Bio-Inspired Analog Visual Systems: The main goal of my research is to develop novel methods for improving visual processing for autonomous systems (robots, planetary explorers, etc). To achieve this, we feel that the best approach is to use nature for our model; animal visual systems are able to perform extremely well under the constraints of limited power (food for energy) and space (eye and brain size). Thus, my group is creating biologically-inspired silicon vision chips. The research combines analog chip design, optoelectronics, and neural networks. The modular aspect is the key to developing different types of processing components with interchangeable parts that can simplify reconfiguration.
Bioinstrumentation: Bioinstrumentation is the application of novel devices to detect biological signals from the human body. This can be done remotely, using various electromagnetic energy sources (such as X-ray or optical wavelength signals), or it can be done directly on the body, such as using electrodes to measure EKG. We are working on new sensors for measuring signals from the body for diagnostic and early detection of diseases, and for disease management. We have also worked on developing new imaging systems - specifically high resolution x-ray imaging using solid-state devices
Integrated Sensors: My original work on visual processing uses on-chip optical detectors as the means for converting light to electrical signals. One of the most exciting aspects of research is that our activity often leads us into new areas, and this is the case with my sensors-related work. We have grown a new research area that is an off-shoot of the vision work, in which our detector chips are used as part of novel sensor systems, including biochemical sensing. The process of detecting chemical agents requires many parts: a sensor to produce a signal indicating something of interest is present, a stimulator to activate the sensor, a transducer to convert the signal into another, more easily manipulated signal, and a processor to make that signal understandable and usable.