On Nov 17th 2020 we present two interesting lectures.
Deep Learning in Image Diagnostics at DIAG
by Prof. Bram van Ginneken (DIAG Radboud UMC, Nijmegen )
For more than half a century, researchers have attempted to analyze medical images with a computer as accurately and precisely as human experts can do this. Only as of recently, after the field switched from machine learning techniques to deep learning techniques, this has actually become achievable. What’s more, it has become scalable, i.e. such systems can be developed quickly when a sufficient amount of well-annotated training data is available. This development will be illustrated with examples from ophthalmology, pathology, and radiology that my group has been involved in. I believe this will have a profound impact on medicine, but also on the field of medical image analysis research.
Dementia: brain imaging, machine learning and open science
by Dr. Esther Bron (University Medical Center, Rotterdam)
Brain diseases – including dementia and stroke – impose an enormous burden to the individual and to society. As a consequence, there is an urgent need to develop effective preventive and therapeutic strategies. It is therefore essential to improve the understanding of the progression of diseases, patient selection in clinical trials, and patient monitoring in clinical practice and clinical trials. Neuroimage analysis and machine learning play a herein a crucial role, i.e. for developing robust quantitative brain imaging biomarkers and for developing data-driven models for diagnosis and prediction.
In this talk, Dr. Esther Bron will present some of her group’s research on methodology for quantifying brain imaging biomarkers, machine learning approaches for aiding diagnosis of brain diseases and novel approaches for modeling and predicting disease progression. But most importantly, she will focus on the validation of such methods and how proper validation is essential for translation to clinical practice and use in clinical trials. Therefore, Dr. Bron will also present the international TADPOLE grand challenge that compared state-of-the-art prediction methods. In addition, she initiated the TADPOLE-SHARE project in which various international research teams collaborate to make the prediction algorithms available for the scientific community.
Share this Post