Djalel wants me to relay this awesome postdoc opportunity, I'll oblige !
Applications are invited for a postdoctoral fellow of two years, at the echOpen non-profit organization.Keywords: deep learning, ultra-sound imaging, image segmentation, model compressionDeadline:The objective of the post-doc is to leverage Deep Learning methods for Ultra-Sound imaging (echography). The work will be directly implemented in the echOpen device prototype, potentially resulting in a tremendous impact on medical diagnosis (more information below).Several subjects are pertinent to the position, they include- image segmentation and organ detection- anomaly detection- model compression- image denoising and enhancementPractical information:- Location: Hôtel-Dieu, at the heart of Paris, within the multi-disciplinary team of echOpen.- Duration: 2 years- Start: March-April 2017- Salary: competitive with the industry- Deadline for application:Medical dataThe project benefits from a strong support from the AP-HP (the organization grouping Paris hospitals) which translates into a privileged access to a large pool of medical data and images, along with their labels.Echopen project descriptionEchopen is an ambitious open research project with a high potential impact on society. It aims at producing the first low-cost portable (open source) echograph, that consists of a lightweight sensor plugged into a smartphone, bringing to the populations a universal diagnosis health tool. This probe is the stethoscope of our modern times.Aimed at health professionals, this initiative was developed to address the tragedy that 2/3 of the world population do not have access to medical imagery tool. We think especially of emergency healthcare, general medicine in both urban and low infrastructure regions, pregnant women and childbirth healthcare.The project benefits from a strong academic support (including AP-HP, ENS Cachan, Satie, Telecom ParisTech, Insititut Pasteur, Brussels Uninversité Libre, and many others research centers and universities). Furthermore, a functional prototype is already in use and allows to directly assess the different algorithms.QualificationsThe minimum requirements are:- a PhD in computer science, statistics, machine learning, signal processing, or computer vision- strong programming skills (especially in Python and its scientific stack)- strong communication abilitiesWhat could strengthen the application:- experience in training large neural networks- a strong enthusiasm in medical applications- an eagerness to contribute to the open-source community- French langage is optionalContactIf you are interested, please send an email to:- Djalel Benbouzid (email@example.com)- Mehdi Benchoufi (firstname.lastname@example.org)Including:- your CV,- a list of your publications- your github username if you have one- one paragraph (~4-6 lines) describing your motivation and achievements.
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.