簡易檢索 / 詳目顯示

研究生: 郭昱辰
Kuo, Yu-Chen
論文名稱: 結合電腦視覺與人機協作之經鼻內視鏡顱底手術暴露範圍與長度量測輔助系統
Human-Computer Vision Collaborative Measurement of Surgical Exposure and Length in Endonasal Endoscopic Skull Base Surgery
指導教授: 梁勝富
Liang, Sheng-Fu
學位類別: 碩士
Master
系所名稱: 電機資訊學院 - 資訊工程學系
Department of Computer Science and Information Engineering
論文出版年: 2026
畢業學年度: 114
語文別: 中文
論文頁數: 47
中文關鍵詞: 電腦視覺人機協作經鼻內視鏡手術腦下垂體巨腺瘤手術暴露範圍
外文關鍵詞: computer vision, human–computer collaboration, endonasal endoscopic approach, pituitary macroadenoma, surgical exposure
相關次數: 點閱:4下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 經鼻內視鏡手術(endonasal endoscopic approach, EEA)廣泛應用於鞍區與鞍旁區病灶,但手術暴露範圍與操作技巧的定量比較仍相當耗時,既有研究多仰賴屍體解剖或神經導航,本研究分析 50 部腦下垂體巨腺瘤手術影片,影片平均長度為 94.07 分鐘,範圍為 59 分鐘至 3.14 小時,影片分為參數設定資料集 10 部、驗證資料集一 20 部及驗證資料集二 20 部。
    本研究建立使用者導引式人機協作電腦視覺(computer vision, CV)量測系統,以畫面中的 6、8、10 或 12 Fr Frazier 吸引管作為比例參考,使用者標註長度端點或目標區域多邊形後,系統依像素數量計算長度與面積,並以參數設定資料集建立水平與垂直方向的經驗性校正係數,固定套用於兩個驗證資料集,再以神經導航量測的兩側內頸動脈間距及蝶鞍高度驗證解剖距離量測結果。
    電腦視覺與神經導航在兩側內頸動脈間距及蝶鞍高度的量測值均未觀察到顯著差異(p = 0.829 及 p = 0.671),依兩項解剖距離計算之相對量測準確度,在參數設定資料集、驗證資料集一及驗證資料集二分別為 95.1%、95.8% 及 96.2%,整體為 95.8%,電腦視覺與神經導航的平均長度量測時間分別為 75.8 秒與 76.1 秒,兩者未觀察到顯著差異(p = 0.913),平均面積量測時間則分別為 80.5 秒與 674.6 秒(p < 0.001),相較於神經導航,電腦視覺的面積量測時間縮短 88.1%。
    殘餘腫瘤組的橫向暴露(p = 0.001)與校正後二維投影面積(p = 0.024)較小,縱向暴露在兩組間未觀察到顯著差異(p = 0.767),且術前腫瘤體積-暴露面積比值較高(p = 0.004),三種手術技巧間及早期組與經驗組間的橫向暴露、縱向暴露與校正後二維投影面積均未觀察到顯著差異,本系統可量化 EEA 手術影片中的解剖距離與校正後二維投影面積,並縮短面積量測時間,供術後影片回顧及手術技巧與術者經驗之探索性分析。

    Quantitative assessment of surgical exposure during the endonasal endoscopic approach (EEA) is time-consuming and commonly relies on cadaveric studies or neuronavigation. We developed a user-guided computer-vision system that uses visible 6-, 8-, 10-, or 12-Fr Frazier suction tubes as scale references to measure anatomical length and corrected two-dimensional projected exposure area in operative videos. Fifty EEA procedures for pituitary macroadenoma were divided into a 10-case parameter-setting dataset and two 20-case validation datasets. Intercarotid distance and sellar height measured by neuronavigation served as anatomical reference values.
    Computer-vision and neuronavigation measurements did not differ significantly for intercarotid distance (p = 0.829) or sellar height (p = 0.671), and relative measurement accuracy was 95.1%, 95.8%, and 96.2% across the three datasets. Length-measurement time was comparable (75.8 vs. 76.1 seconds, p = 0.913), whereas area-measurement time decreased by 88.1% (80.5 vs. 674.6 seconds, p < 0.001). Residual tumor was associated with smaller lateral exposure (p = 0.001), smaller corrected projected exposure area (p = 0.024), and a higher preoperative-tumor-volume-to-exposure-area ratio (p = 0.004), but not with longitudinal exposure (p = 0.767). Exposure did not differ among operative techniques or surgeon-experience groups. This system provides accurate, efficient measurements for postoperative video review and exploratory comparison of surgical exposure.

