02.11.2016

Manish Sahu received the 2nd prize at the "Modeling and Monitoring of Computer Assisted Interventions" (M2CAI) 2016 Challenge on Surgical Tool Presence Detection. He was awarded on October 21, 2016 in the course of the M2CAI workshop, held at the "Medical Image Computing and Computer Assisted Intervention" (MICCAI) Conference in Athens, Greece.

The objective of the challenge was an online detection of surgical tools in endoscopic images. Following machine learning approaches for tool presence detection, the team proposed a deep neural network architecture and trained it through images obtained from a set of endoscopic videos. The analysis on the challenge dataset yields a mean average precision in tool detection of slightly above 60% which places them in 2nd position.

Manish's research is funded by the German Ministry of Education & Research (BMBF) via the BiOPAss project (Image, ontology and process based assistance for minimally invasive endoscopic surgery) within the BMBF program "Adaptive, lernende Systeme - Für eine verständliche Interaktion zwischen Mensch und komplexer Technik".