On Nov 24th 2020 we present two interesting lectures.
Medical Image Analysis in the AMC Quantitative Medical Image Analysis (QIA) group
by Prof. Ivana Išgum (UMC, Amsterdam)
Analysis of medical images is an essential component in patient diagnosis and prognosis. Recent increase in the number of acquired medical images has led to a tremendous increase in the expert workload for image interpretation. Deep learning methods have shown the potential to automate the routine analysis and to support clinical research, thereby alleviating expert workload. In this presentation, AI methods for medical image analysis enabling patient diagnosis and prognosis developed in the AMC QIA group will be discussed. First, our recently developed methods for improving image quality and subsequent analysis of cardiac CT and MR enabling detection of heart disease in cardiac CT and MR will be shown. Next, methods for detection and quantification of stroke in infant brain MR will be presented. Finally, our research on the quantification of gastrointestinal motility in MR exams will be shown.
Next Move in Movement Disorders (NEMO)
by Dr. Ioannis Giotis (ZiuZ Visual Intelligence, Gorredijk)
NEMO is a research collaboration between the University Medical Centre Groningen and ZiuZ Visual Intelligence. The aim of NEMO is to distinguish different hyperkinetic movement disorders by means of computer vision, pattern recognition and machine learning techniques. Hyperkinetic movement disorders are disorders characterized by an excess of involuntary movements, such as tremor, myoclonus, dystonia, tics, chorea, spasticity and ataxia. In practice, even for trained neurologists these different types of involuntary movements are not always easy to recognize and distinguish from each other.
In NEMO we are using different data modalities, such as 3D video, motion sensors and surface electromyography in order to represent and model the different movement disorders and their symptoms (phenotypes). In this talk I will present the medical (i.e. developing a controlled protocol of movement tasks) and technical advances (data acquisition, feature extraction and classification methods) we were able to realize in this project up to this point and the outlook for its future development.
24 Nov: Advances in Medical Deep Learning – Virtual Fall Meeting Session 2 of 3
Share this Post