Education
BSc in Biology, University of Crete, 2016
MSc in Bioinformatics, University of Crete, 2018
Titles and Appointments
Research Analyst
Address
HITS Building, Rm 3121
410 W 10th Street
Indianapolis, IN 46202
Research Keywords
Federated Learning, Ethical AI, Biomedical AI
Links
Google Scholar
LinkedIn
GitHub
Research Summary/Bio
Descended from Greece, worked in labs all over Europe, Akis is a research analyst at the Division of Computational Pathology, focusing on privacy preserving applications to enable large-scale multi-center studies on biomedical AI. Before joining IU, he had amassed four years of research experience in computer vision in various labs and institutions across Europe (including Universitat Politècnica de Catalunya, Dublin City University, Universität Tübingen) and medical imaging (University of Barcelona). Throughout this time, the models he developed on image and video saliency prediction achieved state of the art performance, with corresponding papers published in ICCV and BMVC. Afterwards, his interests shifted to his current focus: ethical AI and federated learning for health care, striving to bring together large-scale multi-center studies for diverse and unbiased diagnostic models. His work on the medical sector of AI has been published by Scientific Reports, Physica Medica, and MICCAI workshop among others. By night he is a (published) writer of science fiction and fantasy, where he uses his AI expertise to imagine the future.
Featured publications (journals, conferences, abstracts)
- Linardos, Akis, Kaisar Kushibar, Sean Walsh, Polyxeni Gkontra, and Karim Lekadir. Federated learning for multi-center imaging diagnostics: a simulation study in cardiovascular disease. Scientific Reports, 12(1):1–12, 2022.
- Linardos, Akis, Matthias Kümmerer, Ori Press, and Matthias Bethge. Calibrated prediction in and out-of-domain for state-of-the-art saliency modeling. ICCV, 2021.
- Panagiotis Linardos, Eva Mohedano, Juan José Nieto, Noel E. O’Connor, Xavier Giró-i-Nieto, and Kevin McGuinness. Simple vs complex temporal recurrences for video saliency prediction. In 30th British Machine Vision Conference 2019, BMVC 2019, Cardiff, UK, September 9-12, 2019, page 182. BMVA Press