Last Friday, we had the distinct pleasure of attending the doctoral thesis defense of Amaia Pikatza Huerga, titled: “Human-Centred Machine Learning for Health and Cognitive Modelling.”

Her research tackles one of the greatest challenges in contemporary technology: how can we trust Artificial Intelligence in critical sectors like healthcare? Amaia’s answer is definitive: by keeping people at the center.

Core Pillars of Research

Through four high-impact case studies, Dr. Pikatza Huerga demonstrated how predictive modeling can coexist with human oversight:

• ❤️ Heart Failure: Improving predictive accuracy while ensuring clinical relevance.

• 🏥 Multimorbidity: Navigating complex, co-occurring health conditions.

• 🧠 Eating Disorders: Developing data models to better understand behavioral health.

• 💡 Human Creativity: Explored through the lens of cognitive modeling.

The Impact of Explainable AI (XAI)

Amaia has proven that it is entirely possible to build high-performance predictive models that remain transparent and interpretable. Her work highlights the immense value of Explainable AI (XAI), allowing medical and psychology professionals to understand the “why” behind every prediction. This transparency paves the way for safer, more confident, and inherently human decision-making.

Congratulations, Dr. Amaia Pikatza Huerga, on this incredibly well-deserved milestone! Your contribution to a more ethical, comprehensible AI that aligns with real-world needs is truly inspiring. We wish you the absolute best in your next chapter! 🚀

On May 6 and 7, 2026 , the AcceptRemote project consortium met in Athens, Greece, for its Kick -off Meeting , marking the official start of this Horizon -funded initiative. Europe . The meeting brought together partners from all over Europe just days after the launch of the project, which will work over the next three years on the development and validation of remote operation solutions for connected, cooperative and automated mobility (CCAM) .

The University of Deusto – DeustoTech was represented by Leire Serrano and Erick Rodríguez , who actively participated in the consortium’s working and coordination sessions.

First face-to-face meeting to align objectives and activities

The meeting provided the first opportunity for the consortium members to meet in person, establish working relationships, and align a common vision regarding the project’s objectives, activities, and expected outcomes.

During the first day, the partners presented the various technical work packages and unveiled the three pilot demonstrations that will take place in Istanbul, Toulouse , and Munich . Discussions focused on key aspects related to remote operations, such as trust , scalability , communication reliability , and user acceptance , helping to establish a shared understanding of the challenges and opportunities that AcceptRemote addresses .

Preparing for the next three years of the project

The second day was dedicated to laying the groundwork for the project’s implementation over the next three years. The partners discussed the pilot validation methodology, communication and dissemination activities, data management and Open Science practices , as well as the overall project management framework.

Special attention was also paid to collaboration processes, coordination mechanisms, and the practical aspects associated with the execution of a Horizon project. Europe under the lump sum funding model .

The meeting also allowed for the clarification of responsibilities, the presentation of upcoming planned activities, and the coordination of the first steps of this new European collaboration.

An important step for the future of remote operations at CCAM

This two-day meeting laid the groundwork for effective cooperation across all work packages and marked a significant milestone in the launch of AcceptRemote . The project is now entering a phase of intensive collaboration aimed at advancing innovative solutions that contribute to the safe, reliable, and widely accepted deployment of remote operations in the field of automated mobility.

The consortium thanks all partners for their active participation, the collaborative spirit shown and the valuable contributions made during the working sessions in Athens.

🌱 Call to action at the University of Deusto! Propose a challenge and win great prizes!
The LOTUS team invites the entire UD community (students, researchers, faculty, and staff) to propose real-world challenges to lead the green and digital transition.

🚀 The goal? To find “seeds of innovation” that will then be explored in a hands-on Design Thinking and Artificial Intelligence course at the Bilbao Campus.

🔍 We’re looking for challenges in:

● Circular economy and sustainable models.

● Green technologies and climate innovation.

● Smart cities and digital tools.

