Gerardo Loza

Research Topic: Characterisation of tissues and tools in images from robotic minimally invasive procedures for interaction modeling.

Supervisors: Professor Pietro Valdastri, Dr Sharib Ali

About Gerardo: I graduated from The Interdisciplinary Professional Unit in Engineering and Advanced Technologies in Mexico with an undergraduate degree in Bionics. I worked in Siemens Healthineers for one year as a field service engineer and in TATA Consultancy Services for over one year as a software engineer for a project in Kaiser Permanent.

Project description: Robotic minimally invasive surgery (RMIS) is one crucial procedure in treating and diagnosing many diseases. Enhanced perception capabilities, alongside advanced automated technology, will lead to an active collaboration between robots and surgeons. However, the need for a precise surgical scene description during the operations hinders the effort to provide autonomy. This project aims to provide a semantic understanding of the surgical scene by a) applying unsupervised and weekly supervised learning for segmenting, tracking, and characterizing entities (tools and tissues) within laparoscopic images, b) modeling the interaction between entities, and c) predicting future interactions.