In 2020, the IEEE Portugal Section launched the IEEE Outstanding MSc Thesis Award aiming to award the best master’s dissertations carried out in Portuguese higher education institutions. 6 dissertations were chosen in different categories, among 56 applications. 4 winning dissertations were carried out at Técnico by Pedro Trindade, Pedro Moreira, Hemaxi Narotamo and Miguel Ferreira.
The Técnico alumnus Pedro Moreira (Electrical and Computer Engineering) was distinguished in the “Energy and Power” category, for his disseration titled “Analysis of the Transitory Regime and Experimental Validation”, supervised by professor Sónia Ferreira Pinto and professor José Alves da Silva.
This is not the first that the work of Técnico alumnus Pedro Trindade (Aerospace Engineering) has been recognised. His thesis had already been recognised by the Portuguese Robotics Society (SPR). “I am very pleased to receive this award, which I see as a revalidation of the quality of my dissertation. It is a privilege to see my work recognised, which is something that motivates me to continue with scientific research in this area”, stresses the alumnus.
The master’s dissertation titled “Distributed Formation Control of Double-Integrator Vehicles in the Presence of Unknown Constant Disturbances”, carried out by Pedro Trindade, won the award in the “Robotics & Control” category. Under the supervision of professors Pedro Batista and Rita Cunha, this work addresses the topic of vehicle formation control. “By dealing with this topic, I have identified a relevant problem, which consists of controlling these formations when vehicles are affected by external disturbances”, explains the alumnus. To solve this problem, Pedro Trindade proposed a new algorithm that, unlike the existing solutions, is able to correct the effect of these disturbances. “In order to show that the proposed algorithm can take the vehicles to the desired formation, I used an unusual technique, which allowed me to present innovative ideas”, he stresses.
Through several experiments, which aimed to complement the results obtained from the theoretical point of view, Pedro Trindade was able to successfully validate the proposed algorithm and demonstrate the benefits of the proposed theoretical solutions applied to aerial robotics. “Although the experiments were carried out with four multi-engines, the results are scalable to any number of vehicles”, highlights the alumnus.
A new method for the segmentation of cell nuclei in microscopy images
Hemaxi Narotamo won the award in the “Engineering in Medicine and Biology” category. “My thesis consisted of developing an approach based on deep learning for the segmentation of cell nuclei, as well as a technique based on machine learning for cell cycle staging from DAPI-stained nuclei. “Given the relevance of nuclei segmentation, which has numerous applications in the field of biomedicine, it is important to develop fast and robust methods for large-scale image processing”, clarifies the alumna.
Aware of this importance, Hemaxi Narotamo developed a new method, based on deep learning, for the segmentation of cell nuclei in microscopy images, supervised by professor Maria Margarida Silveira and professor João Sanches. “The results showed that the approach developed achieves a performance comparable to state-of-the-art models for segmenting cores”, says the alumna. However, the proposed method is much faster and computationally more efficient, which is important once it will be used in clinical practice”, says Hemaxi Narotamo.
According to the Técnico alumna, “the methods proposed in my thesis, which will be of great value in the diagnosis of hereditary cancer based on microscopic images”, might have positively contributed to jury’s decision. “In addition, the methods that have been developed are automatic, fast and computationally efficient,” she adds.
After finishing the Biomedical Engineering course, she continues to do research in microscopy image analysis using deep learning. “Logically, the next step is doing a PhD to continue developing my work in this area. I started with 2D images, but now I’m going to work with 3D data, which has a whole new dimension of challenges”, says Hemaxi Narotamo.
The resolution of a fundamental problem in network routing
The dissertation titled “Routing on Multiple Optimality Criteria: Theory and Protocols”, carried out by the Técnico alumnus Miguel Alves (Electrical and Computer Engineering), under the supervision of professor João Luís Sobrinho won the award in the “Communications” category.
According to Miguel Alves, “the main innovation of our research is a solution to the problem of routing on multiple optimality criteria”. Besides developing a new theory, this solution also includes, “the design of new routing protocols, which meet the set of dominant paths from each source to each destination and allow to send data packets by any path in this set”, highlights the alumnus.
Despite being “grateful to receive this distinction from IEEE Portugal”, Miguel Alves confesses “this is not a complete surprise to me”. “Last August, a scientific article resulting from my joint work with professor Sobrinho received one of the two Best Paper Awards at SIGCOMM 2020, the main conference in the field of computer networks”, he shares. According to the Técnico alumnus, this work “stands out for the simplicity and generality with which it solves a primary problem of network routing, thus enhancing better and more flexible routing choices for future network operators”. “Finally, I must point out that this recognition also rewards two decades of professor Sobrinho’s original research, to which I had the pleasure to associate myself and to give my scientific contribution”.