Training U-net model

Background

Regional nerve block is a common anaesthesia technique used for surgery on the extremities. A successful block requires excellent anaesthesia experience including the ability to identify the appropriate nerves and surrounding tissues on ultrasound and good skills with a needle.

Previous studies have primarily focussed on the usage of ultrasound which has shown that ultrasound increases the success rate of regional nerve blocks. Some studies, however, have found that even with ultrasound assistance, a relatively high failure rate persists. This failure rate has largely been attributed to operators with limited experience and insufficient ultrasound skills.

A failed nerve block not only results in a bad experience for the patient, it might even lead to damage to the patients’ health and in some cases complications could even threaten life.

It’s essential to recognise ultrasound anatomy when performing nerve blocks, however this may sometimes be hampered by patients’ habitus.

Aim

In this study we will attempt to use the dataset of ultrasound images depicting the brachial plexus to train an U-net model in order to identify the region of interest in these images, which may potentially be used in clinical practise.

Primary objective:
1. Train an U-net deep learning model using the dataset and teach it to properly identify the brachial plexus

People

Rob Tolboom

Rob Tolboom

Radboudumc, Nijmegen

Nicola Pezzotti

Nicola Pezzotti

Assistant Professor

TU Eindhoven / Philips Research

Ruud van Sloun

Ruud van Sloun

Assistant Professor

TU Eindhoven / Philips Research