Three Oral Presentations at ISBI in London
Mert Ikinci
Master Thesis Student
Mert is currently pursuing an M.Sc. in Informatics at the Technical University of Munich, specializing in Computer Vision and Machine Learning. He is currently working on his master’s thesis, Development of a Machine Learning Algorithm for Histopathological Classification of Conjunctival Melanocytic Intraepithelial Lesions. Also he is working as a Computer Vision and Deep Learning Engineer at Signatrix in Berlin, focusing on real-time segmentation applied research in retail. Previously, he gained experience as a Computer Vision Engineer at the Turkish Defence Industry, working on applied research for remote sensing and object tracking with UAV’s. He holds a Bachelor’s degree in Computer Science from TOBB University of Economics and Technology, Ankara.
Aya ELGebaly
Master’s Thesis Student
Aya Elgebaly is a Master’s Thesis Student at Albarqouni Lab, where she works on multi-rater medical image segmentation using probabilistic modeling and deep learning. Her research addresses challenges in clinical imaging—such as annotation ambiguities and observer-related uncertainties—with the aim of developing robust and reliable segmentation models for practical healthcare applications. Aya is currently pursuing an Erasmus Mundus Joint Master’s in Medical Imaging and Applications (MAIA), an interdisciplinary program offered in collaboration with the University of Girona (Spain), the University of Bourgogne (France), and the University of Cassino (Italy). With a solid background in medical image analysis, she has contributed to several projects, including semi-supervised learning for breast lesion detection, multi-modality stroke lesion segmentation, surgical video anonymization for hospital integration, and skin image classification using knowledge distillation. Her work on chest CT image registration for lung imaging further demonstrates her diverse expertise in medical AI. Her research interests encompass medical image segmentation, and uncertainty modeling, with a focus on developing AI-driven tools that support diagnostic accuracy and improve patient outcomes. Through her efforts at Albarqouni Lab, Aya is committed to bridging the gap between research and clinical practice.
Malek Al Abed
Research Associate
Malek is a Research Associate at the University Hospital Bonn under the supervision of Prof. Dr. Shadi Albarqouni. He holds an M.Sc. in Biomedical Computing from TU Munich and completed his masters thesis at the University Hospital Bonn and TU Munich under the same supervision. His master’s thesis focused on image quality transfer for ultra-low-field MRI, with particular emphasis on the Hyperfine SWOOP system. Previously, he completed his undergraduate studies in Electrical Engineering at Kuwait University and undertook a research internship at Sony Stuttgart as a deep learning researcher, where he worked on audio recognition and human avatar technologies. Currently, his research focuses on the use of medical imaging in radiation therapy (iMRT and VMAT), particularly for dose optimization and personalized treatment planning. In addition, he works on developing an AI service and its corresponding API as part of the OMI protocol.
Elodie Germani
Postdoctoral Researcher
Elodie Germani works with Prof. Shadi Albarqouni as a postdoctoral researcher. She did her PhD at the University of Rennes, under the supervision of Dr. Camille Maumet and Prof. Elisa Fromont. After four years of medicine school at the University of Versailles, she took a shift in her career and started a Master’s degree in bioinformatics. Her research focuses on exploring, modelling and building solutions to take into account the variability of data in medical imaging, particularly using deep representation learning. During her PhD, her goal was to facilitate the re-use of data shared on public databases by taking into account the different sources of variability. In the future, she would like to focus more in the use of real-world data and on the robustness of machine learning models to dataset shifts and privacy attacks.
