Unique ID issued by UMIN | UMIN000058609 |
---|---|
Receipt number | R000066962 |
Scientific Title | Evaluation of the usefulness of an intraoperative focus point automatic identification model using deep learning in laparoscopic cholecystectomy |
Date of disclosure of the study information | 2025/07/30 |
Last modified on | 2025/07/27 09:48:35 |
Evaluation of the usefulness of an intraoperative focus point automatic identification model using deep learning in laparoscopic cholecystectomy
Evaluation of the usefulness of an intraoperative focus point automatic identification model using deep learning in laparoscopic cholecystectomy
Evaluation of the usefulness of an intraoperative focus point automatic identification model using deep learning in laparoscopic cholecystectomy
Evaluation of the usefulness of an intraoperative focus point automatic identification model using deep learning in laparoscopic cholecystectomy
Japan |
Gallstone disease, Gallbladder polyps, Chronic cholecystitis
Surgery in general | Hepato-biliary-pancreatic surgery |
Others
NO
In recent years, deep learning has attracted attention from researchers in engineering and medical fields, not only in the field of image recognition, due to its high performance and wide applicability compared to conventional algorithms.In our previous research, we developed a model that automatically predicts a surgeon's gaze point during surgery based on labeling data of anatomical structures and artificial objects in surgical videos. Specifically, in laparoscopic cholecystectomy, we noted that the tip of the right-hand forceps is often close to the surgeon's gaze point and devised an algorithm (new algorithm) that predicts the gaze point based on the historical position information of the right-hand forceps tip.Originally, one of the objectives of constructing a system to predict gaze points was to automate camera operations during laparoscopic surgery and reduce the human resources required by surgeons. To achieve this, we considered it useful to fix the laparoscopic camera externally and proceed with surgery while enlarging the gaze point. Furthermore, by combining this with technology that adjusts the enlarged field of view via voice input, we established a system that can be used in actual surgery.As we aim to introduce this system into clinical practice, the objective of this study is to evaluate the usefulness of the aforementioned new algorithm. While it is possible to perform surgery by simply tracking the tip of the right-hand forceps as the point of focus, the use of the new algorithm is expected to enable camera movements that more closely resemble those performed by surgeons during actual laparoscopic surgery.
Safety,Efficacy
We will compare the times deemed appropriate and inappropriate for surgical videos for each video and evaluate the usefulness of the new algorithm and select the most appropriate algorithm. We will also compare postoperative outcomes such as the length of hospital stay including patient information, the incidence of complications within 30 days, and the readmission rate within 90 days.
Observational
18 | years-old | <= |
Not applicable |
Male and Female
New patient cohort: Patients who undergo laparoscopic cholecystectomy at the Department of Hepatobiliary and Pancreatic Surgery and Organ Transplantation, The University of Tokyo Hospital, between the approval date and November 30, 2026, and who have provided informed consent.Specifically, this study excludes cases with severe gallbladder inflammation, such as acute cholecystitis following PTGBD or cholangitis with yellow granuloma, and instead focuses on cases with mild gallbladder inflammation, such as gallbladder polyps, gallstone disease, or mild chronic cholecystitis.
Control group: Patients who underwent laparoscopic cholecystectomy at the Department of Hepatobiliary and Pancreatic Surgery, Tokyo University Hospital, between January 2008 and May 2025. The study focuses on cases with minimal gallbladder inflammation.
Patients aged 17 years or younger. Patients who are unable to give consent. Patients who have previously undergone laparoscopic cholecystectomy and have refused to participate.
20
1st name | Kiyoshi |
Middle name | |
Last name | Hasegawa |
The University of Tokyo Hospital
Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery
113-0033
7-3-1 Hongo, Bunkyo-ku, Tokyo
03-3815-5411
kihase-tky@umin.ac.jp
1st name | Ryo |
Middle name | |
Last name | Oikawa |
The University of Tokyo Hospital
Hepato-Biliary-Pancreatic Surgery Division, Department of Surgery
113-0033
7-3-1 Hongo, Bunkyo-ku, Tokyo
03-3815-5411
oikawar-sur@h.u-tokyo.ac.jp
The University of Tokyo hospital
Ministry of Education, Culture, Sports, Science and Technology
Japanese Governmental office
Ethics Committee, Graduate School of Medicine, The University of Tokyo
7-3-1 Hongo, Bunkyo-ku, Tokyo
03-5841-0818
ethics@m.u-tokyo.ac.jp
NO
東京大学医学部附属病院(東京都)
2025 | Year | 07 | Month | 30 | Day |
Unpublished
Open public recruiting
2025 | Year | 07 | Month | 17 | Day |
2025 | Year | 07 | Month | 17 | Day |
2025 | Year | 07 | Month | 17 | Day |
2026 | Year | 11 | Month | 30 | Day |
1 This study includes patients who were determined to be eligible for laparoscopic cholecystectomy within the scope of routine clinical practice at the Department of Hepatobiliary and Pancreatic Surgery and Artificial Organ Transplantation, Tokyo University Hospital.
2 Prior to the procedure, patients undergoing the surgery will receive preoperative explanations about the procedure and obtain informed consent. Following this, they will be informed about the study and obtain informed consent for participation.
3 The surgery will be performed by the surgeon, assistant, and laparoscopic surgeon as usual. Two monitors will be used, with one displaying the standard camera view and the other showing real-time magnification of the right-hand forceps tip.
4 After port insertion and adhesion separation are completed, and the standard cholecystectomy technique is initiated, the surgeon will fix the scope using an external scope holder and proceed with the surgery while viewing the magnified image of the right-hand forceps tip or fixed images as needed, starting from approximately 5 minutes after the initiation of the cystic duct neck processing and approximately 5 minutes after the division of the cystic duct and cystic artery.During all other times, the scope operator operates the scope as usual to proceed with the surgery.
5 After surgery, the enlarged images from the right-hand forceps tip and the fixed images are recorded. By applying a new algorithm with multiple patterns to the fixed images, additional enlarged images are created.
6 While the surgeon reviews the surgical video, they are asked to continuously press a button at their fingertips during scenes deemed inappropriate as the surgical field and release it during appropriate scenes.
7 Compare the times deemed appropriate or inappropriate for each video to evaluate the usefulness of the new algorithm and select the most suitable algorithm.
2025 | Year | 07 | Month | 27 | Day |
2025 | Year | 07 | Month | 27 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000066962