Computer Vision News - November 2023

Computer Vision News 60 Pioneers of Medical Robotics camera design where the view is aligned with the endoscope axis and the vision sensors are placed at the distal end of the camera body. Our novel design allows us to change the (baseline) distance between the two vision sensors, without the need to increase the size of the incision where the camera is inserted into the patient’s abdomen. Our results showed that increasing this baseline distance improves the depth perception based on the views from our camera, with a precision of 2.5 mm. Moreover, robots, unlike humans, can focus on two (or more) different locations at the same time when performing a task autonomously. Such capability allows robots to perform multiple steps of the task in parallel. We leveraged the surgical robot’s capability to move multiple arms in parallel to devise autonomous execution models that go beyond the humans’ way of performing repetitive surgical tasks. In other words, we applied the “parallelism” concept to automate surgical tasks. This work challenges the dominant idea in autonomous robotic surgery literature that the best execution model to automate a surgical task is the one used by humans. We applied this idea to automate parts of the suturing task and showed that our work leads to an order of magnitude faster execution compared with the state-of-the-art methods. Moving forward, I will research how automation would fit into (in the short term) and revolutionize (in the long term) the current surgical practice in RAS. For example, I am interested in developing criteria with which one can decide if a surgical task is a good candidate for automation or not. I am also interested in developing the algorithms and systems needed to automate such tasks. My work will leverage other robot’s unique capabilities to design and deploy human-centered collaborative systems in RAS. Shan Lin, the other Pioneer of Medical Robotics selected by the workshop.

RkJQdWJsaXNoZXIy NTc3NzU=