    摘要 i SUMMARY ii 誌謝 v 目錄 vii 表目錄 ix 圖目錄 x 第壹章 緒論 1 1.1 研究背景 1 1.2 手術暴露量測與手術影片分析 1 1.3 既有方法之限制與研究動機 2 1.4 研究目的 3 第貳章 研究方法 4 2.1 研究設計與流程概述 4 2.2 研究資料、手術影片與比較基準 5 2.3 系統模組與人機協作流程 6 2.4 軟體介面與系統功能 7 2.5 參考器械與影像量測方法 9 2.5.1. 長度量測 9 2.5.2. 面積量測 10 2.5.3. 方向性校正 11 2.6 影像品質控制與處理原則 13 2.7 系統驗證與統計分析 14 第參章 研究結果 16 3.1 量測輔助系統驗證 16 3.2 系統輸出與代表性案例 18 3.2.1. 參數設定資料集案例:完整量測流程 18 3.2.2. 驗證資料集一案例:目標區域與雙向長度量測 21 3.2.3. 驗證資料集二案例:不同長度量測模式 23 3.3 臨床探索結果 24 3.3.1. 手術暴露與殘餘腫瘤 24 3.3.2. 不同手術技巧與術者經驗 24 第肆章 討論 26 4.1 主要貢獻與結果解釋 26 4.2 與既有研究比較 27 4.3 研究限制 29 4.4 未來方向 30 第伍章 結論 31 參考文獻 32

    Abdelsattar, J. M., Pandian, T. K., Finnesgard, E. J., El Khatib, M. M., Rowse, P. G., Buckarma, E. H., … Farley, D. R. (2015). Do you see what I see? How we use video as an adjunct to general surgery resident education. Journal of Surgical Education, 72(6), e145–e150. https://doi.org/10.1016/J.JSURG.2015.07.012

    Agha, R., Abdall-Razak, A., Crossley, E., Dowlut, N., Iosifidis, C., & Mathew, G. (2019). STROCSS 2019 guideline: Strengthening the reporting of cohort studies in surgery. International Journal of Surgery, 72, 156–165. https://doi.org/10.1016/J.IJSU.2019.11.002

    Ali, S., Bailey, A., Ash, S., Haghighat, M., Allan, P., Ambrose, T., … Braden, B. (2021). A pilot study on automatic three-dimensional quantification of Barrett’s esophagus for risk stratification and therapy monitoring. Gastroenterology, 161(3), 865–878.e8. https://doi.org/10.1053/J.GASTRO.2021.05.059

    Bove, I., Solari, D., Bruneau, M., Berhouma, M., Jouanneau, E., Cornelius, J. F., … Cavallo, L. M. (2023). Endoscopic endonasal pituitary surgery: How we do it. consensus statement on behalf of the EANS skull base section. Brain and Spine, 3, 102687. https://doi.org/10.1016/J.BAS.2023.102687

    Castle-Kirszbaum, M., Wang, Y. Y., King, J., & Goldschlager, T. (2022). Predictors of visual and endocrine outcomes after endoscopic transsphenoidal surgery for pituitary adenomas. Neurosurgical Review, 45(1), 843–853. https://doi.org/10.1007/S10143-021-01617-Y

    Cebula, H., Kurbanov, A., Zimmer, L. A., Poczos, P., Leach, J. L., De Battista, J. C., … Keller, J. T. (2014). Endoscopic, endonasal variability in the anatomy of the internal carotid artery. World Neurosurgery, 82(6), e759–e764. https://doi.org/10.1016/J.WNEU.2014.09.021

    Chadebecq, F., Vasconcelos, F., Mazomenos, E., & Stoyanov, D. (2020). Computer vision in the surgical operating room. Visceral Medicine, 36(6), 456–462. https://doi.org/10.1159/000511934

    Chiurillo, I., Sha, R. M., Robertson, F. C., Liu, J., Li, J., Le Mau, H., … Stolyarov, R. (2023). High-accuracy neuro-navigation with computer vision for frameless registration and real-time tracking. Bioengineering, 10(12), 1401. https://doi.org/10.3390/bioengineering10121401

    Eugen, H., & Donciu, C. (2020). Video distance measurement based on focus. In 2020 international conference and exposition on electrical and power engineering (pp. 582–588). https://doi.org/10.1109/EPE50722.2020.9305593

