 |
||
|
||
Home > TERATEC FORUM > Workshops > Workshop 6
Wednesday October 14, 2020 - Workshops
Workshop 06 - 11:00 to 12:30
AI in scientific computing : accelerating innovation in industrial and academic domain
Chared by Eric Petit, Research engineer, HPC application specialist, Intel and Michaël Krajecki, Agence de l’Innovation de Défense (AID)
How Deep Learning became the tool of choice for human level AI in radiology
 |
Deep Learning as a paradigm and wide family of machine learning models has passed through a series of evolutions to finally become an industrial tool of choice for several important emerging fields.
In this presentation we will try first to get a feeling of what Deep Learning is and what it isn't, discuss which new business opportunities it unlocks and then concentrate on the example of Therapixel, french start-up creating top-notch software for breast cancer detection.
The company was created as a spin-off of research institution INRIA by a team of scientists and software developers. From the very beginning it was oriented towards medical data and thanks to our first touchless visualization software Fluid we have a wide and growing network of partner hospitals today. Since 2016 we progressively developed our AI research team and finally decided to concentrate our efforts on creating an industrial product for breast cancer screening MammoScreen. It originates from one compact research prototype of a deep neural network which won in 2017 a big Digital Mammography data challenge organized by common efforts of Amazon, IBM and several US research institutions. In the next 3 years it quickly evolved from a research draft to an industrial product connected with our own secure medical data cloud easy to deploy in any hospital on the planet. We were missing one last key: an approval for market usage by respective regulatory commissions and we're almost getting there.
 |
 |
Biography : Yaroslav Nikulin is a Russian-French researcher specialized in application of Deep Learning based models to medical imaging problems. Before joining Therapixel in 2016 he spent 8 rich years in two universities in Russia and France and several research and development internships notably at Siemens Corporate Research and Philips Healthcare. During these internships he could acquire some skills and competences which permitted him to design and implement the winning solution of Digital Mammography DREAM Challenge, a major data competition where Therapixel took the 1st place in 2017. Since then he works with a team of engineers and AI researchers on the creation of an industrial product for breast cancer screening MammoScreen. In 2019 he participated in a clinical study which demonstrated a performance increase for radiologists working together with MammoScreen and its autonomous performance of the same level as professional radiologists specialized in mammography reading. |
|---|
Register now and get your badge here
- TERATEC Forum is strictly reserved for professionals.
- Participation to exhibition, conferences and workshops is free (subject to seats available)
- On line registration is obligatory.