Deep learning channel

If you're interested in federated learning, particularly in medical imaging, we invite you to join our seminar tomorrow (Friday) at 11:00 a.m. Iran time! Zoom: https://oist.zoom.us/j/95908496615?pwd=akxZNmprLzNXY212TFh0ZWQ1ZlNyUT09
Meeting ID: 959 0849 6615
Passcode: 767685

Speaker: Prof. Shadi Albarqouni, Computational Medical Imaging Research, University of Bonn

Title: Unlocking the Potential of Federated Learning in Medical Imaging


Abstract: Deep Learning (DL) stands at the forefront of artificial intelligence, revolutionizing computer science with its prowess in various tasks, especially in computer vision and medical applications. Yet, its success hinges on vast data resources, a challenge exacerbated in healthcare by privacy concerns. Enter Federated Learning, a groundbreaking technology poised to transform how DL models are trained without compromising data security. By allowing local hospitals to share only trained parameters with a centralized DL model, Federated Learning fosters collaboration while preserving privacy. However, hurdles persist, including heterogeneity, domain shift, data scarcity, and multi-modal complexities inherent in medical imaging. In this illuminating talk, we delve into the clinical workflow and confront the common challenges facing AI in Medicine. Our focus then shifts to Federated Learning, exploring its promise, pitfalls, and potential solutions. Drawing from recent breakthroughs, including a compelling MR Brain imaging case study published in Nature Machine Intelligence, we navigate the landscape of secure and efficient AI adoption in healthcare.


Bio: Shadi Albarqouni, a pioneering figure in Computational Medical Imaging, serves as a Professor at the University of Bonn and an AI Young Investigator Group Leader at Helmholtz AI. With significant roles at Imperial College London, ETH Zurich, and the Technical University of Munich (TUM), Shadi's impact reverberates through his 100+ publications in esteemed journals and conferences. His expertise extends beyond academia, with contributions as an Associate Editor at IEEE Transactions on Medical Imaging and evaluator for national and international grants like DFG, BMBF, and EC. Recognized with awards like the DAAD PRIME Fellowship, Shadi fosters collaboration through AGYA and ELLIS memberships and initiatives like the Palestine Young Academy and the RISE-MICCAI community, focusing on innovative medical solutions and knowledge transfer to emerging countries. Explore more about his work at https://albarqouni.github.io/.

Zoom Video

Join our Cloud HD Video Meeting

Zoom is the leader in modern enterprise video communications, with an easy, reliable cloud platform for video and audio conferencing, chat, and webinars across mobile, desktop, and room systems. Zoom Rooms is the original software-based conference room solution…

www.group-telegram.com/ms/irandeeplearning.com/508

7.54K viewsMohammad Sabokrou, Feb 8, 2024 at 19:06

If you're interested in federated learning, particularly in medical imaging, we invite you to join our seminar tomorrow (Friday) at 11:00 a.m. Iran time! Zoom: https://oist.zoom.us/j/95908496615?pwd=akxZNmprLzNXY212TFh0ZWQ1ZlNyUT09
Meeting ID: 959 0849 6615
Passcode: 767685

Speaker: Prof. Shadi Albarqouni, Computational Medical Imaging Research, University of Bonn

Title: Unlocking the Potential of Federated Learning in Medical Imaging


Abstract: Deep Learning (DL) stands at the forefront of artificial intelligence, revolutionizing computer science with its prowess in various tasks, especially in computer vision and medical applications. Yet, its success hinges on vast data resources, a challenge exacerbated in healthcare by privacy concerns. Enter Federated Learning, a groundbreaking technology poised to transform how DL models are trained without compromising data security. By allowing local hospitals to share only trained parameters with a centralized DL model, Federated Learning fosters collaboration while preserving privacy. However, hurdles persist, including heterogeneity, domain shift, data scarcity, and multi-modal complexities inherent in medical imaging. In this illuminating talk, we delve into the clinical workflow and confront the common challenges facing AI in Medicine. Our focus then shifts to Federated Learning, exploring its promise, pitfalls, and potential solutions. Drawing from recent breakthroughs, including a compelling MR Brain imaging case study published in Nature Machine Intelligence, we navigate the landscape of secure and efficient AI adoption in healthcare.


Bio: Shadi Albarqouni, a pioneering figure in Computational Medical Imaging, serves as a Professor at the University of Bonn and an AI Young Investigator Group Leader at Helmholtz AI. With significant roles at Imperial College London, ETH Zurich, and the Technical University of Munich (TUM), Shadi's impact reverberates through his 100+ publications in esteemed journals and conferences. His expertise extends beyond academia, with contributions as an Associate Editor at IEEE Transactions on Medical Imaging and evaluator for national and international grants like DFG, BMBF, and EC. Recognized with awards like the DAAD PRIME Fellowship, Shadi fosters collaboration through AGYA and ELLIS memberships and initiatives like the Palestine Young Academy and the RISE-MICCAI community, focusing on innovative medical solutions and knowledge transfer to emerging countries. Explore more about his work at https://albarqouni.github.io/.

BY Deep learning channel