MISP

The 6th edition of Medical Imaging Symposium for PhD Students was held on 6th May, 2011 at the Erasmus Medical Center (EMC), Rotterdam. This Medical Imaging Symposium is organized for PhD students or young researchers in the field of Medical Image Computing, Processing and Analysis and other closely related fields. The goal of this symposium is to bring PhD students together to exchange experiences in medical imaging research in an informal setting with lots of opportunities for discussion and feedback.

For this year, celebrating, we had prepared a special program including:

• a free workshop on “Pattern recognition in medical diagnostics” by Pavel Paclik (PhD) & Carmen Lai (PhD)

• a key-note presentation given by Dr. Oleh Dzyubachyk on “Model-Based Cell Tracking and Analysis in Fluorescence Microscopy”

• 5 short presentations from PhD students from all over the Netherlands on various topics

VENUE: Colloquium room A (Ec-412) [Above the main restaurant],  Erasmus Medical Center, Rotterdam

TIME: 9: 30 AM

Dr. Carmen Lai

Dr. Carmen Lai received her MSc in Electronic Engineering from Cagliari university, Italy, in 2002. In 2003 she joined the Bioinformatic group at TU Delft. She worked in collaboration with the Dutch Cancer Institute on human breast cancer in high throughput data, and defended her PhD in 2008. She joined PR Sys Design and is now a full-time entrepreneur, continuing to cooperate with TU Delft on pattern recognition research, and teaching in industrial pattern recognition courses.

Dr. Pavel Paclik

Dr. Pavel Paclik received both his MSc (cum laude) and PhD from Czech Technical University in Prague in transportation automation (pattern recognition for traffic sign recognition) in 1998 and 2004, respectively. In 2000 Pavel joined the Pattern Recognition Group at TU Delft, working on multi-band and hyperspectral imaging, texture classification, dissimilarity-based recognition and classifier fusion. In 2006 Pavel started PR Sys Design company helping industrial and academic researchers to design and implement pattern recognition systems. Since 2007 he is a full-time enterpreneur, continuing to do research on advanced ROC analysis and system design. In the medical field, Pavel has helped to design algorithms for tissue characterization, medical segmentation and neurological brain-computer interfacing for both industrial and academic partners.

Dr. Oleh Dzyubachyk

Dr. Oleh Dzyubachyk was born in Lviv, Ukraine, on August 20, 1976. He received a M.Sc. degree (cum laude) in Mathematics from Ivan Franko National University of Lviv, Ukraine, in 1998. From 1998 to 2003, he carried out research on modeling coupled fields in ferromagnetic materials at Pidstryhach Institute for Applied Problems of Mechanics and Mathematics (Lviv, Ukraine). From 2003 to 2005, he was a Research Assistant (postmaster program “Mathematics for Industry”) at the department of Mathematics and Computer Science of Technical University of Eindhoven, the Netherlands. After graduating in 2005 he obtained a Professional Doctorate in Engineering (PDEng) degree. From 2006 to 2010, Oleh was a Ph.D. student at the Biomedical Imaging Group Rotterdam, part of Departments of Medical Informatics and Radiology of the Erasmus University Rotterdam, the Netherlands. In October 2010 he joined the Laboratorium voor Klinische en Experimentele Beeldverwerking (LKEB) at the Leiden University Medical Center.

Vincent van Ravesteijn

Vincent van Ravesteijn studied Applied Physics at the Delft University of Technology. Continuing his work for the master thesis he has been pursuing his PhD for the last four years on “Automated Polyp Detection and Electronic Cleansing” for CT Colonography and expected to finish by the end of this year. At the moment he will be continuing his academic career as a post-doc on image processing and signal analysis for the improvement of electron microscopy.  His interests are in the field of quantitative image processing and pattern recognition.

Roy van Pelt

Roy van Pelt is a Ph.D. candidate at the Biomedical Image Analysis group of the department of Biomedical Engineering of Eindhoven University of Technology. In particular, he is affiliated with the Multi-valued Image Analysis and Visualization group, headed by Anna Vilanova. In addition, he is part-time involved with Philips Healthcare in Best. His research focuses on illustrative medical visualization, with a specific interest for multi-valued image data. His recent work deals with visualization of cardiovascular blood-flow velocity data, measured through magnetic resonance imaging. More specifically, he actively investigates the merit of illustrative rendering and real-time interaction techniques for comprehensive exploratory analysis of unsteady blood-flow fields. He received his master’s degree in Computer Science and Engineering from Eindhoven University of Technology in 2007. He was granted the ‘best-paper award’ at the Volume- and Point based Graphics conference in 2008.

Renske de Boer

Renske de Boer obtained her MSc degree in Biomedical Engineering in 2006 at the Eindhoven University of Technology. That same year she started as a PhD student at Biomedical Imaging Group Rotterdam, Erasmus MC. The first part of her PhD project consisted of the automatic segmentation of white matter lesions and brain tissues in MR images. Currently she is working on brain connectivity using diffusion MRI. She will defend her thesis on June 17, 2011.

Oscar Debats

Oscar Debats is a PhD student at the Department of Radiology of Radboud University Nijmegen Medical Centre (RUNMC). He studied Medical Biology and Medicine, both at the Vrije University (VU) in Amsterdam. In June 2008 he started his PhD project at the Diagnostic Image Analysis Group in Nijmegen. He works with Henkjan Huisman on Computer-aided detection of metastatic pelvic lymph nodes.

Hugo Kuijf

Hugo Kuijf (1986) completed his Computer Science bachelor at the Utrecht University in 2007. During his master Game and Media Technology, he developed a special interest in medical imaging. He added some courses from the Biomedical Image Sciences master to his curriculum, focussing on medical imaging, human anatomy, and image processing. After a two-month research project at the Image Sciences Institute (ISI), he started his master thesis at the Netherlands Cancer Institute (NKI-AvL) in Amsterdam. After graduation, he started as a PhD student at the ISI in September 2009. This project focuses on automatic segmentation of the human brain using MR images.

Pattern recognition in medical diagnostics

Pavel Paclik (PhD) & Carmen Lai (PhD)

Medical imaging today is characterized by large and ever-growing number of imaging modalities. The most of the medical diagnostic problems rely on robust processing of sensory data using statistical pattern recognition approaches. This talk will outline pattern recognition techniques typically used in medical imaging and point to specific problems arising in medical applications.

The presentation will be structured in two parts. Firstly, we will discuss statistical pattern recognition methodology and survey number of basic concepts and techniques. These will include detection and classification, dimensionality reduction, performance evaluation and optimization. In the second part, we will highlight number of challenging issues that are typically appearing in medical imaging applications. We will discuss how to approach data normalization, presence of noisy labels, rich meta-data associated with medical images and specific needs of performance optimization. Practical examples will be included.