    Evins, A. I., Sistiaga, I. L., Quispe-Flores, A. H., Castro, M. M., Atchley, T. J., Perez-Fernandez, S., … Bernardo, A. (2024). A quantitative comparative surgical analysis of the endoscopic transorbital approach and frontotemporal-orbitozygomatic approach for extradural exposure of the cavernous sinus. Neurosurgical Focus, 56(4), E4. https://doi.org/10.3171/2024.1.FOCUS23860

    Gopalakrishnan, A., Ganesan, S., Ramesh, A. S., Ramesh, A., Penubarathi, L. K., Raja, K., … Alexander, A. (2024). Do anatomical variations of sphenoid sinus influence sella exposure and residual disease in pituitary surgery? a study in an indian population. International Archives of Otorhinolaryngology, 28(4), e643–e649. https://doi.org/10.1055/s-0044-1788313

    Gujjunoori, S., Reddy, S. S. S., & Gouthaman, K. V. (2017). Tracking and size estimation of objects in motion. In 2017 international conference on machine vision and information technology (pp. 87–92). https://doi.org/10.1109/CMVIT.2017.8

    Khan, D. Z., Luengo, I., Barbarisi, S., Addis, C., Culshaw, L., Dorward, N. L., … Marcus, H. J. (2022). Automated operative workflow analysis of endoscopic pituitary surgery using machine learning: development and preclinical evaluation (IDEAL stage 0). Journal of Neurosurgery, 137(1), 51–58. https://doi.org/10.3171/2021.6.JNS21923

    Kutlay, M., Durmaz, A., Ozer, I., Kural, C., Temiz, C., Kaya, S., … Izci, Y. (2018). Extended endoscopic endonasal approach to the ventral skull base lesions. Clinical Neurology and Neurosurgery, 167, 129–140. https://doi.org/10.1016/J.CLINEURO.2018.02.032

    Lalys, F., Riffaud, L., Morandi, X., & Jannin, P. (2010). Automatic phases recognition in pituitary surgeries by microscope images classification. In Medical image computing and computer-assisted intervention (Vol. 6135, pp. 34–44). https://doi.org/10.1007/978-3-642-13711-2_4

    Leach, P., Abou-Zeid, A. H., Kearney, T., Davis, J., Trainer, P. J., & Gnanalingham, K. K. (2010). Endoscopic transsphenoidal pituitary surgery: evidence of an operative learning curve. Neurosurgery, 67(5), 1205–1212. https://doi.org/10.1227/NEU.0B013E3181EF25C5

    Lofrese, G., Vigo, V., Rigante, M., Grieco, D. L., Maresca, M., Anile, C., … De Bonis, P. (2018). Learning curve of endoscopic pituitary surgery: Experience of a neurosurgery/ENT collaboration. Journal of Clinical Neuroscience, 47, 299–303. https://doi.org/10.1016/J.JOCN.2017.09.011

    Loymak, T., Belykh, E., Abramov, I., Tungsanga, S., Sarris, C. E., Little, A. S., & Preul, M. C. (2022). Comparative analysis of surgical exposure among endoscopic endonasal approaches to petrosectomy: An experimental study in cadavers. Journal of Neurological Surgery Part B: Skull Base, 83(5), 526–535. https://doi.org/10.1055/S-0041-1741067

    Marigil Sanchez, M., Karekezi, C., Almeida, J. P., Kalyvas, A., Castro, V., Velasquez, C., & Gentili, F. (2019). Management of giant pituitary adenomas: Role and outcome of the endoscopic endonasal surgical approach. Neurosurgery Clinics of North America, 30(4), 433–444. https://doi.org/10.1016/J.NEC.2019.05.004

    Noiphithak, R., Yanez-Siller, J. C., Barbero, J. M. R., Otto, B. A., Carrau, R. L., & Prevedello, D. M. (2018). Quantitative analysis of the surgical exposure and surgical freedom between transcranial and transorbital endoscopic anterior petrosectomies to the posterior fossa. Journal of Neurosurgery, 131(2), 569–577. https://doi.org/10.3171/2018.2.JNS172334

    Park, H. R., Kshettry, V. R., Farrell, C. J., Lee, J. M., Kim, Y. H., Won, T. B., … Paek, S. H. (2017). Clinical outcome after extended endoscopic endonasal resection of craniopharyngiomas: Two-institution experience. World Neurosurgery, 103, 465–474. https://doi.org/10.1016/j.wneu.2017.04.047

    Patel, D., Saindane, A., Oyesiku, N., & Hu, R. (2020). Variant sella morphology and pituitary gland height in adult patients with Chiari II malformation: potential pitfall in MRI evaluation. Clinical Imaging, 64, 24–28. https://doi.org/10.1016/J.CLINIMAG.2020.02.014

