Graft failure of kidneys after transplantation is most often the consequence of the acute rejection. Hence, early detection of the kidney rejection is important for the treatment of renal diseases. In this chapter, authors introduce a new automatic approach to classify normal kidney function from kidney rejection using dynamic contrast enhanced magnetic resonance imaging (DCE-MRI). The kidney has three regions named the cortex, medulla, and pelvis. In their experiment, they use the medulla region because it has specific responses to DCE-MRI that are helpful to identify kidney rejection. In the authors’ process they segment the kidney using the level sets method. They then employ several classification methods such as the Euclidean distance, Mahalanobis distance, and least square support vector machines (LS-SVM). The authors’preliminary results are very encouraging and reproducibility of the results was achieved for 55 clinical data sets. The classification accuracy, diagnostic sensitivity, and diagnostic specificity are 84%, 75%, and 96%, respectively.
TopIntroduction
According to National Kidney Foundation, patients with renal diseases have increased from 100,000 to 400,000 over the last two decades (Tu, 2004). The patients with renal failure have three choices: 1) dialysis treatment, 2) medication, and 3) renal transplantation (Hamilton, 1999). Dialysis therapy is used for removing fluid, potassium, and uremic toxins to relieve symptoms of diseases. However, dialysis therapy may not prevent kidney failure (Tu, 2004). Patients who have acute renal disease can be treated with drugs. However, treatment with drugs has some side effects such as infections, and increasing incidence of cancer. Since the first successful organ transplantation was made in 1954 (Tu, 2004), renal transplantation has been applied for treatment of renal failure. However, patient might have some problems such as acute rejection, delayed-appearing antibody-mediated rejection, and acute tubular necrosis after kidney transplantation (Solez, 1999). In order to detect acute rejections, molecular diagnostic strategies have been investigated by the National Institutes of Health and the Immune Tolerance Network (Tu, 2004). However, this field still needs more experimentation and improvement.
Long-term renal allograft survival is dependent on the early detection of acute rejection. Currently, the diagnosis of rejection is done via biopsy, but biopsy has the negative effect of subjecting the patients to risks like bleeding and infections. Moreover, the relatively small needle biopsies may lead to over or underestimation of the extent of inflammation in the entire graft (Farag, 2006). Also, transplanted kidneys face a number of surgical and medical complications. These cause a decrease in kidneys functionality (Yuksel, 2005). Therefore, a noninvasive and repeatable technique is not only helpful but also needed in the diagnosis of acute renal rejection.
In DCE-MRI, a contrast agent called Gd-DTPA is injected into the bloodstream, and as it perfuses into the organ, the kidneys are imaged rapidly and repeatedly. During the perfusion, Gd-DTPA causes a change in the relaxation times of the tissue and creates a contrast change in the images (Yuksel, 2005). As a result, the patterns of the contrast change give functional information, while MRI provides good anatomical information which helps in distinguishing the diseases that affect different regions of the kidneys (Yuksel, 2007). The kidney has mainly three regions: 1) medulla, 2) cortex, and 3) pelvis. In our experiment, they use the medulla and cortex regions. A typical protocol using dynamic MRI involves the following steps: 1) collect a sequence of images from the kidney region as the contrast agent perfuse through the kidney; 2) follow the contrast difference in the kidney with time (image intensity information from the cortex and the medulla regions vary with time as the contrast agent perfuse through the kidney); and 3) establish a correspondence between change of contrast in the image sequence and the status of the kidney.