Department of Biomedical Eng

Awards during HSI 2018!

This year, the 11th edition of the Human System Interaction conference was held in Gdańsk, Poland from the 4th to 6th July. The conference covered the theory, design and application of human‐system interaction in the areas of science, education, business, industry, services, humanity, environment, health, and government. Additionally, the parallel event, Summer School on Deep Learning was organized to make the HSI Conference even more attractive for students and young professionals. As an effect, almost 200 people from 20 countries attended both events, mostly young participants.

During the event 4 IEEE IES awards have been presented for students and young professionals. One of the awards was granted to Alicja Kwasniewska from our Department for the paper:

Mingshan Wang, Tejaswini Sirlapu, Alicja Kwasniewska, Maciej Szankin, Marko Bartscherer, Rey Nicolas, “Speaker Recognition using Convolutional Neural Network with Minimal Training Data for Smart Home Solutions”, presented during the conference by Alicja Kwasniewska and Maciej Szankin from Gdansk University of Technology, Poland and from Intel, San Diego, USA

Additionally, best papers awards were presented to a few teams that received the best evaluation results (3 blind reviews for each paper). Two teams from our Department received the Best Paper Awards:

1.) Krzysztof Czuszynski, Alicja Kwasniewska, Maciej Szankin, Jacek Ruminski
Optical Sensor based Gestures Inference using Recurrent Neural Network in Mobile Conditions

2.) Adam Bujnowski, Kamil Osinski, Artur Polinski, Tomasz Kocejko, Jerzy Wtorek, Alexey Andrushevich, Electrical propterties of bath solution mixtures

Congratulations!

Awards

Awarded student project – ReFlex

Group project “ReFlex” was awarded and took second place in the Group Project contest.

Group project “ReFlex – Rehabilitation support system using EMG signal to control the game” was awarded in Group Project contest, edition 2017. The goal of the project was to prepare the rehabilitation support system designed to increase:
– Attractiveness of exercises,
– Effectiveness of rehabilitation process,
– Motivation to exercise,
– Muscle activity.
All above was realized by ensuring continuous measurement of electrical muscle activity. The signal is sent wirelessly to the mobile device, where it is used to control the dedicated application.
Project was realized by Anna Gozdan, Dorota Dettlaff, Paweł Troka and Hubert Toczko during last year, and was commissioned by Dynamic Precision sp z o. o.

Events

Summer School on Deep Learning – registration closed

We have just closed registration system for the International Summer School on deep Learning. So far, it has been a great success. Participants from 21 countries submitted their applications. Due to some limits we have to chose only 60 participants. Other candidates were added to the waitning list.

International Summer School on Deep Learning (2-6 July, 2018) will introduce participants with fundamentals of deep learning methods. Outside of the basecamp, special sessions and keynotes are planned to keep the audience up to date with the latest advances in this fascinating research area. Outstanding speakers and experienced instructors from all over the world will present scientific background, practical issues and perspectives for the future of deep learning methods. The keynote lectures and mini courses will cover topics ranging from fundamentals of neural networks to practical implementations and applications of deep learning. We invite students, researchers and professionals to participate in this unique event that will keep you on track, help unleashing new ideas and give you a chance to be a part of mind-blowing conversations.

More information: www.dl-lab.eu

Events

E-nose system tests in PKN Orlen

Multisensor electronic nose system developed at the Department of Biomedical Engineering is being tested in the air quality monitoring station in Płock. Station in operated by PKN ORLEN and equipped with high precision reference analyzers for continuous measurement of the most important air parameters such as sulphur dioxide or nitrogen oxides concentration. Tests are part of the partnership between PKN ORLEN and Gdansk University of Technology. The main purpose of the measurements is to check the possibility of using the electronic nose technology in an air quality monitoring.

Events

Deep Learning Lab

The DGX Station was successfully installed and started at our Faculty. We will soon start the experiments in our department that will be focused on training models for gesture recognition for the needs of a linear optical sensor (a part of smart glasses), for the recognition of face features (e.g., a nose) in thermal images and for object recognition of image sequences. The server will be also used during the International Summer School on Deep Learning that we are organizing in July 2018 (www.dl-lab.eu).

