Awards for The best student projects related to artificial intelligence | Gdańsk University of Technology

– Looking at your commitment, it is obvious that our students are passionate about carrying out innovative work, which is important for our university as a research university - said prof. Jacek Stefański, dean of the FETI during the awarding of prizes to authors of the best group projects.

Group projects are a subject that is supposed to teach students to co-operate while solving a specific problem. It has been implemented for 16 years as part of the curricula of all second-cycle studies at FETI.

– Groups of 3 to 5 students prepare solutions to real problems for two semesters - says PhD, Eng. Krzysztof Nowicki, GUT professor, coordinator of group projects at FETI. - Some of them are commissioned by external companies.

This year, four projects were awarded and six projects were distinguished. Three of the four awarded projects are related to artificial intelligence.

– Artificial intelligence is an area that is developing rapidly. All projects, and in particular those awarded, are very interesting, advanced and connected with real applications - said Dawid Gębka from IHS Markit.

The prizes were funded by the Dean of FETI (PLN 10 000 in total) and the company IHS Markit (PLN 5 000)

Projects awarded:

• Mobile application for imaging and classification of skin lesions - 1st place and IHS Markit award

The aim of the project was to develop, implement and validate a mobile application for the assessment of skin lesions, in particular discoloration, for the diagnosis of skin melanoma. The application is intended for doctors and medical personnel - it is to facilitate their work and shorten the time of diagnosis. As the authors of the project emphasize, in the longer term, the use of the application may contribute to an increase in the percentage of people participating in screening tests for the diagnosis of melanoma. It also allows for the creation of a larger database of photos of skin lesions that could constitute input data for further analyzes, e.g. using artificial intelligence methods (used in this project).

Team: Martyna Giziewicz (leader), Filip Brzeziński, Julia Guzik, Emilia Lewandowski.

Research supervisor: prof. Mariusz Kaczmarek

• Image recognition using simulators - 2nd place, and the IHS Markit award

The students' goal in the project was to create an image recognition algorithm, and more specifically semantic segmentation, assuming that there are no annotations available for the target environment that could be used to train the neural network. Such algorithms are used in autonomous vehicles, robotics and medical imaging.

Team: Bartłomiej Wróblewski (leader), Radosław Słowiński, Karol Woźniak.

Research supervisor: Sebastian Cygert

• Mobile robot for critical infrastructure inspection - 3rd place

Students created an integrated inspection robot platform and adapted it to the implementation of autonomous inspection missions of critical infrastructure (e.g. port areas, airports, refineries). The authors of the project established contact with representatives of the Port of Gdynia, thanks to which they could learn about the required functionalities and challenges faced by universal inspection robots, and then designed a very complex advanced software as a whole.

Team: Michał Baranowski (leader), Kamil Romanowski, Mikołaj Barcikowski, Łukasz Pawlicki, Jan Olencki

Research supervisor: prof. Łukasz Kulas

• Endoleoma segmentation in CT imaging - 3rd place

The aim of the project was to develop a system to support medical workers in the diagnosis of endoleaks in the abdominal aorta (creating a service of automatic aortic segmentation and aneurysm detection). As regards the provision and labeling of data, the project was carried out jointly with the Department of Cardiac and Vascular Surgery of the Medical University of Gdańsk.

Team: Artur Skrzynecki (leader), Paweł Kozielec, Marek Perdeusz

Research supervisor: PhD Tomasz Dziubich

Distinguished projects:

• NlightVR ™ - the use of virtual reality in conjunction with biosensors

Team: Anita Dworzanska (leader), Wiktor Pieńkowski, Tomasz Idczak, Kamil Grabowski, Ivan Vakulko

Research supervisor: prof. Jacek Lebiedź

Client: NlightniN Production Ltd.

• A mobile application for planning and performing tasks in a master - slave relationship

Team: Stanisław Góra (leader), Mikołaj Mirko, Mateusz Janicki, Jan Czubiak

Research supervisor: prof. Agnieszka Landowska

• Configurable guitar multi-effects on the Raspberry PI platform

Team: Jakub Chmielewski (leader), Bartosz Borczuch, Przemysław Pająk, Piotr Ziętkowski

Research supervisor: MSc Dawid Weber

• Mobile voting system for meetings of the Students' Parliament of Gdańsk University of Technology

Team: Maciej Żurek (leader), Szymon Nowacki, Paulina Cebulska, Adam Brejtfus, Mateusz Wszeborowski

Research supervisor: PhD Jakub Miler

Client: Student Government of Gdańsk University of Technology

• Research on the possibility of using neural networks based on physical laws (PGNN) in the analysis of spatial data

Team: Bartłomiej Borzyszkowski (leader), Marcin Świniarski, Karol Damaszke, Jakub Romankiewicz

Research supervisor: prof. Marek Moszyński

• Robot bartender

Team: Jakub Wysocki (leader), Krzysztof Włódarczak, Paweł Wróblewski, Damian Wypych, Łukasz Antoniszyn

Research supervisor: MSc Tomasz Chludziński,

Inauguration of second-cycle studies and of the subject - research project

Presentation of projects and awarding the prizes to the authors of the best group projects at FETI took place as part of the inauguration (online) of the 2nd cycle studies, which was commenced by over 350 students.

In the following years, group projects will be implemented at an earlier stage of studies - first-cycle studies. On the other hand, at the second cycle of studies, a new subject was launched - a research project.

– This is a two-semester subject in which students will have a research hypothesis to verify - said PhD, Eng. Paweł Czarnul, GUT professor, vice-dean for cooperation and promotion of FETI.

A team of students supervised by a GUT employee will verify the hypothesis submitted by the client. The task of students will consist, for example, in the analysis of a given problem, optimization of algorithms, or determination of the implementation possibilities of an application project.

PhD, Eng. Krzysztof Nowicki, GUT professor, the dean's representative for research projects, emphasizes that the obligatory result of a research project for projects with a research hypothesis is a report in the form of a publication formatted according to the IEEE template, prepared in English. For application projects, a report in the form of a patent application is also allowed.