came to UCL after reading Maths and work in Investment Banking. The aim of my project is to assess the damage to brain nerves from diseases. In the first year, the “MRes”, I compared models that describe better the permeability of the nerves, working with synthetic data to fine-tune the models, and with real Diffusion-MRI data to test their utility. This year I am working on ranking the models of Diffusion MRI. It exploits the natural diffusion of fluids inside the biological tissue to deduce the restrictions encountered by these fluids. There is a multitude of models, from parameters of which we can make inferences about the underlying microstructure. These inferences vary in accuracy and precision, largely based on the model which is used. So far, ex-vivo data has been used to build a hierarchy of the models. I will follow a similar route, but with in-vivo data.
I joined CMIC in 2011 after completing a BSc in physics at King’s College London and an MSc in computer science at Imperial College. During my DTP I will be looking at applying techniques of computer vision to the field of robot-assisted surgery. Research in this field aims to overcome the challenges surgeons face in terms of tool localisation and scene visualisation within the patient’s body. This can be achieved through the visual tracking of both the camera and the instruments as well as performing reconstruction of the surrounding surfaces. One of the research goals I hope to achieve is to use this data to build a system that is capable of assessing the skills of trainee surgeons. This will allow their ability and progress to be measured in quantifiable way.
After finishing my Dipl.-Ing. (FH) and Master’s degree in image engineering at the University of Cologne, I worked at the Research Centre Juelich and became highly interested in medical image registration. That’s way I joined UCL and the DTP programme at CMIC in 2010. In breast cancer treatment the number of breast conserving surgeries are increasing. Dynamic Magnetic Resonance Imaging is used to localise and stage lesions, as well as to provide important information for the surgeon with respect to planning and excision margins. However the patient position is completely different during imaging and surgery. We are aiming to recover the deformation which the breast undergoes due to the position change of the patient from image acquisition (prone) to surgery (supine). To solve this large scale deformation problem we build and integrate bio-mechanical models and image registration and finally will make the registration approach physically more realistic.
After completing my undergraduate master’s degree in Medicinal Chemistry at UCL, I wanted to research in a more pre-clinical environment. I joined the Doctoral Training Program in 2010, collaborating between the Radiochemistry department and the Centre for Advanced Biomedical Imaging. My research focuses on the development of radiotracers for targeting Voltage Gated Sodium Channels (VGSC). Results from numerous studies have implicated VGSC expression in the pathophysiology of various diseases such as Multiple Sclerosis and Epilepsy. Development of VGSC radiotracers may allow the use of nuclear imaging to quantify changes in the Na+ channel expression using SPECT/CT imaging. In doing so, this may enable diagnosis of disease, monitor response to treatment and study the pathological processes of a wide range of diseases.
I graduated with a masters in Engineering Mathematics from the University of Bristol before working as a research assistant at CMIC, focussing on vessel-based image registration techniques. I then worked for a private company dealing with imaging for clinical trials. I am now studying for a PhD in motion correction in simultaneous Positron Emission Tomography (PET) and Magnetic Resonance imaging (MRI). The project is split between The centre for Medical Image Computing (CMIC) and the institute of nuclear Medicine (INM) at University College Hospital (UCH). MRI provides high resolution anatomical information while PET provides functional information, but the images are of lower resolution. My work will focus on exploiting the fact that both imaging modalities can be acquired simultaneously on the new Siemens mMR scanner, and motion information found in one can be used to correct the other.
Before coming to the UCL Centre for Advanced Biomedical Imaging, I completed four years of the Natural Sciences Tripos (BA & MSci.) at Cambridge University, where I specialised in experimental and theoretical physics. It was during my final year that I developed an interest in biological and medical physics. During my DTP I will be investigating resting-state fMRI as an early biomarker for Alzheimer’s Disease, and testing this with a transgenic mouse model. There will be many physics and neuroscience experiments, combined with a number of data analysis techniques to map correlations between brain areas and establish functional networks. It may be that subtle differences within these networks can provide an insight into how Alzheimer’s Disease develops, and potentially lead to a diagnostic tool in humans.
Hi I’m Frank, I studied Physics (Solid State, Quantum & Medical) at King’s (MSci) before coming to UCL. I’m from Germany originally and like beer-houses :) I very much recommend making use of the great choice of courses available at UCL and can recommend Anatomy and Scientific Computing, as well as Advanced Imaging P3 if you are going to do MRI – as they teach elementary image processing skills. My DTP project is on pushing 23Na-MRI in the brain to derive quantitative measures of sodium in healthy and diseased (Multiple Sclerosis) human brains. I like my project as it encompasses developing sequences, acquiring, processing and analysing data as well as fitting and designing models.
