Image analysis is the extraction of useful information from digital images and has applications in many fields from astronomy to zoology, including biology, medicine and industrial inspection.
In the Centre for Imaging Sciences, physicists, chemists, computer scientists, bioscientists and clinical researchers work together to develop new methods and apply cutting edge imaging and computational techniques for the understanding of disease, its management and treatment.
We help make the best use of the University's extensive imaging facilities (including Magnetic Resonance Imaging and Positron Emission Tomography equipment, a cyclotron and radiochemistry facilities and extensive bio-imaging equipment). We develop and apply novel computer algorithms to understand and interpret medical and biological imaging data.
Tim develops statistical models of both shape and appearance, which have proved very useful for interpretting images of many different kinds. He has pioneered novel algorithms (such as `Active Shape Models' (ASMs) and `Active Appearance Models' (AAMs)) which use such models to find the outlines of structures in images. These have many applications, including locating bones and organs in medical images, for face and gesture recognition and for industrial inspection.
Tim has a particular interest in musculoskeletal applications with projects aiming to identify people with osteoporosis (www.stopfrac.com), measure bone shape (www.bone-finder.com) and understand how best to monitor and treat osteoarthritis. For a more information see his list of current projects.
His research often involves assessing the basic principles on which the subjects of computer vision and image analysis are founded. Recent research has involved; developing a statistically self-consistent solution to the problem of analysing point based shape models for genetics, the first fully quantitative quantitative pattern recognition system, and methods for calibration of MRI based diffusion measurement for clinical practice. All work is done using principles of quantitative use of probability backed up with Monte-Carlo testing, generally bootstrapped from real world data samples.
Some of Neil's recent work included designing a new approach, Linear Poisson Models with which to analyse MR imaging data, to assess the volume of a tumour which responds to treatment in pre-clinical cancer trials. The method generated an improvement in statistical sensitivity of a factor of sixteen over a conventional T-test in the same data.
School of Computer Science
Chris is Director of Manchester Informatics, and has been a leading figure in health informatics in the UK for over 15 years. He has also been at the forefront computer vision research for over 35 years with some of the most highly cited publications in the field and a strong record in technology transfer. His core research is in computer vision and medical image analysis – with a central interest in developing generic methods to underpin practical applications in medicine, industry, and commerce. His interests in image analysis cover Mammography, Bone and Joint imaging (OA, RA), Nailfold Capillaroscopy along with Facial Recognition/analysis.
Tingting focuses on developing advanced mathematical modelling and large-scale optimisation techniques to (1) simulate human intelligence and (2) analyse real-world complex data. For (1), she aims at constructing effective machine learning models to automate tasks such as matching, recognition, prediction, ranking, inference, characterisation, language and vision understanding. For (2), she develops algorithms to discover latent structure and extract information from large-scale, noisy and unstructured data, e.g., text, image, video, signal and network data to support development of text mining systems and other related research areas such as bioinformatics.