Navigation of Microparticles in CAM vasculature (2011)
Navigation microparticles against the flow in the human vasculature is a challenge that many researchers try to address mainly through modeling. We are using the vasculature of a chicken embryo and microcannulas to insert feromagnetic and permanent magnetic microparticles and guide them at branching points etc.
Visually Servoing Wireless Magnetic Intraocular Microrobots (2006 - 2011)
During my Ph.D. studies, I work on project for performing drug delivery and sensing in the posterior eye segment using microrobots. My work focuses on using new computer vision algorithms to extract the position of intraocular microdevices from ophthalmoscopic images, and use this information to control the electromagnetic field, and servo the microrobots. This work was conducted at the Multiscale Robotics Lab (MSRL) of the Institute of Robotics and Intelligent Systems (IRIS) at ETH Zurich.
Steerable Intravitreal Implants: Mobility Evaluation (2010 - 2011)
Drug delivery with intraocular implants is a promising strategy for the length treatment of diseases in the posterior eye segment. We propose magnetic drug-carrying reservoirs as a method to perform localized drug delivery. Mobility of magnetic reservoirs can allow an unprecedented level of control, and minimally invasive treatment. With our work, we evaluate the possibility of using magnetic intraocular implants in the vitreous humor of the human eye.
Visual Tracking for Microrobots and Microrobotic Soccer (2009)
MagMites are resonant magnetic actuators that fit in a square of 300 um dimensions. During spring and summer 2009, I assisted the course of “Advanced Robotics and Mechatronics”, at ETH Zurich. The students were tasked to redesign MagMites and create algorithms for their path planning. As the ETH Zurich team, we participated at the RoboCup competition in Graz, Austria; we defended and won the title of World Champion.
E. coli Bacteria with Reprogrammable Memory (2007)
Our combined team of biologists and engineers coped with the problem of implementing memory capabilities in bacterial colonies. Firstly, E. coli were made able to respond differently (with distinct fluorescent proteins) to two different chemical inputs. Secondly, they remembered which input was presented to them. Finally, when confronted with a new input, they were able to recognize whether it is the one that they were trained with or not. We successfully presented this projected at the International Competition on Genetically Engineered Machines (iGEM) in 2007; we won a gold medal, and the best presentation award. Our old webpage is still available here.
Tracking of Moving Objects with Emphasis on Human Gestures (2006)
During my Master’s Thesis, I worked in the area of tracking. I developed code for tracking using geodesic active contours using the narrow-band propagation algorithm, and I incorporated an optical-flow-based force to segment only the moving objects in images. Additionally, I worked with shape priors of human hands, in order to perform the segmentation and subsequent tracking of human gestures. This work was conducted at the Computer Vision and Signal Processing Lab (CVSP) of the Electrical and Computer Engineering School (ECE) at the National Technical University of Athens (NTUA).