I completed my Bachelor degree in Physics and Earth and Space Sciences at Jacobs University Bremen in Germany and wrote my Physics Bachelor thesis in cooperation with Fraunhofer MEVIS on a project dealing with compartment modelling of the liver and deconvolution techniques based on perfusion and hepatic extraction. I joined the DTP programme at UCL in 2012 as an MRes student. Currently I am working in the field of Neonatal Imaging, on a project dealing with MRI brain data of 18 to 28 weeks foetuses. The aim of the project is to perform segmentation and registration of the images in order to investigate the development of the early human brain. This information is crucial for the early prediction of brain abnormalities.
My background is in Physics for undergraduate’s degree and Computer Science/Electrical Engineering for master’s degree which I obtained both in France. Before joining UCL, I did a research internship in signal processing at CNRS in Nice, France, where I worked on denoising of astrophysical hyperspectral data. My research at UCL involves the use of novel signal and image processing methods known as compressed sensing and sparsity in the field of dynamic medical images. This is closely related to topics such as inverse problems, optimisation and in particular image reconstruction from undersampled data. General aim is to help in the acquisition, reconstruction and/or interpretation of medical images with potential benefits for clinicians and/or patients. For example, compressed sensing has been applied in MRI to speed up scan time and in CT to reduce radiation dose.
I came to the DTP in 2011 from a background in physics. I am looking for ways to improve the early diagnosis of dementia through the combination of data from different imaging modalities. This year I shall be working between CMIC the INM at UCLH. My research here aims to improve and create computer aided diagnosis tools using techniques from computer vision. Right now, I am using exploring a novel way to combine PET and MRI data to recognise changes associated with disease progression in Alzheimer’s. Using pre-existent data and random forest classifiers, a machine learning technique, I produce similarity measures between the data points associated with different cases of disease or health. These measures can be used to map diagnosed and undiagnosed data points to a new space where machine learning techniques may be used more effectively.
My joint interest in neurobiology and biomedical imaging has led me to pursue a research role in neuroimaging. I particularly like studying diffusion MRI and what this tells us about the nerve pathways of the human brain. For the first year of my PhD I am doing the MRes in Medical and Biomedical Imaging, working partly on gaining new skills in computing and neuroanatomy, and partly on my PhD project, which is a joint collaboration between CMIC, the Institute of Child Health and the Dementia Research Centre. My project will initially involve testing various methods for obtaining measures of structural connectivity using diffusion-weighted images and for analysing the network using graph theoretical analysis. The method of choice will be developed as a novel clinical tool for studying language development in children and for neurodegenerative diseases such as Alzheimer disease.
Miguel Rodrigues Gonçalves
I joined the DTP in 2010 after finishing an MSc in Biomedical Engineering in Portugal. Over the last year of my Master’s I became very interested in the medical imaging field and so I applied to the UCL Centre for Advanced Biomedical Imaging, where I am based at currently. My project aims at characterising the microenvironment of tumour xenograft models and evaluating characteristics associated with acute hypoxia. Acute hypoxia is a characteristic of solid tumours and is associated with resistance to chemotherapy and radiotherapy, as well as contributing to tumour progression and development of metastatic disease. Oxygen fluctuations can be measured non-invasively with MRI using gradient-echo sequences sensitive to the changes in oxygen and blood flow. These findings will be used to construct mathematical models of tumour growth and progression, with a view to applying corresponding techniques in clinical trials of radiotherapy efficacy.
After I finished my degree in Chemistry at the University of Liverpool, I wanted to broaden my horizons and pursue my Masters in a more varied, multidisciplinary area. I went on to Imperial College to study Bioimaging Sciences, where I fell in love with MRI. My decision to study at UCL was largely shaped by the facilities and projects available at the Centre for Advanced Biomedical Imaging (CABI), where I am now based. The initial MRes year will expose me to different areas of research as CABI works with collaborators across a range of disciplines. I will also be working towards optimising in vivo imaging of neurodegenerative mouse models with respect to the 3Rs; this will form the foundations of my work for the next 4 years.
Stian Flage Johnsen
My academic background is in computational science, computer graphics, and medical imaging. I received a first MSc degree in computer science from ETH Zurich in 2008, and a second one in medical image computing from UCL in 2010. Since November 2010, I have been employed by CMIC as a research associate, at first on a temporary basis, and since February 2011 on a permanent contract. I also enrolled as a part-time PhD student with UCL in February 2011, carrying out research in the area of simulating biomechanics for aiding surgical image guidance at CMIC. This entails dealing with biomechanical simulation by means of FEM, contact detection and modelling, and registration problems. This research has given me the very unique opportunity to combine what I learned in both my MScs, and apply it to a range of challenging and interesting problems.