● Inclusive entrepreneurship and gender perspective.

🎁 Why participate? In addition to seeing your idea transformed into a real solution, the 3 best proposals will receive prizes and a certificate of recognition! 🏆
📅 Deadline: June 30, 2026. 📝 How to register? https://docs.google.com/forms/d/e/1FAIpQLSegzpUMnKZ4r30soMXOLgUvisJlZ7dUbw506sZUV-nFyzU55A/viewform
No prior experience needed, just a desire to make an impact! ⬇️
#DeustoUniversity #LOTUSProject #Sustainability #DigitalInnovation #DesignThinking

The University of Deusto launches its Research Portal, designed as a strategic tool to strengthen the visibility, reach and impact of its institutional research activity.

This platform is a public gateway to systematised information on the university’s research activity. In its first phase, it allows users to explore, in a clear and visual way, information on research staff, research teams and scientific outputs, including over 34,600 publications (around 12,600 open access), defended doctoral theses, collaborations, and overall indicators. This provides an up-to-date and accessible view of our research activity and its evolution over time.

These data are displayed automatically from the reference sources WoS, Scopus and Dialnet. In subsequent phases, the portal will continue to be enriched with data on projects, establishing itself as a comprehensive tool for managing research outputs and the profiles of research staff.

The portal enables the academic community and the wider public to access relevant information on research areas, results and contributions. Through this initiative, the University highlights the value its teams bring to society and the scientific community, while providing up-to-date and reliable indicators.

It is also a key tool for promoting collaborations, responding to requests for specialised knowledge, and strengthening the transfer of results to the socio-economic environment. With its launch, the University takes a step towards making its contribution more visible and reinforcing its commitment to open science, transparency, and the return of knowledge to society.

More information: portalinvestigacion@deusto.es 

Modern industry faces a critical challenge: the need to make intelligent decisions in real time directly on the production floor. Artificial Intelligence solutions based on the cloud often encounter barriers of latency and connectivity. To overcome this, at DeustoTech we are leading a project that redefines the limits of Industrial Edge-AI.

The “Dúo Dinámico” of Advanced Automation

We are carrying out a pioneering integration that combines the best of two technological worlds. On the one hand, the deterministic reliability and proven robustness of the Siemens S7-1215C automaton, a standard in industrial control. On the other hand, the immense flexibility and parallel processing capacity of the FPGAs of the AMD Kria KV260 family.

We call this combination the “dynamic duo” of IA Industrial. By combining these technologies, we can run complex Deep Learning models directly on the machine. The result? A drastic reduction in latency, bringing response times down to microseconds, something fundamental for the control of critical processes, advanced robotics and industrial artificial vision.

The Research Challenge: Performance vs. Consumption

However, taking artificial intelligence to the extreme (Edge) is not just about computing power. As part of the work of our research group at DeustoTech, we are addressing the core of the problem in embedded systems: the delicate balance between computational performance and energy consumption.

A massive industrial deployment of AI will not be viable if it is not sustainable. Therefore, our research focuses on the development and application of advanced techniques to optimize Deep Neural Networks (DNNs). Through strategies of compression, quantization and efficient design of algorithms, we ensure that the inference is not only ultra-fast, but also:

Energy efficient: Minimizing the consumption of watts for each inference made in the FPGA.

Thermally stable: Reducing heat dissipation, crucial in closed industrial environments.

Highly scalable: allowing its integration in diverse production lines without requiring oversized electrical infrastructures.

Looking towards Industry 5.0

The work we are developing from DeustoTech not only solves a current technical problem, but also lays the foundations for Industry 5.0, where machine intelligence collaborates efficiently, quickly and, above all, sustainably.

 

The Technological Institute of the University of Deusto has recently accepted the defense of Asier González’s doctoral thesis, entitled “Reinforcement learning in collaborative robotics: a study on the influence of human parameters in the development of robotic skills”.🤖

✅The research addresses one of the main current challenges in human-robot interaction: the development of robotic systems that are not only efficient from a technical point of view, but also are safe and comfortable for the people they interact with.

