1 code implementation • 21 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
In order to use the loss function in practice, we propose an algorithm that progressively improves the clinical acceptability of the model, so that we can achieve the best tradeoff possible between accuracy and given clinical criteria.
no code implementations • 8 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
First, using the history of glucose predictions made by several models, the future glucose variation at a given horizon is predicted.
no code implementations • 8 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
Research in diabetes, especially when it comes to building data-driven models to forecast future glucose values, is hindered by the sensitive nature of the data.
no code implementations • 8 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
Then, we show the importance of making stable predictions by smoothing the predictions made by the models, resulting in an overall improvement of the clinical acceptability of the models at the cost in a slight loss in prediction accuracy.
no code implementations • 8 Sep 2020 • Maxime De Bois, Hamdi Amroun, Mehdi Ammi
This paper presents a 3-step system that estimates the real-time energy expenditure of an individual in a non-intrusive way.
no code implementations • 8 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
We show the usefulness of its interpretable nature by analyzing the contribution of each variable to the final prediction.
no code implementations • 8 Sep 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
As informative for the patients as for the practitioners, it can enhance the understanding of the predictions made by the model and improve the design of future glucose predictive models.
no code implementations • 29 Jun 2020 • Maxime De Bois, Mehdi Ammi, Mounîm A. El Yacoubi
In a field where data are hard to obtain, and where the comparison of results from different studies is often irrelevant, GLYFE gives the opportunity of gathering researchers around a standardized common environment.
no code implementations • 29 Jun 2020 • Maxime De Bois, Mounîm A. El Yacoubi, Mehdi Ammi
To improve the quality of the transfer between multiple sources of data, we propose a multi-source adversarial transfer learning framework that enables the learning of a feature representation that is similar across the sources, and thus more general and more easily transferable.