醫學影像完整上傳是影像醫學部門的重要品質項目，任何影像的漏傳都可能造成病患診斷與治療的延誤，亦可能引發醫療糾紛，影響醫院聲譽。本研究以新竹縣地區醫院影像醫學部為例，該單位2023年5月至11月電腦斷層檢查量1607筆中有17筆檢查未完整上傳，完整上傳比率為98.94%。分析其原因有「無專人負責上傳」、「工作忙碌」、「口頭交班遺忘」、「網路阻擋上傳」等疏失導致無法完整上傳影像。本研究運用QC Story手法來分析無法完整上傳的主要原因，對策擬定與實施步驟中，運用卷積神經網路(CNN)建構電腦斷層影像分類系統，以人工智慧輔助核對影像，建立可視化上傳列表。最終經由「電腦斷層工作專責處理」、「開發可視化上傳列表」及「紀錄交班內容」三項策略實施後，達到影像完整上傳率100%的目標。效果確定後，建立標準化流程以維持改善成效，防止錯誤再發生。

The complete transmission of medical images is a critical quality measure for the radiology department. Failure to upload any image can cause delays in patients' diagnosis and treatment, potentially leading to medical disputes and damaging the hospital's reputation. This study uses the radiology department of a regional hospital in Hsinchu County as an example. During May and November 2023, out of 1607 CT scans, seventeen were not fully uploaded, resulting in a complete upload rate of 98.94%. The analysis identified several reasons for not fully uploading images, including a lack of dedicated personnel handling uploads, heavy workloads, forgetfulness during handovers, and network blockages. The study utilized the QC Story method to analyze the capital causes of incomplete uploads. In the strategy formulation and implementation steps, a tool called "Convolutional Neural Network (CNN)" was utilized to develop a CT image classification system. This system, assisted by artificial intelligence, streamlines image verification, and offers a visual upload interface, reducing manual verification time and minimizing human errors. By implementing three strategies— dedicated CT task handling, developing a visual upload interface, and creating handover records, the goal of achieving a 100% complete image upload rate was accomplished. After confirming the effectiveness, establishing a standardized process maintains the improvement results to prevent the recurrence of errors.

第一章 緒論 9
1.1 研究背景與動機 9
1.2 研究目的 9
1.3 研究架構 10
1.4 研究限制 10
第二章 文獻回顧 12
2.1 醫學影像儲存與傳輸系統 12
2.2 QC STORY 12
2.3 卷積神經網路 13
第三章 研究方法 16
3.1 QC STORY 16
3.1.1 PDCA循環 16
3.1.2 問題解決型QC STORY 17
3.2 卷積神經網路 18
3.2.1 模型架構 19
3.2.2 模型訓練與評估 20
第四章 個案研究 22
4.1 問題描述 22
4.2 運用QC STORY進行流程改善 22
4.2.1 主題選定 22
4.2.2 活動計畫擬定 24
4.2.3 現況把握及目標設定 25
4.2.4 要因解析 26
4.2.5 對策擬定與實施 28
4.2.6 效果確認 36
4.2.7 標準化 37
4.2.8 檢討與改進 37
4.3 績效比較 38
第五章 結論與未來發展 39
5.1 結論 39
5.2 本研究貢獻 39
5.3 未來研究 40

1.蘇朝墩，2010，品質管理，前程文化事業有限公司。
2.蘇朝墩，2002，品質工程，中華民國品質學會。
3.楊宗龍，陳為忠，潘慧本，楊建芳，2003，無片環境標準數位影像防漏機轉─論MPPS之必要性，中華放射線醫學雜誌，28(4)，243-247。
4.戴君妃，2021，應用QC STORY提升乳癌篩檢乳房攝影量-以某區域教學醫院為例，國立清華大學碩士論文。
5.游展碩，2023，基於CNN的影像處理技術來通報及記錄病患的需求，健行科技大學碩士論文。
6.Chen, Q., Jiang, M., Zeng, M., Jiesisibieke, Z. L., Chen, P. E., Chien, C. W., Tung, T. H., 2020, Constructing a new health education model for patients with chronic hepatitis B. , Medicine, 99(50), e23687.
https://doi.org/10.1097/MD.0000000000023687
7.Ding, S.W., Chang, C.W., Tsai, Y.F., Chen, C.C., Li, Z.Y., Chen, L.C., Chu, H.C., Chung, W.S., Li, W.C., 2020, Promoting the Quality of Teaching by Using Task-Oriented QC Story The Challenge of Making Teaching Materials Paperless, Journal of Healthcare Quality, 14(3), 52-57.
8.Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., Thrun, S., 2017, Dermatologist-level Classification of Skin Cancer with Deep Neural Networks, Nature, 542(7639), 115-118.
9.Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., Kim, R., Raman, R., Nelson, P. C., Mega, J. L., & Webster, D. R., 2016, Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA, 316(22), 2402-2410.
10.Ku, F. L., Huang, C. P., Tsai, C. C., Chen, T. W., 2018, Strategies to improve the completeness of a system of substitute therapist in a rehabilitation center, Annals of Physical and Rehabilitation Medicine, 61, 519-520.
11.Sajjad, M., Khan, S., Muhammad, K., Wu, W., Ullah, A., Baik, S. W., 2019, Multi-grade brain tumor classification using deep CNN with extensive data augmentation, Journal of Computational Science 30, 174-182.
12.Sha, D., Hsu,S., Chen, H.Y., 2016, Creative Problem-Solving QC Story Based on Systematic Innovation TRIZ, Journal of Engineering, 35(2), 123-130.
13.Zhou, X., Li,Y., Liang, W., 2020, CNN-RNN Based Intelligent Recommendation for Online Medical Pre-Diagnosis Support, IEEE/ACM Transactions on Computational Biology and Bioinformatics, 18(3), 912-921.
https://doi.org/10.1109/TCBB.2020.2994780