Computer vision for total laparoscopic hysterectomy
| Autor: | Masakazu Sato, Maki Nakata, Minoru Nakabayashi, Kei Inaba, Nao Itaoka, Yoko Hasumi, Natsuki Nagashima, Hiroshi Ki, Minako Koizumi, Chiharu Ueshima, Yu Takahashi |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Laparoscopic surgery
genetic structures medicine.medical_treatment ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Total laparoscopic hysterectomy Hysterectomy Field (computer science) 03 medical and health sciences 0302 clinical medicine Medicine Humans Computer vision Retrospective Studies Sutures business.industry Ethics committee Robotic Surgical Procedures General Medicine Open source Surgery Computer-Assisted 030220 oncology & carcinogenesis OS X 030211 gastroenterology & hepatology Augmented reality Female Laparoscopy Artificial intelligence Ureter business Algorithms |
| Zdroj: | Asian journal of endoscopic surgery. 12(3) |
| ISSN: | 1758-5910 |
| Popis: | Introduction Laparoscopic surgery is widely performed in various surgical fields, but this technique requires time for surgeons to master. However, at the same time, there are many advantages in visualizing the operative field through a camera. In other words, we can visualize what we cannot see with our own eyes by using augmented reality and computer vision. Therefore, we investigated the possibilities and usefulness of computer vision in total laparoscopic hysterectomy. Methods This study was approved by the Mitsui Memorial Hospital ethics committee. Patients who underwent total laparoscopic hysterectomy at Mitsui Memorial Hospital from January 2015 to December 2015 were enrolled. We evaluated 19 cases in which total laparoscopic hysterectomy was performed by the same operator and assistant. We used the Open Source Computer Vision Library for computer vision analysis. The development platform used in this study was a computer operating on Mac OS X 10.11.3. Results We created panoramic images by matching features with the AKAZE algorithm. Noise reduction methods improved haziness caused by using energy devices. By abstracting the color of the suture string, we succeeded in abstracting the suture string from movies. We could not achieve satisfactory results in detecting ureters, and we expect that creative ideas for ureter detection may arise from collaborations between surgeons and medical engineers. Conclusions Although this was a preliminary study, the results suggest the utility of computer vision in assisting laparoscopic surgery. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Plný text