Hugo Yèche,

"Time you enjoy wasting is not wasted time" - Marthe Troly-Curtin

PhD Student

ETH Zürich
Department of Computer Science
Biomedical Informatics Group
Universitätsstrasse 6
8092 Zürich
CAB F53.1

I am interested in solving problems arising from deep learning approaches in health care. With this purpose I focus on interpretable approaches that exploit unlabeled data and tackle the issue of class imbalance.

After two years of preparatory classes I entered Telecom Paris as my engineering school for the following three years.  During my first year, I obtained my B.Sc in the field of “Science of Engineering”.  I then moved to Sophia-Antipolis for my second year to join EURECOM as a double degree with Telecom Paris. Finally I completed my last year of master at ENS Paris Saclay, joining the “Mathematiques, Vision, Apprentissage”  (MVA) master’s program where I studied applied mathematics and computer vision in machine learning. During my last 6 months of master, I completed a research internship within Imagia, a Montreal-based start-up. There I worked on interpretable deep learning in the context of medical Imaging.  I joined the Biomedical Informatics group in 2020 for my Ph.D.

Abstract The recent success of machine learning methods applied to time series collected from Intensive Care Units (ICU) exposes the lack of standardized machine learning benchmarks for developing and comparing such methods. While raw datasets, such as MIMIC-IV or eICU, can be freely accessed on Physionet, the choice of tasks and pre-processing is often chosen ad-hoc for each publication, limiting comparability across publications. In this work, we aim to improve this situation by providing a benchmark covering a large spectrum of ICU-related tasks. Using the HiRID dataset, we define multiple clinically relevant tasks in collaboration with clinicians. In addition, we provide a reproducible end-to-end pipeline to construct both data and labels. Finally, we provide an in-depth analysis of current state-of-the-art sequence modeling methods, highlighting some limitations of deep learning approaches for this type of data. With this benchmark, we hope to give the research community the possibility of a fair comparison of their work.

Authors Hugo Yèche, Rita Kuznetsova, Marc Zimmermann, Matthias Hüser, Xinrui Lyu, Martin Faltys, Gunnar Rätsch

Submitted NeurIPS 2021 (Datasets and Benchmarks)


Abstract Intensive care units (ICU) are increasingly looking towards machine learning for methods to provide online monitoring of critically ill patients. In machine learning, online monitoring is often formulated as a supervised learning problem. Recently, contrastive learning approaches have demonstrated promising improvements over competitive supervised benchmarks. These methods rely on well-understood data augmentation techniques developed for image data which do not apply to online monitoring. In this work, we overcome this limitation by supplementing time-series data augmentation techniques with a novel contrastive learning objective which we call neighborhood contrastive learning (NCL). Our objective explicitly groups together contiguous time segments from each patient while maintaining state-specific information. Our experiments demonstrate a marked improvement over existing work applying contrastive methods to medical time-series.

Authors Hugo Yèche, Gideon Dresdner, Francesco Locatello, Matthias Hüser, Gunnar Rätsch

Submitted ICML 2021