NOTE! This site uses cookies and similar technologies.

If you not change browser settings, you agree to it. Learn more

I understand

Learn more about cookies at : http://www.aboutcookies.org/Default.aspx?page=1

October 14, 2019 | Complex computations on encrypted data now possible

CEA List calculs complexes donnees chiffrees 250In e-healthcare, keeping personal data confidential is crucial. But when you store and analyze data on servers that are connected to the internet, that data becomes vulnerable to hackers, especially during transmission. A brand new homomorphic encryption system has made complex computations on encrypted data possible.

Cingulata was developed by List, a CEA Tech institute, to create applications capable of performing computations on encrypted data. This single software compilation chain previously supported only the Brakerski/Fan-Vercauteren (BFV) homomorphic encryption scheme, the most commonly used within the community. It now has a new, more powerful cryptosystem library, TFHE (Fast Fully Homomorphic Encryption over the Torus), designed by List (that earned the Carnot seal in 2006) and academic research labs.

TFHE's fast bootstrapping reduces the time it takes to complete a multiplication operation so that time is constant—it does not depend on the number of operations that precede it. This means that for the same application, using TFHE in Cingulata can speed up computation times by a factor of at least ten compared to BFV. Ultimately, Cingulata will integrate other cryptosystems and serve as a single interface.

The new version of Cingulata has already been used in the EU Horizon 2020 project KONFIDO, for example, where it is helping to develop solutions to allow doctors located in different European countries to read patients' medical files online without putting confidentiality at risk. And companies like Orange and Thales are looking at how Cingulata can help them respond to the needs of their respective markets.

An open source version of Cingulata with TFHE is available on GitHub: https://github.com/CEA-LIST/Cingulata

Read article at http://www.cea-tech.fr/

 

 

L’institut Carnot TN@UPSaclay (CEA List) aux rendez-vous Carnot 2019

RdvCARNOT 710

carnot 2019 1Au croisement de la recherche et du développement industriel, l’institut Carnot TN@UPSaclay (CEA List) focalise ses programmes de R&D sur l’intelligence artificielle, l'usine du futur, les systèmes cyberphysiques et la santé numérique. Il contribue à la compétitivité des entreprises par l’innovation et le transfert technologique et collabore chaque année avec plus de 200 entreprises, dont 50% de PME.

Nos 3 atouts ? Un important portefeuille de technologies génériques protégées par des brevets, une approche d’« ensemblier de l’innovation » et des plateformes technologiques de haut niveau.

Retrouvez-nous sur le stand H33 (Espace Champerret) et sur les conférences suivantes :

carnot 2019 2Conférences plénières

16 octobre, 11:00-11:40 | Quelles opportunités pour l’IA dans le secteur manufacturier ?

Pour le Carnot TN@UPSaclay :

  • Eric Bévillard, Bureau Etude Marketing, CEA Tech
  • Florent Pénet, Consultant en marketing et stratégie de l’innovation technologique, CEA Tech

Pitchs Espace Manufacturing (J32)

  • 16 octobre, 09:00 | « Isybot, l’automatisation agile » | Yvan Measson, Isybot
  • 16 octobre, 14:00 | « La perception automatique, fonction primaire des machines intelligentes » | Stéphane David, Carnot TN@UPSaclay (CEA List)

Pitchs Village Findmed

  • 16 octobre, 10:30 | « Collecte et analyse des données avec l’IA : comment aider les professionnels de la santé ? » | Romain Farel, Carnot TN@UPSaclay (CEA List)
  • 17 octobre, 10:35 | « Nouveaux outils logiciels pour l’optimisation des pratiques en imagerie médicale » | Guillaume Boissonnat, Carnot TN@UPSaclay (CEA List)

Pour en savoir plus venez nous rencontrer sur le stand H33 à l’espace Champerret lors des RDV Carnot les 16 et 17 octobre 2019 !

Pas encore inscrit ? https://rdv-carnot.vimeet.events/fr/

À très vite !