    Pillai, P., Kurozumi, K., Filipce, V., Tang, M., Mackeise, O., & Ammirati, M. (2009). Quantitative analysis of surgical exposure and surgical freedom obtained by endoscope and microscope in the endonasal transsphenoidal approaches to sella using CT-based frameless stereotaxy: An anatomical study. Skull Base, 19(1), A042. https://doi.org/10.1055/S-2009-1222157

    Prajapati, H. P., Jain, S. K., & Sinha, V. D. (2018). Endoscopic versus microscopic pituitary adenoma surgery: An institutional experience. Asian Journal of Neurosurgery, 13(2), 217–221. https://doi.org/10.4103/AJNS.AJNS_160_16

    Qureshi, T., Chaus, F., Fogg, L., Dasgupta, M., Straus, D., & Byrne, R. W. (2016). Learning curve for the transsphenoidal endoscopic endonasal approach to pituitary tumors. British Journal of Neurosurgery, 30(6), 637–642. https://doi.org/10.1080/02688697.2016.1199786

    Ryu, J., Choi, J., & Kim, H. C. (2013). Endoscopic vision-based tracking of multiple surgical instruments during robot-assisted surgery. Artificial Organs, 37(1), 107–112. https://doi.org/10.1111/J.1525-1594.2012.01543.X

    Safavian, N., Toh, S. K. C., Pani, M., & Lee, R. (2024). Endoscopic measurement of the size of gastrointestinal polyps using an electromagnetic tracking system and computer vision-based algorithm. International Journal of Computer Assisted Radiology and Surgery, 19(2), 321–329. https://doi.org/10.1007/S11548-023-03011-Z

    Tang, C. T., Kurozumi, K., Pillai, P., Filipce, V., Chiocca, E. A., & Ammirati, M. (2013). Quantitative analysis of surgical exposure and maneuverability associated with the endoscope and the microscope in the retrosigmoid and various posterior petrosectomy approaches to the petroclival region using computer tomography-based frameless stereotaxy: a cadaveric study. Clinical Neurology and Neurosurgery, 115(7), 1058–1062. https://doi.org/10.1016/J.CLINEURO.2012.10.023

    Tsuneoka, H., Tosaka, M., Yamaguchi, R., Tanaka, Y., Mukada, N., Shimauchi-Ohtaki, H., … Yoshimoto, Y. (2023). The significance of the intercarotid distances for transsphenoidal pituitary surgery: A magnetic resonance imaging study. World Neurosurgery, 175, e704–e712. https://doi.org/10.1016/J.WNEU.2023.04.009

    Vakil, N. (1995). Measurement of lesions by endoscopy: an overview. Endoscopy, 27(9), 694–697. https://doi.org/10.1055/S-2007-1005790

    Wong, C. E., Chen, P. W., Hsu, H. J., Cheng, S. Y., Fan, C. C., Chen, Y. C., … Liang, S. F. (2024). Collaborative human-computer vision operative video analysis algorithm for analyzing surgical fluency and surgical interruptions in endonasal endoscopic pituitary surgery: Cohort study. Journal of Medical Internet Research, 26, e56127. https://doi.org/10.2196/56127

    Wong, C. E., Kuo, Y. C., Huang, D. W., Chen, P. W., Hsu, H. J., Lee, W. T., … Liang, S. F. (2025). Human-computer vision collaborative measurement of surgical exposure and length in endonasal endoscopic skull base surgery. IEEE Open Journal of Engineering in Medicine and Biology, 6, 480–487. https://doi.org/10.1109/OJEMB.2025.3587947

    Wong, C. E., Wang, W. H., Lan, M. Y., Lee, P. H., Huang, C. C., Su, P. F., & Lee, J. S. (2022). Predicting the need for desmopressin treatment during inpatient and after discharge following endoscopic sellar surgery. Frontiers in Neurology, 13, 843646. https://doi.org/10.3389/fneur.2022.843646

    Xu, H., Han, T., Wang, H., Liu, S., Hou, G., Sun, L., … Zhou, J. (2022). Detection of blood stains using computer vision-based algorithms and their association with postoperative outcomes in thoracoscopic lobectomies. European Journal of Cardio-Thoracic Surgery, 62(5), ezac154. https://doi.org/10.1093/EJCTS/EZAC154

    Zhu, Y., Liu, W., Zhang, L., & Hu, B. (2024). Endoscopic measurement of lesion size: An unmet clinical need. Chinese Medical Journal, 137(4), 379–381. https://doi.org/10.1097/CM9.0000000000002882

    QR CODE