✅In this context, the thesis proposes the integration of human variables – such as comfort – directly in the process of training robots through reinforcement learning techniques. For this, the work combines the use of simulation environments, the design of user-centered reward functions and their subsequent experimental validation with human participants. 🧑

The thesis has been directed by Andoni Eguiluz and Iñaki Vazquez, and constitutes a relevant contribution towards the development of collaborative robotics systems that are more adaptive and aligned with human needs.

This type of research reinforces the commitment of the University of Deusto to the advancement of innovative technologies centered on people, especially in areas such as collaborative robotics and artificial intelligence.

From the Technological Institute, we want to congratulate Asier for this excellent work and for his contribution to the progress of research in this field. 👏✨

Kaijing Lyu is a PhD candidate at the Beijing Institute of Technology and recently concluded a one-year visiting research stay at the Deusto CCM – Chair of Computational Mathematics at DeustoTech, UD – University of Deusto. Her stay has been made possible through a collaborative agreement between the BIT and the UD.
✅Her research focuses on the control of partial differential equations and neural operators has been funded by the BIT and DASEL- Ciencia de Datos para Redes Eléctricas – Data Science for Electrical Grids, and the ERCAdvanced Grant Control for Deep and Federated Learning (CoDeFeL) led by her co-supervisor, Prof. Enrique Zuazua.
✅During her time at DeustoCCM – Chair of Computational Mathematics University of Deusto, Kaijing made significant contributions to four research papers. Two of these were co-authored with Umberto Biccari, Co-Principal Investigator of the DASEL project.
👏✨We hope that Kaijing’s productive stay with us will contribute to her ongoing success and further strengthen the collaborative ties between our institutions. We were delighted to have her as a member of our Deusto UD community and wish her the very best in all her future academic and personal endeavors.👏✨

María Isabel and Octavio, students of the Master’s in Innovation and Technology (dual track), won first place in the prestigious European hackathon (https://thelotushackathon.eu/) of the LOTUS project (https://thelotusproject.eu/), held last December 2025. Their project, named CityPulse, stood out among various international proposals for its human-centered and technological approach to urban mobility.

The hackathon, which seeks advanced technological solutions to address different challenges framed within the Sustainable Development Goals (SDGs), recognized CityPulse as the most disruptive tool for transforming the pedestrian experience.

CityPulse: Much more than a map

Unlike conventional navigation apps focused solely on speed, CityPulse prioritizes user well-being and safety. It’s a platform that generates a Safety & Livability Score for each street, analyzing complex variables such as:

. • Infrastructure: Public lighting, sidewalk width and presence of ramps.

• Physical environment: Steep slopes and pavement condition.
• Social activity: Land use and commercial density.

The “Conversational Bot” Revolution

What captivated the jury of the LOTUS project was its conversational interface. CityPulse includes a bot that accompanies the user in real time. If a person feels unsafe on a dark stretch of road or needs to find an area with more people, they can communicate this to the system using their voice. The platform dynamically recalculates the route based on the pedestrian’s expressed feelings and preferences.

“We wanted to create a tool that not only tells you where to go, but also listens to you and adapts to how you feel in the city,” explained María Isabel and Octavio, award recipients during the ceremony.

Social impact and accessibility

CityPulse’s versatility allows it to offer customized solutions for different user profiles:

1. Women: Prioritize routes with greater urban presence and better lighting, avoiding isolated areas.
2. People with reduced mobility: Eliminate physical barriers and steep inclines, ensuring 100% accessible routes.
3. Tourists: Transform the journey into a guided experience that highlights local businesses and cultural heritage.

Data for more humane cities

In addition to the direct benefit for citizens, CityPulse serves as a source of strategic insights for public administrations. Through the analysis of aggregated and anonymized data, governments can identify critical points in the city that require improvements in lighting or accessibility, fostering more proactive and inclusive urban planning.

