This award is to support a cooperative research by Dr. Aly Farag, Department of Electrical Engineering, Computer Vision & Image Processing Laboratory (CVIP Lab), University of Louisville, Louisville, Kentucky, and Dr. Tarek El-Diasty, Urology and Nephrology Center, Mansura University, Mansura, Egypt. They plan to investigate image registration, segmentation and analysis of the contrast enhanced MR images with specific application to kidney images. Renal imaging using Gd-DTPA Enhanced Dynamic MRI and Color Doppler Ultrasonography can be used for the early detection of rejection of kidney transplants. As rejection develops, transplanted kidneys show a noticeable increase in size and develop abnormal flow patterns that are not uniformly distributed throughout the whole kidney. Quantification of these changes, using image analysis can be used to detect rejections, thus replacing risky biopsy procedures. The PIs plan to combine the two powerful imaging techniques - Gd-DTPA Enhanced Dynamic MRI and Color Doppler Ultrasonography - in order to automatically detect the renal dysfunction of kidney transplants. The image analysis approach will be validated using manual approaches by expert physicians.
Intellectual Merit: Dynamic MRI is a fast acquisition protocol that produces quite noisy and low resolution images compared to traditional MRI. Therefore, the proposed image analysis approach must deal with these restrictions and be able to distinguish the parts of the kidney (such as the cortex and medulla) that portrays its function, mainly blood purification. The focus in this project is on image registration, segmentation and analysis of the contrast enhanced MR images. While the research is of substantial importance in clinical diagnostic science it will stimulate new approaches in automated computer analysis and detection. The use of these innovative methods creates a robust and accurate registration method for various imaging conditions with varying geometrical and intensity status. The methodologies used are unique, and are highly developed at the PI's laboratory.
Broader Impact: Image segmentation and registration cuts across nearly all applications of medical imaging analysis; hence, these methodologies will find greater usage in this dynamic field of research. The project will require transfer of many images over the internet; thus will benefit from information technology initiatives like the vBNS and Internet2 which have been supported by the NSF. The project will enhance collaboration between the US and developing counties, like Egypt, and promote positive exchanges of ideas and appreciation of cultures. The expected outcome is the creation of an image analysis approach for accurately tracking the kidney function after an implant. The findings will improve healthcare delivery of the increasing number of kidney transplant patients in Egypt, the US and worldwide.
This project is being supported under the US-Egypt Joint Fund Program, which provides grants to scientists and engineers in both countries to carry out these cooperative activities.
