隨著全球人口快速老齡化和醫療人力資源日益緊缺，提高醫療作業效率已成為當務之急。在醫療資訊管理領域中，護理紀錄扮演著舉足輕重的角色。
目前的傳統護理紀錄在提升現代護理人員效率方面存在一定局限性。傳統方法雖強調紀錄的完整性，但缺乏有效的紀錄模式和生成式人工智慧的支持。例如，現有的傳統護理紀錄多採用手寫方式降低了時間效率，加重了護理人員的工作負擔。此外，不同護理人員的專業程度差異也導致紀錄品質參差不齊，無法有效促進醫療資訊的互通與整合。
本研究旨在解決現有護理紀錄系統的不足，特別聚焦於護理紀錄和生成式人工智慧應用方面的挑戰。研究目標是提出一套整合護理紀錄和生成式AI的解決方案，以提升效率。具體而言，本研究希望增加護理人員在記錄過程中使用生成式AI，所以引入生成式AI以強化紀錄的效率；同時利用生成式AI將護理紀錄標準化，加速護理紀錄的建立。
本研究提出了一套利用生成式人工智慧轉換護理紀錄的方法論，主要分為三個階段：首先，對傳統的護理紀錄進行簡化處理；其次，利用生成式人工智慧整合病患生理數值和主訴；最後，生成一套完整的護理紀錄。研究結果顯示，此方法論在時間效率、一致性、完整性方面均具有顯著優勢。

With the rapid aging of the global population and the increasing shortage of medical human resources, improving the efficiency of medical operations has become a top priority. In the field of medical information management, nursing records play a pivotal role.
The current record of traditional nursing has certain limitations in improving the efficiency of modern nursing staff. Although traditional methods emphasize the integrity of records, they lack the support of effective recording models and generative artificial intelligence. For example, existing traditional nursing records are mostly written by hand, which reduces time efficiency and increases the workload of nursing staff. In addition, differences in the professionalism of different nursing staff also lead to uneven record quality, which cannot effectively promote the exchange and integration of medical information.
This study aims to address the shortcomings of existing nursing record systems, specifically focusing on challenges in nursing records and generative artificial intelligence applications. The research goal is to propose a solution that integrates nursing records and generative AI to improve efficiency. Specifically, this study hopes to increase the use of generative AI by nursing staff in the recording process, so generative AI is introduced to enhance the efficiency of recording; at the same time, generative AI is used to standardize nursing records and accelerate the creation of nursing records.
This study proposes a methodology for using generative artificial intelligence to transform nursing records, which is mainly divided into three stages: first, simplifying the traditional nursing records; second, using generative artificial intelligence to integrate patients' physiological values and chief complaints; Finally, a complete set of nursing records is generated. The research results show that this methodology has significant advantages in terms of time efficiency, consistency, and level of detail.

論文審定書i
致謝ii
摘要iii
Abstractiv
目 錄v
圖目錄vii
表目錄viii
第一章緒論1
第一節研究背景與動機1
第二節研究目的與範圍3
第二章文獻探討4
第一節護理紀錄4
第二節AI與生成式AI簡史9
第三章研究方法17
第一節設計科學研究法17
第二節研究流程與步驟20
第四章設計與發展解決方案23
第一節利用ChatGPT4 轉換護理紀錄之方法23
第二節用ChatGPT案例評估34
第五章展示與評估49
第一節案例展示49
第二節方法評估68
第六章討論與結論70
第一節研究成果70
第二節研究貢獻73
參考文獻76
英文部份76
中文部份78

