PITA Fiscal Year 2008 Projects - Biomedical and Health Engineering

Signal Processing For Improved Artery Identification Using Infrared Imaging Systems

InfraRed Imaging Systems is using Infrared imaging technology to visualize blood vessels in the human body. The company has immediate needs to improve the imaging performance. The company feels that much improvement can be achieved though signal processing. Signal processing can be employed to improve the resolution of the images by employing superesolution approaches. The various sequences of images can be employed to improve resolution and to perform image fusion. Edge detection, region growing and segmentation can be employed to extract accurate position estimates of various important objects in the images. These positions can be used for more accurate positioning of instruments. When the objects and their edges are identified, they can be appropriately colored to make their visualization easier. It seems very feasible to be able to differentiate or colorize differently veins (e.g. purple color) and arteries (bright red) which would take advantage of the fact that oxygenated and deoxygenated blood absorb Infrared light differently. Current thinking indicates that such colorizations would make the current product easier and safer to use and thus provides significant gains. It also seems possible to use sets of images to build a three dimensional model of the tissues being imaged. Finally, it appears quite possible that one can image with slightly different frequencies to exploit the different responses of the tissue and vessels to the different frequencies. We are hoping that data fusion can be employed to extract the different information that is brought out by the different frequencies. It would be very interesting to experiment with these and other signal processing methods that can be used to improve and augment the imaging process.