醫用語音電腦輔助系統之設計研究——以輸尿管鏡手術為例

• Main Authors: 謝世婷, Hsieh, Shih-Tyng
• Other Authors: 工業設計學系碩博士班, 謝孟達, Shieh, Meng-Dar
• Format: Thesis
• Language: English
• Published: 2013
• Subjects: 泌尿科; 輸尿管鏡手術; 語音介面; 電腦輔助手術; 設計; 使用性; Urology; Ureteroscopic Surgery; Voice User Interface; Computer Aided Surgery; Design; Usability; Kung; sami
• Online Access: http://ir.lib.ncku.edu.tw/handle/987654321/134353; http://ir.lib.ncku.edu.tw/bitstream/987654321/134353/-1/index.html

本研究設計與探討於手術室使用之醫用語音電腦輔助系統，取名為Sami，並進行輸尿管鏡手術之使用性測試以瞭解Sami與目前成大醫院使用之一般介面之差異。本研究首先搜集六種語音介面設計方針，並藉由觀察與訪談瞭解成大醫院泌尿科醫師的需求。在問卷調查醫師們對於需求的重要程度後，萃取出符合醫師需求之設計目標以開始設計流程。在第一設計階段，本研究基於設計方針設計第一版本的Sami。第二設計階段邀請了三位跨領域的專家參與?發式評估(Heuristic Evaluation)，專家們將以Nielsen的十項使用性準則與語音介面設計方針作為評估的基礎。評估後，針對專家的使用性評估結果改良完成第二版本的Sami。接著第三設計階段實施綠野仙蹤法 (Wizard of Oz Test)，受試對象為三位成大醫院泌尿科的手術醫師，醫師們將使用平常習慣的英文會話置入完成器材控制任務之語音指令資料庫。並在實驗完成後，參與訪談以表達對於Sami語音介面的建議。在最後的設計流程中確立系統架構與萃取出的語音指令關鍵字，並且經由Java程式語言完成Sami的實驗模型設計。 在使用性測試中，本研究邀請三位手術醫師參加測試。於使用性測試結果我們發現，縱然測試過程中未曾發生辨識錯誤的問題，但醫師們仍對於語音辨識技術層面持保守態度。此外，Sami確實可以幫助醫生專注於凝視病灶並進行醫療行為。在方便性與快速性兩項評估標準中，醫師們認為Sami語音介面相較於一般介面擁有較佳評價得分。 In this research there will be a design of computer aided surgery (CAS) system with voice user interface (VUI) Through a whole design process this research will propose Sami the designed CAS system with VUI Then usability test will be applied to Ureteroscopic surgery to realize the comparison result of conventional UI in National Cheng Kung University Hospital (NCKUH) and VUI of Sami First this research collected six VUI design guidelines Then by implementing observations and interviews requirements of the surgeons from Urology department in NCKUH were realized After the results of important requirements were investigated by questionnaires the design objects were defined and the design process was started In the first step of the design process the preliminary design of system framework was accomplished Three experts from three different academic fields were invited to join the second step Heuristic Evaluation Experts would examine the designed system framework with Nielsen's 10 usability principles and VUI design guidelines Then the revision design of Sami was proposed based on the results of Heuristic Evaluation In the third step three surgeons from NCKUH were invited to Wizard of Oz test (WOZ test) During the WOZ test surgeons used voice commands to complete tasks of controlling devices by English Furthermore interviews were applied to realize the satisfaction issues for Sami In the end of the design process the system framework of Sami as well as keywords of voice commands were defined Also the experimental prototype of Sami was constructed using Java Language In the usability test three surgeons were invited Though there was no error in voice recognition during usability test surgeons had conservative attitude about the technology of voice recognition in CAS system However Sami could provide an opportunity for surgeons to control interface without their hands which lets surgeons gaze at the nidus of the patient and execute the treatment In the comparison result surgeons considered that two evaluation criteria of VUI had a higher score than conventional UI which are convenience and quickness