I am a computational scientist with a doctoral and Master’s degree in computer vision and image analysis. My research is mostly focused on algorithm development for analysis and interpretation of medical and biomedical data. I have an extensive experience in building cutting edge computer vision technologies for object detection, semantic segmentation, image restoration, depth estimation, optical flow, image registration and 3D reconstruction. I have developed tools in these topics with both classical mathematical models and deep learning models. I am a passionate researcher with knowledge, energy and enthusiasm to contribute towards healthcare technologies that can benefit patients and assist clinicians around the world. I am a promotor of translational research and equality in access to digital healthcare for all. I am a flexible, considerate and humble human being.
I am currently a post-doctoral research associate at the Department of Engineering Sciencee, University of Oxford. My current research involves around building robust models for computer assisted endoscopy in gastroenterology, especially the oesophagus and the colon. The project is a part of NIHR Oxford biomedical research centre imaging theme on Computer aided endoscopy for Barrett’s oesophagus, a defined cancer precursor in the oesophagus, for which I am collaborating with consultant gastroenterologists from the Oxford University Hospitals. At Oxford, I am also co-supervising two DPhil students together with Prof. Jens Rittscher.
I am also working on developing novel deep learning methods for polyp detection and segmentation on tackling issues such as generalisability, minature networks and networks capable of dealing with small size data. To accomplish this, I am collaborating with Prof. Pål Halvorsen from the Simula Research Laboratory, Oslo, Norway, and senior consultant Thomas de Lange from Sahlgrenska University Hospital Gothenburg, Sweden. I am mentoring some students as part of this collaboration. I am also co-supervising one PhD student with Gilberto Ochoa Ruiz in surgical computer vision.
I have worked as researcher in different capacities in leading research teams in France, Spain, Germany, Switzerland and the UK. I am initiator and lead organiser of series of endoscopic computer vision challenge that started in 2019 (follow-us here). I am very thankful to all the collaborators for believing in me and assisting me in data collection, curation, and participating in discussions for planning and execution. I am keen towards learning about new technologies and open to meeting new people, establishing interesting and innovative collaborations (both academia and industry), mentoring students and looking forward for new and exciting opportunities.
Thanks to all colleagues at Oxford who have been always supportive of me and helped me grow. I am very thankful and lucky to have also received constant support from many other researchers especially Prof. Danial Stoyanov, Prof. Lena Maier-Hein and Prof. Adrien Bartoli.
“Tell me and I forget, teach me and I may remember, involve me and I learn.” – Benjamin Franklin
Post-doctoral Researcher @dkfz.de 2015 - 2018 I joined Biomedical and Computer vision group of Prof. Karl Rohr at the University of Heidelberg and DKFZ, Germany. During my work there I collaborated extensively with neuroscience researchers (led by Prof. Katrin Amunts) and physicists (led by Markus Axer) at Forschungszentrum, Jülich, Germany. I designed mathematically plausible phyics-based deformable multi-modal image registraiton techinque that allowed for precise 3D reconstruction of ultra high resolution (1.3mu) 2D histology brain images between CCD acquired image and high-throughput polarized light microscopy (more info here).
Research Associate - Doctoral thesis @univ-lorraine.fr 2012 - 2015 I obtained my PhD from the University of Lorraine supervised by Prof Christian Daul and Prof. Walter Blondel from CNRS - CRAN laboratory at Nancy, France, where I was involved in the development of robust computer vision algorithms for monitoring bladder cancer progression by accurately creating mosaics of endoscopic videos. The key element in this work was to enable computer to match reliable features based on frame motion and perform mosaicking of the surface under observation. (Thesis here)
Master’s in computer vision - Master thesis @univ-bourgogne.fr 2012
I did a master’s thesis between Oak Ridge National Laboratory, USA and Le2I, Le Creusot at the University of Bourgogne supervised by Prof. Fabrice Meriaudeau) and Thomas P Karnowski on “Retinal image analysis from fundus imaging”.
Contact: ali[dot]sharib2002 [at] gmail[dot]com