腦中風是臨床常見疾病，中醫介入治療在台灣已漸為常態。本研究藉由現代化儀器及統計學探討中風患者中醫舌象評估腦中風狀態的可行性。本研究回溯2010年8月1日至2014年12月31日間於彰化基督教醫院腦中風會診中醫患者與至本院進行健康檢查的一般民眾中，收集基本資料與舌象，利用客觀化的自動化舌診儀器輔助，與資深中醫師的舌象判讀，比較梗塞性腦中風患者與健檢平常人的舌象，並運用統計分析找出兩組間差異與腦中風患者預測病情狀態的可行性。結果總共收集梗塞性腦中風組有309例，健檢平常人組有405例。在多變項邏輯迴歸分析結果發現，有舌體瘦小(odds rations(OR):10.1, p-value=0.003)、舌苔黃(OR:6.1, p-value=0.001)、舌苔膩(OR:17.7, p-value<0.001)、舌津偏少(OR:8.6, p-value=0.012)和舌下絡脈明顯者(曲張/集結成珠) (OR: 7.4, p-value<0.001)者較平常舌象者罹患梗塞性腦中風勝算比皆大於2，並有顯著統計差異。研究顯示部分舌象異常與罹患梗塞性腦中風有顯著的正相關，或許未來可將舌象作為預防、診斷和治療腦中風的評估模式之一。

Stroke is a common clinical disease. Recently, the Traditional Chinese Medicine (TCM) inter-vention is popular worldwide. Inspired by TCM auxiliary treatment, many kinds of modern instruments and new statistical methods are applied to evaluate whether a patient has stroke or not. In particular, TCM tongue diagnosis is now found powerful in rating the possibility of stroke for a potential patient. We collected the characteristic of tongues of the cerebral infarction patients consulting TCM as well as those of some participants that took health examinations during August 1st, 2010 and December 31st, 2014. Several items of ordinary examinations including TCM tongue diagnosis are also recorded in the data set. The data concerning tongue diagnosis was collected by the objective automatic tongue diagnosis system (abbreviated as ATDS) developed by our team, and was rated by senior TCM physicians. Statistics testing is conducted to judge if there are significant differences between the two groups, the experimental group and the control group. We also discuss the possibility of predicting the severity of the cerebral infarction of a patient by the characteristic of tongue. We recruited 714 participants, including 405 individuals without cerebral infarction and 309 patients with cerebral infarction. According to the logistic regression analysis, the odds of having thin tongue is 10.1 times compared to normal size tongue (p-value=0.003), yellow fur is 6 times compared to normal fur (p-value=0.001), slimy fur increases to 17.7 times in odds ratio (p-value<0.001), tongue saliva depletion increases 8.6 times in odds ratio(p-value=0.012), and twisted sublingual collateral vessels is 7.4 times of odds ratio (p-value<0.001). The odds ratio of suffering cerebral infarction is twice for the tongues of the abnormal group compared with the normal group, which shows a significant difference. In this study, we find that the abnormality of the characteristics of the tongues of the patients is highly correlated with their cerebral infarction. Tongue diagnosis might emerge as a tool on rating whether a patient has stroke or not.