請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/61743
標題: | iPad電子病歷對醫師工作的效益 Mobility Patterns of Doctors Using Mobile Electronic Health Records on iPads |
作者: | Allan Chang Lin 林亞倫 |
指導教授: | 陳彥仰(Mike Yen-Yang Chen) |
關鍵字: | 電子病歷,醫學資訊,iPad,行動裝置互動, Electronic Health Records,Medical Informatics,iPad,Mobile Device Interaction, |
出版年 : | 2013 |
學位: | 碩士 |
摘要: | 現今的電子病歷都局限於桌上型電腦。而且當醫生們需要看病患的電子病歷時他們都必須在醫院的辦公室或護理站才接觸得到這些電腦。但是醫院的電腦的數量是有限的,所以導致醫生們會需要多花而外的時間尋找沒正在被使用的電腦並影響工作效率。為了解決這個問題有一些醫院開始使用「行動電子病歷」,也就是結合了電子病歷與現今市場上的觸碰式行動裝置。我們希望了解醫生們使用這種行動電子病歷時的行動狀態。
於是我們追蹤了台灣某大醫院的住院醫生使用Dr. Pad 的使用和行動模式。Dr. Pad 是一款為Apple iPad所設計的行動電子病歷應用程式。我們在Dr. Pad編碼內寫入了記錄器記錄著住院醫生們使用Dr. Pad時的時間、次數、時間長度和行動數據。在這次的研究中我們收集了179位住院醫生使用Dr. Pad四個星期(28天)的使用數據。然後我們運用機器學習演算法建了一個J48決策樹(decision tree)來猜測醫生們的行動類別。這讓我們可以知道醫生在使用Dr. Pad時是否是靜止或行走中。 在這28天我們收集了16,157筆記錄。這些記錄所顯示的趨勢支持了醫生們所告訴我們的使用以及我們在現場所觀察到的使用趨勢。根據Dr. Pad所收集到的數據,醫生們在使用Dr. Pad時大部份是在行動中。使用時間也大多是聚集在早上,而且行動中的比例也突顯地比較高。另外,醫生們使用Dr. Pad的時間長度也比我們預期的還要長,甚至是在我們預期是巡房和病患互動的時段也是如此。 Before Electronic Health Records (EHRs) were available on touch-panel tablets, doctors were confined to accessing patient records on the hospital’s computer stations, in their offices or at nurse stations. This situation is further aggravated by a limited number of workstations, which leads additional workload on doctors looking for an available computer. Mobile EHRs were introduced in order to curb this problem, allowing doctors to access records almost anywhere. This means the doctors can access records while on the go. The most recent incarnation of the mobile EHR comes in the form of touch surface tablets such as the Apple iPad. We want to know what the doctors’ mobile behaviors are like while using such a portable EHR. Thus we studied the use of Dr. Pad, a mobile EHR application on the iPad used by resident doctors of one of Taiwan’s largest hospitals. We inserted a logger that allowed us to extract direct usage and motion data from a large-scale in-the-wild use of a mobile EHR by 179 resident doctors over 4 weeks. Using machine-learning techniques, we built an unpruned J48 decision tree classifier with which we can predict the doctors’ mobile behaviors while using Dr. Pad. We could tell whether the user was moving or stationary during this time, which was previously unobserved and self-reported to an extent. We captured 16,157 sessions worth of data across 179 doctors who used Dr. Pad over 28 days. Our data revealed trends in the doctors’ use of the mobile EHR, which supported reports from doctors and our observations of their work routines. The doctors were in fact mobile for the majority of the time they used Dr. Pad. Their use was also concentrated in the morning, as we had expected, and were significantly more mobile as well. Furthermore, we also found that the doctors used Dr. Pad more frequently than we had expected, even during times when we expected them to be engaged with patients. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/61743 |
全文授權: | 有償授權 |
顯示於系所單位: | 資訊網路與多媒體研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-102-1.pdf 目前未授權公開取用 | 7.96 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。