Fraunhofer Center for Assistive Information and Communication Solutions – AICOS
26.09.2024
The Closing Event of the 15th edition of the Fraunhofer Portugal Challenge took place last Wednesday (Sep 25) at the Faculty of Engineering of the University of Porto (FEUP), recognizing the most innovative technological ideas from Portuguese universities of 2024. The idea contest, organized by Fraunhofer Portugal AICOS, seeks to bridge the gap between academic research and industry by awarding research projects with practical applicability and societal impact.
This year’s Challenge saw the distinguished participation of Prof. Fazel Ansari, from Fraunhofer Austria, who currently serves as Full Professor of Data-driven Maintenance Management at the Technische Universität Wien (TU Wien). Prof. Ansari, renowned for his extensive work in sustainable production and logistics, added invaluable insight into the evaluation process as part of the expert jury.
The panel of experts, which included several prominent figures from Fraunhofer Portugal AICOS and Fraunhofer Portugal, was critical in selecting the winning projects. These experts, many of whom are also deeply involved in the creation and support of spin-offs, provided their perspectives on how the submitted ideas could evolve into successful, marketable technologies. This panel, composed of specialists in areas ranging from artificial intelligence to business development, played a pivotal role in shaping the outcomes of the competition.
The Challenge has continually expanded its scope, most recently adding the Student Award category in 2023 to encourage early-stage Master's students to submit their innovative ideas. In addition to monetary prizes, the winners also benefit from mentoring provided by Fraunhofer Portugal AICOS researchers, fostering future collaborations and innovation-driven entrepreneurship.
Winning Ideas (MSc Thesis Category)
Diogo Pires (1st Prize) | University of Aveiro | “Development of an instrumented implant comprising capacitive technology to monitor the fracture bone healing”
This project proposes the development of a new bioelectronic plate for bone fracture fixation, equipped with capacitive sensors to monitor the fracture healing process. Currently, follow-up methods are based on image analysis, which involves some subjectivity, can only be performed in a hospital setting, exposes patients to high cumulative doses of radiation, and does not provide continuous information about the biomechanical state of the fracture. The new technology aims to overcome these limitations by offering real-time, continuous monitoring based on changes in the electrical properties of bone tissues during different phases of healing. Computational simulations and experimental tests show that this new system can identify the healing progress through variations in capacitive patterns. The potential of this innovation lies in reducing treatment time, preventing additional surgeries, and improving patients' quality of life, with possible clinical applications in the future.
Diogo Martins (2nd Prize) | University of Minho | “Deep Learning-based Posture Recognition for a Holistic Ergonomic Assessment Framework”
This project aims to reduce the risk of work-related musculoskeletal injuries, common in sectors such as agriculture and construction, where poor postures are often maintained for long periods. The solution proposes a technological tool for postural evaluation that helps increase workers' postural self-awareness by intuitively showing which postures they should avoid, as they present a higher ergonomic risk. It addresses the ergonomists' need for objective data and expedited tools to streamline their work. The system uses wearable sensors to continuously calculate ergonomic risk during a work shift in three modalities: global, by joint, and cumulative effect. Simultaneously, it detects the type of posture that led to this risk through Artificial Intelligence models that explain their decisions, ensuring reliability and robustness. At the end of the shift, a graphical interface presents ergonomists and workers with a report that links ergonomic risk to specific posture types. This solution has the potential to improve workers' musculoskeletal health and reduce productivity losses.
Miguel Peixoto | University of Minho | “Anomaly Detection as a Quality Control Tool in an Industrial Context”
Anomaly Detection for Quality Control in Industry: This idea aims to modernize the quality control process in factories, which often relies on manual inspections prone to errors. The proposal develops an automated system capable of identifying defects in products during production using artificial intelligence and computer vision. By detecting defects more quickly and accurately, this system helps reduce waste and improve production line efficiency. The solution is flexible, can be applied across various industries, and is easy to implement, contributing to more sustainable and competitive production. In the context of Industry 4.0, the digitization and automation of manufacturing processes are essential to maintaining global market competitiveness. This proposal focuses on developing an innovative system for industrial quality control, using artificial intelligence to detect defects in products during production. The proposed system replaces manual visual inspections, which are prone to errors and resource-intensive, with an automated solution that uses advanced anomaly detection algorithms and computer vision. This system is highly adaptable, capable of being implemented in different production lines and adjusted for various types of products, ensuring accurate and efficient defect detection. Furthermore, this solution offers a significant advantage by being easily integrated into existing factory systems, functioning as a plug-and-play tool. This means companies can quickly improve their quality control processes, minimizing production downtime and material waste, which in turn enhances the sustainability and efficiency of manufacturing processes. With the potential to transform industrial quality control, this idea not only helps companies become more competitive but also promotes more sustainable production aligned with the challenges of today's global economy.
Student Award Category
Martim Silva
Annually, around 9,926,100 individuals require cardiac surgery. After surgery, patients need rigorous monitoring and precise assessments to confirm post-operative improvement and identify the need for further interventions. However, traditional methods are time-consuming, subjective, and often inconsistent. This project proposes the development of a graphical interface that will allow medical teams to input simple characteristics of the post-surgical Electrocardiographic (ECG) signal. This interface will be integrated with a back-end that uses machine learning models to quantify patients' recovery status and the need for further interventions. The system will provide crucial information to support decisions on the next steps in treatment, optimizing time and medical team efficiency.
The Fraunhofer Portugal Challenge remains an exemplary initiative, driving forward a culture of innovation by recognizing groundbreaking ideas that align with Fraunhofer Portugal AICOS’ mission of advancing research with practical utility.
The Fraunhofer Portugal Challenge, known for encouraging creativity and technological innovation, features two categories: the Master Thesis Award and the Student Award. In the Master Thesis Award category, the best ideas based on academic theses are recognized with scientific prizes totaling six thousand euros, distributed among the top three (1st place - €3,000, 2nd place - €2,000, and 3rd place - €1,000). The Student Award category is open to all master's students who present the most promising technological idea, regardless of whether it is based on an academic thesis. In addition to a monetary prize of three thousand euros, the winner of this category also receives mentorship and guidance from FhP-AICOS researchers, providing a unique opportunity for the development of their idea.
Submitted ideas must align with the scientific areas developed by FhP-AICOS, which include Artificial Intelligence, Human-Centred Design, and Cyber-Physical Systems. This initiative aims to recognize and reward the most innovative ideas, but also to support students in transforming these ideas into practical solutions with real impact.
Organized since 2010, the Fraunhofer Portugal Challenge seeks to encourage cooperation between industry and the scientific community, motivating and rewarding research of practical utility through the awarding of prizes to students and researchers who best contribute to the philosophy of Fraunhofer Portugal AICOS: to develop research with practical applicability for economic development and to improve people’s lives.