肝癌病患文字病歷報告之錯別字辨識與更正方法

Abstract

病歷資料擁有豐富的疾病、醫療程序和治療結果等資訊。在之前的研究裡，我們實做一資訊擷取系統提取肝癌病人文字報告裡肝癌相關資訊。資訊擷取系統提取的結果將用於建立預測肝癌復發的模型。然而，由於這些文字的醫療報告為醫護人員手動輸入，其中難免會有錯別字，這些因素造成醫療擷取系統抽取資訊上的困難，而因此遺漏掉無法被抽取出的珍貴醫療資訊，所以如何減少報告中錯別字對於未來的研究是非常重要的。 本研究的目的在於提供一個有效率的方式去辨識醫療報告中之錯別字並加以更正，以幫助醫療擷取系統能抽取出更多醫療資訊。我們設計一套錯別字辨正與標準化系統，用於校正報告中之錯別字。透過這套方法，能在系統抽取資訊前，將文章中內之錯別字更正，以幫助醫療擷取系統能從中擷取到先前因錯別字與不同表示法的原因而無法被擷取出來之資訊，以提高系統之資訊抽取率。由於醫療報告過於龐大，使用人為方式尋找錯誤是非常耗費人力與時間的，透過這個方法，可有效減少醫療報告之錯別字，並改善病歷之品質。

Textual medical records contain valuable information about diseases, medical procedures and treatment results. In our previous work, we implemented the information extraction system for extracting the desired information from liver cancer patients’ textual reports. These extracted results produced by information extraction system are used for supporting the development of recurrence predictive model. However, these narrative reports are made by human manually. Therefore, improving the correctness of medical reports is very important for further research. In the study, we already implemented the information extraction system which can extract medical information for liver cancer recurrence predicting model. But, detecting and correcting the misspelling words of all medical reports manually would be a time-consuming and labor-intensive task. Therefore, the aim of this study is to provide an efficient way for facilitating the process of checking and correcting the misspelling words. We implemented the error handling system for correcting the misspelled words of each medical report. After the preprocessing procedure executed by the error handling system, the information extraction system can extract out those information which cannot be found due to the misspelling words. In this way, it can highly promote the accuracy of the medical extracted results.