Abdelhady, A. M., & Davis, C. R. Plastic Surgery and Artificial Intelligence: How ChatGPT Improved Operation Note Accuracy, Time, and Education. Mayo Clinic Proceedings: Digital Health, 1(3), September 2023, pp. 299-308.
Ayers, Comparing Physician and Artificial Intelligence Chatbot Responses to Patient Questions Posted to a Public Social Media Forum, 2023.
Buchanan, B., The AI Triad and What It Means for National Security Strategy :Center for Security and Emerging Technology, August 2020.
Cheevakasemsook, A., Chapman, Y., Francis, K., & Davis, C. The study of nursing docu- mentation complexities. International Journal of Nursing Practice, 12(6), 2006, pp. 366-374.
Hansten, R. Streamline change-of-shift report. Nursing Management, 34(8), 2003, pp. 58-59.
INACSL Standards Committee, Watts, P. I., Mc Dermott, D. S., Alinier, G., Charnetski, M., & Nawathe, P. A. Healthcare Simulation Standards of Best Practice TM Simulation Design. Clinical Simulationin Nursing, 58, 2021, pp. 14-21.
Griffiths, P., Debbage, S., & Smith, A. A com- prehensive audit of nursing record keeping practice. British Journal of Nursing, 16(21), 2007, pp. 1324-1327.
Koska, M. T. Quality-the name is nursing care, CEO say. Hospitals, 63(2), 1989, 32.

Kung TH, Cheatham M, Medenilla A, Sillos C, De Leon L, Elepaño C, et al. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models. PLOS Digit. Health 2023, 2, e0000198.
Kosinski, M. Theory of mind might have spontaneously emerged in large language models, 2023.
Rahman, M.M., Watanobe, Y. Chatgpt for education and research: Opportunities, threats, and strategies. Appl. Sci. 2023, 13, 5783.
OpenAI, ChatGPT, Feb 2024, https://chatgpt.com/, [Retrieve 2024/06].
 
吳仁和，物件導向系統分析與設計-結合MDA與UML，8th edition，台北：元照，June 2024.
童恒新、吳昆家。ChatGPT於醫護教育之應用。源遠護理，17(2)，2023, pp. 17-21,。
邱台生、徐南麗。台北榮總護理部推行工作簡化之實施評值.榮總護理， 9 (3)，1992, pp. 309-312。
常玉琴、黃惠姿、王襄華、張雅惠、樊君儀、高麗娟、戴千淑。提昇護理紀錄書寫完整性之改善方案。榮總護理，26(3)，2009, pp. 304-310,。
蔣立琦、葉美玲、蘇遂龍．精準護理─以人為本 的 知 識 轉 譯．護理雜誌，63（6），2016, pp. 23–29,。
凌荃，第五章 護理記錄 重點整理－護理師國考快捷｜痞客邦，2019。
https://n220577775.pixnet.net/blog/post/351458923-%E7%AC%AC%E4%BA%94%E7%AB%A0--%E8%AD%B7%E7%90%86%E8%A8%98%E9%8C%84--%E9%87%8D%E9%BB%9E%E6%95%B4%E7%90%86, [Retrieved 2024/04].
工研院產業學習網，科技與醫療：帶你了解智慧醫療趨勢、應用及願景，2022。
https://college.itri.org.tw/Home/InfoData/f6e19f2d-f81c-421c-bc36-ea6409ba0a5d/644a31e4-9184-4ad9-a2f2-0256b0b908da, [Retrieved 2024/04].
Invest Taiwan 全球智慧醫療市場發展概況與趨勢，2024。
https://investtaiwan.nat.gov.tw/intelInfoPageCht202406060001cht?lang=cht&search=202406060001, [Retrieved 2024/03]
林佳慧、丘周萍。腎組織灌流不全之護理，2005，
http://www.tnna.org.tw/upfile/file/File_20101022104847.pdf,[Retrieved 2024/03]
林瓊華。護理記錄 - 衛生福利部胸腔病院，2000，
https://www.ccd.mohw.gov.tw/public/news/handouts/b38c06ff2d89945df203da1c31cf8b03.pdf, [Retrieved 2024/5]
何泓澐、高婉甄。敏惠醫護管理專科學校，
https://c017.mhchcm.edu.tw/var/file/17/1017/img/197/132500127.pdf, [Retrieved 2024/5]
q19821209，DART護理紀錄，2008。
https://q19821209.pixnet.net/blog/post/204127475, [Retrieved 2024/5]
許育菁。DART 護理紀錄介紹-- 護理之家
https://www.puzih.mohw.gov.tw/public/ufile/887fed1c8f3bc1043446fbd50f8f3fd3.pdf, [Retrieved 2024/5]