Endotracheal intubation (ETI) is a high-stakes, lifesaving procedure that is necessary when patients need assistance in breathing effectively. The procedure involves the insertion of a breathing tube through the mouth or nose to keep the airway open and to deliver oxygen. Failure to complete the procedure correctly could lead to a high risk of death or permanent neurologic impairment. With the use of video laryngoscopy for ETI, there are sizable data archives of successful and failed intubation attempts. The goal of this SuperUROP project is to use these data archives to develop a computer vision model to aid physicians in the ETI procedure. The model will automatically detect anatomical landmarks in real-time and suggest trajectories for optimal placement of the endotracheal tube. This project involves a collaboration with colleagues from Boston Children’s Hospital.

I joined SuperUROP because I am interested in applying hands-on what I have learned through my coursework at MIT through my own research project. Through my SuperUROP experience, I hope to advance my knowledge in computer vision, learn fundamental research-based skills, and grow as an independent researcher. I believe there is great potential in the applications of AI in healthcare to enhance patients’ lives, and I hope to be part of future advancements in this promising field.