With this achievement, Octavio and María Isabel not only highlight the potential of academic talent in the dual education model, but also position CityPulse as a benchmark for the future of urban navigation.

On the 16th of October 2024, the TANGENT project final event was held in Les Ateliers des Tanneurs, Brussels The event covered the presentation of the key outcomes of the project, presented by different partners. The TANGENT dashboard and related solutions, which provide enhanced information services for multimodal transport management, real-time traffic management services, and transport network optimisation for transport authorities. Our research methodologies and tested technologies for traffic management such as travel behaviour analysis, transport prediction and simulation, transport network optimisation and data gathering, harmonisation and fusion. The technical challenges in multimodal network and traffic management which were discussed on a panel moderated by Leire Serrano (Universidad de Deusto) with Eleni Vlahogianni (National Technical University of Athens), Athina Tympakianaki, PhD (Aimsun), Ynte Vanderhoydonc (imec), Antonio D. Masegosa (Universidad de Deusto), Marco Comerio (Cefriel), Tiago Dias (A-to-Be). The view of transport managers on the TANGENT tools tested in the cities: Rennes, Greater Manchester and Lisbon. This was also discussed on a panel which included Hannah Tune (Transport for Greater Manchester), Frédérique Baudouin (Rennes Ville et Métropole) and Joana Cunha (Carris), and moderated by Lucie Tristant (ID4MOBILITY). The policies and governance structures behind multi-actor cooperation in traffic management. We finished with a workshop on the future of the TANGENT tools: exploitation, policy and future roadmap.

A two-day CONDUCTOR General Assembly was organised by technical University of Munich (TUM) in Munich, Germany, 18-19 September 2024.

The consortium reviewed the progress of the work in the first 23 months of the project, updating on the different domains: traffic simulation scenarios in the light of CAVs, microsimulation environments and integration of public transport lines. Integration process of the modules in the overall platform.

DeustoTech presented the achievements on the domain of traffic management solution for signal vehicle couple control with CAVs and optimization of freight deliveries in urban areas using passenger transport, particularly Demand Responsive Transport (DRT).

Europe plan to transition from a fossil fuel-based economy to a renewable energy system by 2050, with electricity as the primary energy carrier. Central to this vision are Positive Energy Districts (PEDs)—urban areas that generate as much energy as they consume annually.

DEUSTO SUSTAINABLE RESEARCH GROUP has recently published a paper that focuses on a PED model featuring centralized energy storage aimed at optimizing electricity self-sufficiency and self-consumption. Utilizing a fuzzy logic-based energy management system, the model adjusts the state of charge of energy storage by considering local electricity production, consumption, and tariff contracts.

Testing this model in a PED of 360 households in Bilbao, Spain, researchers evaluated various scenarios, including different storage sizes, climate change effects, and extreme energy price fluctuations. Findings indicate that the PED can achieve up to 75.6% self-sufficiency and 76.8% self-consumption, with climate change potentially enhancing these figures.

This research underscores the viability of PEDs in advancing Europe’s renewable energy goals, presenting a robust framework for urban energy resilience and sustainability.

A two-day CONDUCTOR General Assembly was organised by Jozef Stefan Institute in Ljubljana, Slovenia, 10-11 April 2024. The consortium reviewed the progress of the work in the first 18 months of the project, updating on the different domains: – Development of microsimulation environments and integration of public transport lines. – Data collection and creation of realistic scenarios for accurate simulations. – Linking FleetPy with Aimsun Next for improved analysis functions. – Review of literature and finalisation of algorithms for optimisation methods. – Extension of software modules and facilitation of software integration. – Further developments in the areas of data management, analysis and smart data utilisation. Our colleagues Leire Serrano and Arka Ghosh presented the achievements on the domain of: – Traffic management solution for signal vehicle couple control – Optimization of freight deliveries in urban areas by using Demand Responsive Transport (DRT)