脂肪肝已成為最常見之肝臟慢性疾病。超音波是臨床上最常用來檢查暨追蹤肝臟疾病之造影工具。然而診斷結果卻受限於操作者之主觀意識判斷。本研究之目的是建立穩健脂肪肝超音波影像分類模型。本研究採用回朔性分組實驗設計；收集299 受試者包括正常、輕度、中度及重度脂肪肝之超音波影像，選取感興趣區域，萃取去斑塊雜訊及修剪像素影像紋理特徵，利用數據挖掘及人工智能方法，使用肝臟影像及多樣影像特徵，建立肝臟脂肪變性影像分類模型。簡單貝氏分類器及支持向量機模型準確度分別為0.67 及0.70，巢狀人工智能網路及卷積神經網路準確度分別為0.79 及0.81。卷積神經網路是可行且穩健的脂肪肝影像分類方法，未來，制訂有效的模型訓練策略及設定適當的模型參數，決定卷積神經網路分類之效率及準確度。

Steatosis is the leading chronic hepatic disorder worldwide. Ultrasound (US) is the most utilized modality and the follow-up imaging tool for visualizing liver clinically. However, the diagnostic result is often limited to operator dependenc,y and subjective evaluation. The main aim of this investigation is to build a robust classification model for hepatic steatosis US images. In this retrospective study, liver US images of 299 subjects, consisting of cases of normal liver and mild, moderate, and severe steatosis, were collected. The speckle noise suppression and trimming intensity algorithm were applied to the extracted regions (ROIs) in images. Numerous features in hepatic US images were utilized to build classification models using data mining and artificial intelligent approaches. The accuracy of classification models by Naïve Bayes and Support Vector Machine (SVM) classifier were 0.67 and 0.70, respectively. The accuracy of Nested Artificial Neural Network (Nested ANN) and Convolutional Neural Network (CNN) were 0.79 and 0.81, respectively. The results showed that CNN was the most feasible approach to classification of hepatic steatosis US images. In the future, we expect to draw up an effective training strategy and set up proper parameters which determines the efficiency and accuracy of CNN.