Jonathan Oakley to Give Tutorial: An Introduction to Image Processing in Optical Coherence Tomography at IEEE HISB on July 29th
Optical Coherence Tomography (OCT) was developed as a result of a multi-disciplinary research effort based around the Massachusetts Institute of Technology. Its first description was an article in Science, and the first instrument was studied in a clinical environment at the New England Eye Center and Tufts University School of Medicine, Boston. Since then, the modality has revolutionized ophthalmic clinical decision-making, but its applicability is not limited to the field of ophthalmology alone. Indeed, it is actively being adopted in cardiology, dermatology, dentistry, gynecology, and microscopy to name just a few.
This tutorial will introduce the physics of OCT, give an overview of its ever widening application domain, and will then concentrate on image processing algorithms used clinically in ophthalmology. The tutorial is aimed at getting researchers with a background in image processing up and running with this relatively new modality. Particular emphasis will be given to graph-based algorithms as they have emerged as the method of choice in both academia and industry. We will give an overview of their theoretical basis, their advantages and disadvantages, and how they are implemented. Image registration and machine learning will comprise the remainder of our tutorial. To close, we will give some perspective on where the technology is heading in terms of hardware development and newer applications, and how these will affect algorithm developers.
About the Authors
Jonathan Oakley has a B.Sc. in Computer Science from the University of York, England, a Masters from the department of Medical Physics at University College, London, and a Ph.D. in medical image processing from the Swiss Federal Institute of Technology. Since then, he has spent over ten years working on image processing algorithm development for KLA-Tencor, Fujifilm and, most recently, Carl Zeiss Meditec Inc., where he worked exclusively in OCT. While at Zeiss, he was responsible for motion correction, image registration and various anatomical segmentation algorithms including the Optic Nerve Head as well as retinal layer segmentation algorithms, some of which have generated new scientific findings and consequently new market opportunities for ophthalmic imaging in the neurological domain. In 2010 he and Daniel Russakoff formed Voxeleron LLC.
Daniel Russakoff received an A.B. in geophysics from Harvard University and his Ph.D. in computer science from Stanford. His research interests are in computer vision and pattern recognition in general, and biomedical image analysis in particular. He has authored numerous conference and journal papers and holds several patents on topics ranging from stereo vision to medical image registration. Since he left academia, he has worked as a Computer Scientist at the National Institute of Standards and Technology and as Chief Scientist at Fujifilm's San Jose Research Lab. His more recent work has been using probabilistic shape analysis and machine learning for segmentation of deformable structures in 2D and 3D radiological images. Currently, he works with Jonathan as a computer vision consultant at Voxeleron LLC where, in particular, he is applying his extensive experience in radiology to the development of new, more advanced algorithms for OCT.