Biography
Dr. Hilal Al-Libawy
Dr. Hilal Al-Libawy
University of Babylon, Iraq
Title: Federated learning in Medical Imaging
Abstract: 
Different issues, such as privacy, computational cost or data heterogeneity were the reasons behind the creation of Federated Learning (FL) framework. FL is a collaborative framework built as a distributed approach to manage training of machine learning and deep learning algorithms in distributed manner without the need for gather all the data in one place. The core mechanism of FL is to handle each set of data locally and the collaboration will be in the level of model and not the data which in turn will keep the data protected against any privacy risks. At the same time, medical imaging devices cannot offer big dataset in individual sites and again because of privacy concerns, it is not allowed to gather medical imaging dataset in one place. The success of smart artificial intelligence model usually depends on size of the dataset especially when using deep learning algorithms. Large number of medical dataset images can be collected from different sites based on Internet of Medical Things (IoMT). So, FL can be the best solution to produce an accurate and efficient deep learning model and in the same time keep the privacy of patient data. This solution involves aggregating training results from multiple sites to create a global model without directly sharing datasets. However, the most challenge of FL in medical imaging field is rises because of the heterogeneity of the medical devices that create heterogeneous images. This challenge is under comprehensive research efforts nowadays.
Biography: 
Assistant Professor Hilal Al-Libawy received BSc degree in Electrical Engineering from Baghdad University, Baghdad, Iraq, in 1991, MSc degree in electronic engineering in 1995. He is a teaching staff in Babylon University, Babylon, Iraq since 2004 till now. Al-Libawy has received his PhD certificate in behavioral analysis and operator fatigue studies in 2018 in the University of Liverpool, Liverpool, UK. His main areas of research interest are behavioral analysis, operator fatigue detection, deep learning, Federated learning  and biological, cognitive modeling including ACT-R architecture and FPGA optimised application.