Eleni Chiou

pic

Machine Learning Scientist

GSK.ai

London

email: elechiou at gmail.com

I am a Machine Learning Scientist with over 6 years of academic and industrial experience in deep learning and AI for healthcare.

Currently, I work at GSK.ai, where I develop cutting-edge computer vision solutions for computational pathology, aiming to advance precision medicine and accelerate drug development.

I completed my PhD in Machine Learning and Medical Imaging at University College London (UCL), where I was advised by Iasonas Kokkinos and Eleftheria Panagiotaki. During my PhD, I worked at the intersection of Deep Learning, Computer Vision, and Medical Imaging. My research primarily focused on the development of domain adaptation methods for semantic segmentation with deep learning, applied to both novel medical imaging datasets and natural images. This included proposing pixel-level domain adaptation approaches that leverage generative adversarial networks and content-style disentanglement to translate images across different domains while maintaining semantic consistency, allowing for unsupervised or semi-supervised model adaptation.

I received my Diploma in Electrical & Computer Engineering from the Technical University of Crete (TUC) and my MSc degree in Biomedical Engineering from the Technical University of Denmark (DTU).

During my PhD, I did a research internship with Samsung, where I worked on efficient reference-based image super-resolution for shadow removal.

Publications

Unsupervised Domain Adaptation with Semantic Consistency across Heterogeneous Modalities for MRI Prostate Lesion Segmentation
E. Chiou, V. Valindria, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
MICCAI Workshop on Domain Adaptation and Representation Transfer (DART), 2021
arxiv
Synthesizing VERDICT Maps from Standard DWI Data using GANs
E. Chiou, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
MICCAI Workshop on Computational Diffusion MRI (CDMRI), 2021
arxiv
Synthetic Q-Space Learning with Deep Regression Networks for Prostate Cancer Characterization with R-VERDICT
V. Valindria, M. Palombo, E. Chiou, S. Singh, S. Punwani and E. Panagiotaki
IEEE International Symposium on Biomedical Imaging (ISBI), 2021
Harnessing Uncertainty in Domain Adaptation for MRI Prostate Lesion Segmentation
E. Chiou, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2020
project page / arxiv / code
Domain Adaptation for Prostate Lesion Segmentation on VERDICT-MRI
E. Chiou, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
International Society for Magnetic Resonance in Medicine (ISMRM), 2020
Automatic Classification of Benign and Malignant Prostate Lesions: A Comparison Using VERDICT-MRI and ADC Maps
E. Chiou, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
International Society for Magnetic Resonance in Medicine (ISMRM), 2019
Prostate Cancer Classification on VERDICT DW-MRI Using Convolutional Neural Networks
E. Chiou, F. Giganti, S. Punwani, I. Kokkinos and E. Panagiotaki
MICCAI Workshop on Machine Learning in Medical Imaging (MLMI), 2018
Spatial Filter Feature Extraction Methods for P300 BCI Speller: A Comparison
E. Chiou and S. Puthusserypady
IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2016 (Oral)

Thesis