Albert K. Hoang Duc
I join the PhD program at UCL in 2010 after working in various universities in the UK, Canada and the USA. I completed a Master’s degree in Computer Science in France and a Master’s degree in Neuroscience in the UK. Radiation therapy of cancer requires the segmentation of organs at risk on various imaging modalities in order to maximize the dose received by the target tumor while controlling the dose received by the surroundings organs at risk. Manual contouring can provide these delineations but is dramatically time consuming and prone to inter-expert variability. My research focuses on applying machine learning techniques for the development of novel atlas-based algorithms to obtain automatic segmentation of organs at risk in the liver and head/neck regions that produce fast, accurate and consistent results. The acquisition of large quantities of imaging studies prior to, during and after treatment may be a significant challenge for hospital logistics but offers a great opportunity for large scale analysis and data mining.
Even at the end of my Natural Sciences degree, I still wasn’t sure about doing a PhD; but the Biomedical DTP has proved to suit me perfectly. The initial training year helped me find my feet, and I took various courses including computer programming, cancer biology and animal handling. These prepared me well for my current multidisciplinary research, which encompasses optics, acoustics and cancer biology and entails activities such as computational modelling, laboratory experiments and small animal studies. The subject of my project is pre-clinical characterisation of tumours using photoacoustic imaging, and so far I have been making Doppler velocity measurements on blood-simulating phantoms. Sometimes it feels as if nothing works, but the sudden breakthroughs keep me going! I was able to present some of my results at two international conferences in San Francisco.
I enrolled in the DTP program in 2009 after spending a decade working in the computer industry. My project involves the use of MRI images acquired during a neurosurgical procedure to accurately localise brain structures and disease lesions during surgery. The main
focus of my work has been enhancement of fast non-rigid registration techniques developed at UCL to allow for multi-modal image registration using both structural and diffusion MRI images. We have successfully applied this technique on temporal lobe epilepsy cases. In addition, I am working on novel EPI distortion correction schemes that can be used on interventionally acquired Diffusion Weighted MRI images. My other research interests include discrete optimisation techniques, estimation of uncertainty in optimisation processes and visualisation of higher dimensional data.
I started the DTP programme in September 2009 to pursue my interest in Medical Imaging. My project researches cell based therapies for cancer (for example, using the bodies white blood cells to fight tumours) and how to label and track cells in the body over long periods of time. We are particularly interested in lung cancer because it is the commonest cause of cancer death in the UK, accounting for 22% of all cancer deaths. I graduated from UCL Medical School in 2007 and during my medical studies I took a year out to gain an Intercalated BSc in Medical Physics and Bioengineering, doing a research project in X-ray diffraction. After graduating, I worked as a junior doctor at King’s College Hospital and then at the Medway Maritime Hospital, where I did rotations in Accident and Emergency, Haematology and Obstetrics and Gynaecology, completing my foundation training in August 2009.
I started my DTP at UCL in 2009 after graduating in Information Engineering at Politecnico di Milano. My project, which springs from the collaboration between the Centre for Medical Image Computing and the Institute of Nuclear Medicine at UCH, researches the integration of multi-modal imaging techniques to observe the state and progression of brain disease. In the last two years I have investigated unified computational models of the uptake of radio-labelled pharmaceuticals in the brain tissue and of PET and MR image acquisition. I am currently investigating the use of such unified models for compensation of motion of the patient during the acquisition of PET and SPECT scans and to improve the quantification of pharmaceutical uptake in the upcoming MR/PET imaging systems.
Originally a biologist by degree I have moved into diffusion MRI and microstructure measurement. My focus can be categorised as basic science and has required constant development of unique and novel techniques and methods for working with UCL’s high field MRI systems. I design and build custom parts and equipment for both my work and my collaborators around UCL. I also work on ‘Science & Art’ projects with artists who have spent time working in UCL’s Centre for Advanced Biomedical Imaging.
I joined the DTP in 2009 after completing an MPhys degree at the University of Manchester. My project is based on ‘global tractography’, which involves the construction of whole-brain white matter connectivity maps from a large-scale optimisation in which the entire connectivity model is solved simultaneously. This utilises diffusion weighted magnetic resonance imaging, which gives directional information about tissue structure in highly coherent configurations like the bundles of axons in white matter. Currently established methods solve connectivity one connection at a time, tracking locally from one voxel to the next. Due to the complexity of white matter structure at scales much smaller than the typical image resolution, these locally based techniques can be easily confounded at bifurcations and divergences. Global tractography has the potential to improve on the accuracy of MRI-derived brain-connectivity maps which can be of great importance to inform our understanding of brain function and pathology.
I began the DTP program in 2009, having spent a couple of years working in industry since completing my Master’s degree in Computer Science at UCL. My project involves using predicting the onset of dementia in elderly people both with Mild Cognitive Impairment, a transitional state between healthy ageing and Alzheimer’s disease, and eventually in subjects showing no symptoms. This is done using image processing tools developed by my colleagues at CMIC to extract features that are predictive of dementia from structural MRI scans and other sources of data, such as patterns of atrophy in cortical grey matter or the shape of the hippocampus. Each subject brain is represented as a point in a very high dimensional space of these predictive features. State of the art techniques borrowed from machine learning are then used to separate groups of training subjects and classify unseen ones in this space.