Transilvania University of Brasov

  • Natalia GARCIA-... profile
    Natalia GARCIA-...
    26 July 2019 - updated 1 year ago
Category winner
Voting is closed. 1378 users have voted.

About the innovator

The Transilvania University of Braşov is a Romanian public academic institution with over 800 full-time scientists and professors, and around 20,000 students (including over 500 students in PhD programs). The Department of Automation and Information Technology is affiliated to the Faculty of Electrical Engineering and Computer Sciences. It has participated in / coordinated numerous European and National research projects with focus on biomedical engineering and medical imaging applications, and has strong know-how in artificial intelligence / deep learning, encryption, physiological modeling, and computer vision.

MH-MD project on Twitter

What is the innovation

Despite their potential in enabling personalized medicine applications, the adoption of Deep Learning (DL) based solutions in clinical workflows has been hindered in many cases by strict regulations concerning the privacy of patient data.
We have implemented a software framework for developing personalized medicine DL solutions based on homomorphically encrypted data. Supervised learning techniques are employed, and the trained DL models are deployed in a cloud environment. In a typical scenario, the client (hospital, patient, etc.) encrypts the data, sends it to the cloud based application, which processes the data with the DL model and outputs encrypted results, which are then sent back to the client. Finally, the client decrypts and interprets the results. Thus, we ensure that both input data and results remain private, and data analysis is performed only on the encrypted version of the data.
Using the above described framework, we have successfully developed multiple personalized medicine applications, including: coronary artery disease assessment and diagnosis based on coronary angiographies (hospital is the client), and long-term cardiovascular monitoring, relying on a reduced-order hemodynamic model and sensor data acquired through a wearable device (patient is the client).

Out of the lab. Into the market

Different IP protection strategies have been considered to provide a value proposition based on physiological modeling, artificial intelligence and homomorphic encryption, to execute personalized medicine applications without disclosing sensitive data. A patent application has been already filed, and possible business strategies include the licensing of the toolset to hardware and software providers, or to provide the toolset as a service. We are also looking for strategic partnerships and investment to support our growth into the market (including the submission of project proposal to public national or EU calls).

Benefits of participation in Horizon 2020

As a research group in the Department of Automation and Information Technology, our participation in the framework programme has allowed us to focus our efforts in areas with the highest innovation potential. Through projects like MD-Paedigree and MyHealthMyData, we have developed a wider perspective on healthcare related applications by working with other organizations that cover the range from medical devices to medical applications (including physiological modeling, artificial intelligence / deep learning based image processing, high performance computing, etc.). As a department we have learned about the value of transferring the technology beyond the lab.

This innovation was funded via H2020 project MH-MD

Team behind the innovation