News and Events

Fraunhofer Portugal Challenge 2014: and the winners are…

The closing event of the Fraunhofer Portugal Challenge 2014 took place last 29th of October. After the six finalists’ presentations, the jury deliberated and announced the winners of this year’s edition.

Each year, Fraunhofer AICOS organizes the Fraunhofer Portugal Challenge, an idea contest intended to reward the scientific investigation with “practical utility” developed in Portuguese Universities. Highlight for the increase of the number of applications which has doubled this year.

 

Best ideas in the PhD Category:

1st - 3.000€

"A Novel Functional Electrical Stimulation System and Strategies for Motor Rehabilitation"
Paulo de Melo - Instituto Superior Técnico

Paulo de Melo aimed to develop new systems and strategies, using Functional Electrical Stimulation (FES), towards a model-based correction of such disabilities, namely drop foot, one of stroke’s most frequent consequences. This new stimulation system, brings the flexibility back to the user, by using a modular stimulation architecture and enabling real time control of all the electrical pulse’s parameters. This makes the system ideal as a basis for the development of novel neuroprostheses, but also for other more generic applications such as muscle rehabilitation systems. With its modular architecture, any desired number of stimulation channels can be easily added and can even be embedded in custom hardware solutions. Successful bench testing and experimental trials, revealed a safe, robust and reliable system, capable of promoting forceful contractions.

 

2nd - 1.500€

"earGram: Composing Music by Selection"
Gilberto Bernardes - Universidade do Porto

Gilberto Bernardes proposed an analysis-synthesis framework for assisting musicians in the tasks of selecting and assembling audio signals, namely by incorporating algorithmic music strategies in the processing chain. Gilberto Bernardes addressed problems raised by the implementation of audio signals in algorithmic composition by creating higher-level descriptions of sound objects, which drastically reduces their audio data representation and yet provides meaningful and highly flexible information. EarGram, the proof-of-concept software developed as a result of this study, integrates the proposed analysis-synthesis framework in a concatenative sound synthesis system.

 

3rd - 1.000€

"Transparent and/or flexible low-cost electronics with a-GIZO TFTs"
Pydi Ganga - Universidade do Porto)

Pydi Ganga developed a framework to support electronic-circuit and system-level design with a-GIZO TFT technology. This work resulted in a novel accurate device model for circuit simulations and high-performance circuits that are experimentally validated through measurements. The proposed device model is simple, accurate, and continuous, with relative small development time. It is based on an equivalent circuit approach with neural networks. As the device model is developed solely from measured data, Pydi Ganga has characterized the static and dynamic behavior of the device. This model is further implemented in Verilog-A and used for circuit simulations. This work is followed by the design of different traditional and novel circuits only with n-type enhancement transistors, taking into account the threshold-voltage variation with respect to bias stress and poor intrinsic mobility compared to crystalline silicon.

 

Best ideas in the MSc Category:

1st - 2.000€

"Microwave Imaging of the Axilla to Aid Breast Cancer Diagnosis"
Ricardo Eleutério - Universidade Nova de Lisboa

Microwave Imaging is one of the most promising imaging modalities for breast cancer early screening and monitoring. The study developed by Ricardo Eleutério tests the feasibility of imaging the axilla region by means of an Ultra-Wideband Microwave Imaging radar system. This approach involves illuminating the axilla with an Ultra-Wideband pulse, and recording the resulting backscattered signals. Subsequently, these recorded signals are processed: firstly, a skin artefact removal is applied to remove the skin reflection; secondly, a beamformer is used to spatially focus the backscattered signals so that returns from high scattering regions (e.g. metastasised lymph nodes) add coherently and returns from clutter add incoherently; and thirdly an image of the dielectric scatterers in the axilla is created.

 

2nd - 1.000€

"Speech and language technologies for aphasia therapy"
Anna Pompili - Instituto Superior Técnico

Anna Pompili developed VITHEA (Virtual Therapist for Aphasia Treatment), an on-line platform for the treatment of aphasic patients, which integrates state of the art speech and language technologies to provide automatic word naming exercises. The VITHEA platform is designed to act as a “virtual therapist”, which simulates an ordinary speech therapy session, evaluating the patient’s answer by means of the use of automatic speech recognition (ASR). The virtual therapist then provides a vocal feedback via text-to-speech synthesis techniques.

 

3rd - 500€

"SmartGuia: Shopping Assistant for Blind People"
Karen Duarte - Universidade de Coimbra

The idea presented by Karen Duarte consists of an assisting technology to help visually impaired persons on theirs trips to public buildings, particularly in shopping centers. Public buildings are changing constantly, often people have to take different routes to reach known destinations, and new spaces and services are available. This feature is clearest on shopping centers, as brands and store owners are constantly changing spaces, structures and services to attract and please customers. This dynamic environment is almost always signaled and labelled with visual marks and signs, and several other useful information is available through visual means. Therefore blind and partially sighted people are deprived from a strong and widely used information channel, leading to their deviation from community daily-life.