The first Medical Imaging Deep Learning (MIDL) conference will take place in our home town Amsterdam. Bringing together the leading researchers and experts will generate new ideas and foster innovation in this fast-developing field. Innovation that will ultimately lead to improved outcomes in healthcare.
Aidence is a sponsor and active participant of this event. Tim Salimans, Research Scientist at OpenAI and Scientific Advisor and co-founder of Aidence is one of MIDL’s keynote speakers. Tim’s talk covers the deep learning techniques Aidence uses to obtain state of the art accuracy of the detection and quantification of lung nodules on chest CT. He will also give an insight in the requirements and challenges faced when developing a deep learning system for use in clinical practice.
Marysia Winkels, data engineer at Aidence, will present a paper on 3D G-CNNs for Pulmonary Nodule Detection during the session on alternative deep learning architectures. The paper is the result of a Master thesis supervised by Taco Cohen and Max Welling from the University of Amsterdam, in collaboration with Aidence. The paper introduces a new type of convolutional layer based on group theory that significantly reduces the amount of training data required by the algorithm to successfully recognise pulmonary nodules. In addition, G-CNNs show a higher sensitivity towards malignant nodules, and have the added benefit of converging faster than regular CNNs.
Aidence is proud of Tim and Marysia‘s contribution to research in Deep Learning for Medical Imaging. Aidence is and intends to stay at the forefront of using the latest Deep Learning techniques. Therefore we continuously work with Key Opinion Leaders in this field.
To stay at the forefront we welcome research interns, graduates and experts to join our team.
We look forward to meeting you in the beginning of July in Amsterdam!