Professor Paul Cardenas will present his work An ensemble learning methodology for predicting medical micro-robot degradation classes at the Annual European Conference on Safety and Reliability (Esrel). This research was developed in collaboration with Zeina Al Masry, from the FEMTO-ST institute, and Sergio Lescano, from Amarob Technologies.

The Esrel is an international conference supported by the European Association for Safety and Reliability (ESRA), to be held from 28 August to 1 September in Dublin, Ireland. It is considered the most important annual event in the areas of risk assessment and management and performance optimization of socio-technological systems in Europe. It is also one of the most important events of its kind, internationally.

The work models the degradation of the microrrobot to ensure optimal performance limits during surgical acts where its degradation status is predicted. Therefore, a new methodology is proposed to predict the degradation classes of the microrrobot, based on learning together, combined with autoconders and with attributes produced by data engineering in a heuristic way. The study shows that the proposed methodology is robust and very resistant to the presence of noise, which is very high in real scenarios.

More news