Projects

A compact air quality monitoring station

Together with PM Ecology company we will be working on the development of a compact air quality monitoring station. Research is partially supported by the National Centre for Research and Development (NCBR) under the so-called “fast-track” R&D funding (POIR 1.1.1). The project is designed to carry out industrial research and development work, which will allow the company to put into production a comprehensive air quality monitoring system with area maps of air pollution generated in real time and available online.

Events

DGX Station – faster Deep Learning (NVIDIA GTC 2017)

DGX Station – faster Deep Learning

Thanks to Faculty Founds we have bought the new NVIDIA DGX Station that will be delivered at the beginning of November. The server will allow efficient computations using newest GPU cards. It will be a great companion to the extraordinary cluster available at out supercomputer centre (TASK – www.task.gda.pl).

The DGX Station and related applications were presented during the NVIDIA GPU Technical Conference at Munich (10-12 October, 2017). The event was almost entirely dominated by presentations of the application of AI in practice. Many talks and demonstrations were dedicated to medical issues like automatic analysis of medical images (e.g., cancer recognition, tissue analysis in pathology, image quality improvements, etc.). Almost all medical imaging modalities were referenced during presentation. The practical role of AI and DL is increasing. Therefore, we are happy from new resources that will be available for our group.

Events

Visit of prof. Jacek Zurada – the candidate for…

Prof. Jacek Zurada has visited us during the National Conference on Databases (Infobazy 2017). In 1968 he graduated from the Gdansk University of Technology. His MSc thesis was in Control and Information Theory. In 1975 he received his PhD degree and became a post-doc at the Swiss Institute of Technology in Zurich. Currently, he is a candidate for the position of IEEE President Elect 2018.

Events

We are organizing HSI 2018!

We are happy to inform that the next edition of the IEEE/IES Conference on Human System Interaction will be organized in Gdansk! During the first week of July 2018 (4-6, July 2018) researchers from all over the world will visit Gdansk University of Technology to exchange new innovative ideas related to the field of human system interaction.

The HSI conference has been prepared as a platform for exchanging ideas, knowledge, skills and experiences in interactions between man and these systems. The HSI conference series has become the very important international conference in the field of Human System Interactions, human system science and engineering, human factors in large complex systems, human space computing and cyberphysical systems, human environment interactions, and social‐technical systems. This 11th international conference on Human System Interaction will be held in Gdańsk from the 4th to 6th July 2018. The conference will cover the theory, design and application of human‐system interactions in the areas of science, education, business, industry, services, humanity, environment, health, and government. Aside from the regular presentations, the conference will include keynote addresses from speakers who work both in industry and academia, and expects to attract more than 120 participants.

The General Chair of the Conference and the Organizing Committee are from the Department of Biomedical Engineering. As organizers we welcome all interested researchers, professionals and students to participate in this great event.

More information at: HSI Website

Awards

Best Paper Award – HSI 2017

Krzysztof Czuszynski, Jacek Ruminski and Jerzy Wtorek received the Best Paper Award during the 10th IEEE/IES International Conference on Human System Interaction. The title of the paper: Pose classification in the gesture recognition using the linear optical sensor.

Abstract:
Gesture sensors for mobile devices, which have a capability of distinguishing hand poses, require efficient and accurate classifiers in order to recognize gestures based on the sequences of primitives. Two methods of poses recognition for the optical linear sensor were proposed and validated. The Gaussian distribution fitting and Artificial Neural Network based methods represent two kinds of classification approaches. Three types of hand poses, differing in the number of fingers joined together, were investigated. The reflected light intensity pattern originated by hand located closely to the sensor was parameterized into 14 features. The change of reflection pattern originated by hand dislocation was reduced by application of two variable functions in the first of the methods. A one and two hidden layers topologies were considered in the neural network related approach. Both methods were designed with the use of a training set of samples and validated with another (testing) set. The results present the average poses recognition rate of 81.19% for Gaussian distribution fitting and 90.02% for ANN based method.
PDF – PRESENTATION
More info at: IEEE Xplore

Blog