The Improvement of Object Detection and Localization for Autonomous Camera Movement; An Investigation of the Social Implications of Medical Robots

Artificial intelligence and deep learning are continually improving processes in almost every field there is. One such field is the field of medicine. My capstone research focuses on the improvements that can be made to the perception component of the RAVEN-II surgical robot, in order to localize and detect objects of interest in its drylab workspace. Upon completion of the technical portion of my thesis, these improvements led to a higher precision in the automated movement of the ZED mini camera. This was achieved through image augmentation, which is the artificial expansion of a dataset, and more generalized labeling. The specific augmentation techniques used include flipping left/right 50% of the time and canny edges. The generalized labeling eliminated the need for distinguishing between the left and right graspers and the colors of the blocks.

This research directly corresponded with my STS research, where the impacts that medical robots would have on our healthcare system and society are explored. While they have been proven to be a valuable tool, accidents with devastating repercussions have occurred. This raises many important questions like who should be held accountable when the technology malfunctions? How would this affect hospitals, insurance companies, surgeons, and patients and how comfortable are patients with surgical robots? The answers to these questions are sought with the help of SCOT theory, users vs. nonusers, and case studies.

As with any technology, surgical robots influence society and society influences its development and future phases. Due to the spike in minimally invasive surgeries, surgical robotics is becoming a focal point. It will be interesting to see where this technology is headed, but we should also remain somewhat cautious as it progresses into its more advanced stages.

School of Engineering and Applied Sciences
Bachelor of Science in Computer Science
Technical Advisor: Homa Alemzadeh
STS Advisor: Richard Jacques