Preoperative MR Imaging Analysis of Osteosarcoma Using Patlak Plot Technique

• Main Authors: Pei-hua Wu, 吳佩樺
• Other Authors: Ruey-Feng Chang
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/07035641648024135478

碩士 === 國立中正大學 === 資訊工程所 === 96 === Osteosarcoma is the most common bone tumor, which often occurs in children and adolescents. Osteosarcoma has not obvious symptoms so it is easily misdiagnosed. In recent years, the dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) is often used for diagnosis of osteosarcoma and monitoring the response preoperative chemotherapy of osteosarcoma. In DCE-MRI, the images are obtained every two seconds to observe the signal intensity change during the injection of contrast agent. For the viable-appearing tumor cells, the signal intensity change will be larger but the signal intensity change will be smaller for the necrotic tumor cells. The proposed computer-aided analysis can help doctors determine the results of chemotherapy treatment, including the tumor necrosis rate and the extent of the tumor necrosis, before the operation. The Patlak plot technique is adopted in the proposed analysis system. It is assumed that the contrast medium will keep in the tumor tissue. The Patlak analysis uses a graphical method to evaluate the permeability values and the related blood volume by the slope and intercept of a derived straight line. Using the pixel-by-pixel Patlak analysis in the tumor, the permeability values of all the pixels in the tumor could be found. A permeability threshold will be automatically selected to decide whether a pixel is necrotic or not. If the permeability value of a pixel is larger than the threshold, then this pixel is classified as a living cell. Otherwise, it is classified as a necrotic cell. Based on the pathological results, the Patlak plot analysis is compared with the tumor analysis without vessel information. In the experiment, there are 15 cases to test the accuracy of the proposed computer aided analysis system. The average accuracy rate of the Patlak plot analysis is 91.36% and the average accuracy rate of the tumor analysis without vessel information is 87.74 %.