November 22nd, 2022 
Daniela Lopez de Luise
We hope you can join us on Thursday December 1 at 12:30pm CET for our next 1-hour free webinar.
Date&Time: December 1st, 2022 at 12:30 p.m. CET [06:30 a.m. New-York] – [12:30 p.m. Paris/Louvain] – [6:30 p.m. Beijing]
Title: Learning coalitions: How an AI and a group of experts with varying opinions can continuously improve their collective decisions made from images.
Title: Learning coalitions: How an AI and a group of experts with varying opinions can continuously improve their collective decisions made from images.
Speaker: Benoit Macq (UCLouvain)
To join the webinar, please register to receive more details on how to connect.
The registration form can be found at: https://forms.gle/9JCc6NBgM1x2kZK6A
or via the website of the journal at: https://jivp-eurasipjournals.springeropen.com/
Contact: Esinu Abadjivor <esinu.abadjivor@springernature.com>
Title: Learning coalitions: How an AI and a group of experts with varying opinions can continuously improve their collective decisions made from images.
Abstract: Deep learning technologies have enabled the development of artificial intelligence systems for the recognition of anomalies in images or for predicting possible effects on the imaged object. For the application to medicine, we propose a new approach to learning and continuous improvement of models, based on a coalition of heterogeneous hospitals sharing the same care methods and continuously improving decisions in a common way to achieve the same quality of clinical decisions for all. The models developed and the humans constitute a learning coalition and the progresses in the quality of the decisions made by the models and by the human experts are quantified in the respect of ethical rules adopted by the coalition.
Contact: Esinu Abadjivor <esinu.abadjivor@springernature.com>
Title: Learning coalitions: How an AI and a group of experts with varying opinions can continuously improve their collective decisions made from images.
Abstract: Deep learning technologies have enabled the development of artificial intelligence systems for the recognition of anomalies in images or for predicting possible effects on the imaged object. For the application to medicine, we propose a new approach to learning and continuous improvement of models, based on a coalition of heterogeneous hospitals sharing the same care methods and continuously improving decisions in a common way to achieve the same quality of clinical decisions for all. The models developed and the humans constitute a learning coalition and the progresses in the quality of the decisions made by the models and by the human experts are quantified in the respect of ethical rules adopted by the coalition.
Speaker Bio: Benoit Macq is a professor at the Ecole Polytechnique de Louvain, UCLouvain, Belgium where he heads the “pixels and interactions lab (www.pilab.be)” and teaches image processing, information theory and medical imaging. He has been a visiting professor at McGill University in Montreal and at Telecom Paris Tech. He was a senior invited scientist at MIT and at the Ecole Polytechnique Fédérale de Lausanne. Benoit Macq is a Fellow Member of the IEEE and a Member of the Royal Academy of Sciences of Belgium. He is Founding Editor of the Journal on Multimodal Interfaces and Senior Associate Editor of the IEEE Transaction on Image Processing. Benoit Macq has been Prorector of UCLouvain and has promoted the creation of 12 spin-off companies based on the research done in his laboratory. He is co-creator of the TRAIL Institute (Trusted AI Labs) which gathers 600 AI researchers in French-speaking Belgium.
Webinar videos are available online at https://vimeo.com/showcase/8005816.
image004.jpg@01D6FBAE.3C1BBB60” width=”435″ height=”227″ border=”0″>
Posted in