Matthieu Jean-Michel Philippe Broisin

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Bio-Hybrid Systems for Ecosystem Level Effects

Rafael Botner Barmak, Matthieu Jean-Michel Philippe Broisin, Jean-Louis Deneubourg, Robert Matthew Mills, Francesco Mondada, Martina Szopek

In a time marked by ecological decay and by the perspective of a severe backlash of this ecosystem decay and climate devastation onto human society, bold moves that employ novel technology to counteract this decline are required. We present a novel concept of employing Artificial Life technology, in the form of cybernetically enhanced bio-hybrid superorganisms as a countermeasure and as a contingency plan. We describe our general conceptual paradigm, consisting of three interacting action plans, namely: (1) Organismic Augmentation; (2) Bio- Hybrid Socialization and (3) Ecosystem Hacking, which together compose a method to create a novel agent for ecosystem stabilization. We demonstrate, through early results from the research project HIVEOPOLIS, a specific way how classic Artificial Life technologies can create such a living, ecologically active and technologically-augmented superorganism that operates outside in the field. These technologies range from cellular automata and biomimetic robots to novel and sustainable biocompatible materials. Aiming at having a real-world impact on the society that relies on our biosphere is an important aspect in Artificial Life research and is fundamental to our methodology to create a physically embodied and useful form of Artificial Life.

MIT Press2021

Evaluation of Deep Learning Strategies for Nucleus Segmentation in Fluorescence Images

Matthieu Jean-Michel Philippe Broisin, Jonathan Roth

Identifying nuclei is often a critical first step in analyzing microscopy images of cells and classical image processing algorithms are most commonly used for this task. Recent developments in deep learning can yield superior accuracy, but typical evaluation metrics for nucleus segmentation do not satisfactorily capture error modes that are relevant in cellular images. We present an evaluation framework to measure accuracy, types of errors, and computational efficiency; and use it to compare deep learning strategies and classical approaches. We publicly release a set of 23,165 manually annotated nuclei and source code to reproduce experiments and run the proposed evaluation methodology. Our evaluation framework shows that deep learning improves accuracy and can reduce the number of biologically relevant errors by half. (c) 2019 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

WILEY2019