Correlation Study of Image Features between Ultrasound and PET on Breast Cancer Scan

• Main Authors: Yen, YU-JEN, 顏郁人
• Other Authors: Huang, Yung-Hui
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/12084505898935265017

碩士 === 義守大學 === 資訊工程學系碩士在職專班 === 101 === In the past decade, the incidence and death rate of female with breast cancer is increasing year after year, and the mean age of patience is becoming younger. According to the study of American Joint Committee on Cancer (AJCC), early detection and treatment of breast cancer can greatly improve the survival rate of patients. The aim of this study was to retrieve the image features and to do the correlation analysis between Positron Emission Tomography (PET) and Breast Ultrasound ianges of breast cancer patients. This study collected 14 cases (mean age: 55) with breast cancer by biopsy and have received both Positron Emission Tomography (PET) and Breast Ultrasound scanning in three months. In the Ultrasound image, both of B-Mode and Elastography would be undergoing for all patients. Mean, SD and Entropy were retrieved in B-Mode Ultrasound, and the elastography factors were be assess in Elastography. Meanwhile, the maximum standardized uptake value (SUVmax) of breast lesion can be retrieved from PET images. According to the biopsy report, the patients based on their breast cancer stage can be divided into two groups. The collected data will be analyzed by the statistics analysis with Mann-Whitney U test, ROC-AUC Analysis and Kappa agreement analysis. The p-value of SUVmax and SD collected by Mann-Whitney U test are 0.0.6 and 0.036 respectively (less thanα=0.05). The result indicates that SUVmax and SD can effectively distinguish whether the lesion belongs to the first group or the second group. The cut-off point of SUVmax and SD collected by ROC-AUC Analysis are 3.050 and 35.182 respectively. The result indicates that SUVmax and SD can effectively distinguish whether the lesion belongs to the first group or the second group. The results from Kappa agreement analyses are SUVmax(Cut-off point: 3.05), Kappa=0.696 of biopsy and SD(Cut-off point: 35.182), Kappa=0.650, and Kappa=0.696 of SUVmax(Cut-off point: 3.05) and SD(Cut-off point: 35.182). The cut-off point from SUVmax and SD were highly consistency to distinguish whether the lesion belongs to the first group or the second group. Ultrasound is one of an economical, convenient and non-ionizing radiation inspection instrument. It not only can be used to identify the stage of breast cancer but also can be used to track whether there is recurrence. In the future, more data would be collected and more information can be gained for diagnosing the lesion in early stage. That will be achieving early detection and treatment of breast cancer and improving the survival rate of patients.