 

 

7 octobre 2019 | Le List développe l’exosquelette EMY pour redonner la mobilité à des patients tétraplégiques

clinatec 1 2
Pilotage de l’exosquelette 4 membres par le patient dans le cadre du projet BCI mené à Clinatec ©Clinatec

L’événement

Pour la première fois, un patient tétraplégique a pu se déplacer et contrôler ses deux membres supérieurs grâce à une neuroprothèse, qui recueille, transmet et décode en temps réel les signaux cérébraux pour contrôler un exosquelette réalisé par le CEA List, institut Carnot. Les résultats de l’étude clinique du projet Brain Computer Interface (BCI), réalisée à Clinatec (CEA, CHU Grenoble Alpes) ont été publiés le 4 octobre 2019 dans la revue The Lancet Neurology. Suite à l’opération de presse organisée par le CEA et Clinatec le 7 octobre, de nombreux articles et reportages ont souligné cette avancée pour les patients.

Le défi du List

Pur l’équipe de robotique ExoBCI installée à Grenoble, au plus près des patients, l’exosquelette EMY (Enhanced Mobility) est le fruit de plusieurs années de développement et d’intégration des briques d’actionnement réversible et de contrôle-commande du List. La conception d’EMY a spécifiquement pris en compte l’interaction d’une personne tétraplégique avec l’exosquelette pour pouvoir la mobiliser en toute sécurité. Les outils de conception logicielle du List ont également contribué à ce succès en assurant la sécurité des cartes électroniques utilisées pour le pilotage de l’exosquelette.

Trouvez les supports de communication et les reprises média de l’événement :

 

 

October 3, 2019 | Real-time 3D image reconstruction for non-destructive testing

CND reconstruire images 3d 250A new image reconstruction algorithm that is 30 times faster than the prior state of the art has placed real-time inspection within reach. The algorithm, based on techniques used in medical imaging, was utilized for non-destructive testing (NDT), where it produced 3D images in real time.

In non-destructive testing (NDT) 3D images are usually constructed by post-processing data from ultrasonic probes. And, with processing times from a few minutes up to an hour (depending on the number of voxels in the final 3D image), "real-time" imaging is simply not possible. Researchers at List, a CEA Tech institute, sped image processing up by a factor of 30, effectively eliminating a major obstacle to real-time 3D imaging.

The research, which was for a PhD dissertation, leveraged existing technologies used in medical imaging. They tweaked a space-time Fourier transform algorithm to factor in the unique characteristics of non-destructive testing, such as a variety of waves, the presence of interfaces between the probe and part to inspect, and the reflection of waves off of the interfaces. An initial proof-of-concept confirmed that the algorithm was valid by performing the calculations in parallel on multicore graphics processors. The 3D images obtained presented the same level of quality as those produced using conventional methods—and they took just seconds to process.

The algorithm makes it possible to see the geometry of a defect "live" and in 3D from a single sensor position. Another PhD research project is focusing on speeding up data transfer to the CPUs with the goal of building a prototype real-time 3D imager.

Read article at http://www.cea-tech.fr/

 

 

September 23, 2019 | Better control of patient exposure to radiation during cardiac catheterization

cardiologie reduction dose 250Reference beams used to measure the doses delivered by X-ray imaging equipment used in interventional cardiology have been developed by France's national radiation testing and metrology lab, LNHB.

In interventional cardiology, a doctor inserts a catheter into an artery or vein and directs it toward the area of the heart that requires treatment. During these procedures, the doctor controls his or her movements using real-time X-ray images. Catheterization has become increasingly popular over the past several years. However, until now, there was no available system for accurately measuring the dose delivered to patients undergoing these procedures. LNHB (Laboratoire National Henri Becquerel), which is both France's national radiation testing and metrology lab and a laboratory of CEA Tech institute List, recently filled this gap.

As part of the EU VERIDIC project, LNHB researchers developed four new reference beams covering the energy ranges used in interventional cardiology. The researchers drew on their spectrum characterization know-how, using X-ray generators to produce beams comparable to those generated by the equipment found in hospitals and treatment centers. They then measured the dosage (air kerma) with the primary instruments developed by LNHB.

These new reference beams will be used to calibrate the measurement instruments utilized in clinical settings, providing an accurate assessment of the doses of radiation delivered to patients during cardiac catheterization procedures.

Read article at http://www.cea-tech.fr/