Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu

INESC TEC researchers internationally acknowledged thanks to work in the health area

Three teams composed of INESC TEC’s researchers were acknowledged at several international competitions in the field of telecommunications and multimedia with direct impact health. The awards were presented by three different entities: International Telecommunication Union (ITU), with the INESC TEC team winning first place; MICCAI Hackathon 2021, where INESC TEC researchers reached the top place, and Fraunhofer Portugal Challenge, with the team reaching third place.

24th November 2021

The team comprising Daniel Granhão, Guilherme Carvalho, Tiago Gonçalves, researchers at INESC TEC’s Centre for Telecommunications and Multimedia (CTM), and José Nuno Grácio Rosa, from the Polytechnic Institute of Leiria, entitled BacalhauNET, won the “ITU-ML5G-PS competition -007: ‘Lightning-Fast Modulation Classification with Hardware-Efficient Neural Networks’, a competition dedicated to Artificial Intelligence, which aims to develop neural networks with awareness of both the cost of computing inferences and accuracy.

More specifically, their work explored the development of neural networks to classify radiofrequency modulations. The results achieved led to the reduction of the network inference cost as much as possible, allowing for the creation of an innovative and compact network capable of addressing the proposed challenge.

INESC TEC researchers Helena Montenegro, Isabel Rio-Torto and Tiago Gonçalves won first place at MICCAI Hackathon 2021, an event co-organised by Wilson Silva (also an INESC TEC researcher) part of MICCAI 2021, the most prestigious conference in the area of Medical Image Analysis.

The winning work, “Show me consistency! Increasing the consistency and quality of annotations”, explored the fact that “annotations in medical imaging are extremely difficult to get, and also very challenging to keep consistent across multiple annotators. How could one increase the consistency and quality of annotations?”. This process led to the development of three machine learning models to measure the consistency of annotations made by different annotators.

Helena Montenegro, also a student at the Faculty of Engineering of the University of Porto (FEUP), reached the third place in the 12th edition of the Fraunhofer Portugal Challenge, in the master’s category. With the work “Privacy-Preserving visual case-based explanations”, the researcher aims to protect the privacy of patients in medical images, thus enabling their use and sharing as explanations to validate the decisions of Artificial Intelligence algorithms.

The existence of this type of explanation in clinical contexts increases the interpretability of deep learning systems, increasing their acceptance in real contexts. It’s worth mentioning that this solution presents a generative model, capable of anonymising images, removing discriminatory characteristics of the patients’ identity, while maintaining the medical characteristics that are essential for diagnosis.

Helena Montenegro’s work had already received the CTM 2021 “Best Master’s Thesis” award, an initiative that aims to highlight the high merit work developed at INESC TEC’s CTM.

The INESC TEC researchers mentioned in this news piece are associated with INESC TEC, FCT and UP-FEUP.