| Summary: | 碩士 === 國立中山大學 === 資訊工程學系研究所 === 101 === Breast cancer ranks second in the cancer fatality rate among females worldwide. One out of every eight women gets BC during lifetime. The reasons leading to breast cancer are multiple. Even after breast cancer is cured, there is still some metastasis probability which may damage body and threat to life. The prognosis of breast cancer is important to predict the metastasis of breast cancer and reach early treatment so that it can increase the survival rate. The tongue features for 60 breast cancer patients and 70 normal persons are extracted by the Automatic Tongue Diagnosis System (ATDS). A total of nine tongue features, namely, tongue color, tongue quality, tongue fissure, tongue fur, red dot, ecchymosis, tooth mark, saliva, and tongue shape are identified for each tongue. Features extracted are further sub-divided according to the areas located, i.e., spleen-stomach, liver-gall-left, liver-gall-right, kidney, and heart-lung area. The purpose focuses on inducing significant tongue features (p&;lt;0.05) to discriminate breast cancer patients from normal persons. The Mann-Whitney test shows that the amount of tongue fur (p=0.007), the tongue fur in the spleen-stomach area (p=0.020), maximum covering area of tongue fur (p=0.002), thin tongue fur (p=0.000), the number of tooth mark (p=0.050), the number of red dot (p=0.000), red dot in the spleen-stomach area (p=0.000), red dot in the liver-gall-left area (p=0.002), red dot in the liver-gall-right area (p=0.000), red dot in the heart-lung area (p=0.003) demonstrate significant differences. Next, the data collected are classified into two groups. The training group consists of 55 breast cancer patients and 60 normal persons, while the testing group is composed of 5 breast cancer patients and 10 normal persons. The logistic regression by utilizing these 10 tongue features with significant differences in Mann-Whitney test as factors is performed. Then we remove one of the 10 tongue features which is not the most significant differences (p&;gt;0.05) and perform logistic regression three times. Among them, the amount of tongue fur (p=0.011), the tongue fur in the spleen-stomach area (p=0.006), the maximum covering area of tongue fur (p=0.003), thin tongue fur (p=0.019), the number of red dot (p=0.017), red dot in the spleen-stomach area (p=0.002), red dot in the heart-lung area (p=0.016) reveal independently significant meaning. To prove the importance of independently significant meaning, we remove an independently significant meaning, thin tongue fur (p=0.019), and perform logistic regression. Among them, the amount of tongue fur (p=0.037), the tongue fur in the spleen-stomach area (p=0.005), the maximum covering area of tongue fur (p=0.001), the number of red dot (p=0.008), red dot in the spleen-stomach area (p=0.002), red dot in the heart-lung area (p=0.005) reveal independently significant meaning. The tongue features of the testing group are employed in the aforementioned three models to test the power of significant tongue features identified in predicting breast cancer. An accuracy of 80% is reached through Model I by applying the 10 significant tongue features obtained through Mann-Whitney test. For the second model employing 7 tongue features induced by logistic regression with independently significant meaning, 80% accuracy is achieved. The third model employing 6 tongue features induced by logistic regression with independently significant meaning, 67% accuracy is achieved and proves the important prediction of independently significant meaning.
|
|---|
