Looking back at

EAISI Summit 2021

On Monday the 15th of November we had our yearly EAISI Summit. We planned to welcome our audience of at least 400 guests in the Evoluon, but due to COVID-19 measures, we decided to switch to a hybrid event last-minute. Our speakers presented from the live studio in the Evoluon and we invited our guests from TU/e and industry to join the program online.

The future impact of AI

This year's theme was The future impact of AI. The keynote by Jim Stolze was a lively and entertaining kick-off for the more in-depth sessions that followed. Six presentations were organized around our 'moonshot' topics; NextGen Industry, Healthcare Anywhere, Trustworthy Data, Responsible Mobility, and Thinking Assistants. We chose the format of ‘duo talks’. First someone from industry setting the scene and presenting an actual problem, directly followed by a scientist talking about his/her work in this context helped the audience understand the current and future impact of AI. These experts showed us how the EAISI Moonshots are already making great progress, thanks to a range of AI techniques. Another highlight was the Pitch Parade; short elevator pitches by PhD's and Posdocs gave us an opportunity to see some of the ways AI is being applied in further new research.

Having all the presenters on stage, with expert moderators asking the questions from the chat gave a lively interaction. With our small group of live attendants we provided the speakers with a visible human audience that helped them to give inspiring lectures about the many aspects of the AI research related to EAISI.

NextGen Industry will allow robots to work with specialists and help service organizations to bring the right information to the right persons. Healthcare Anywhere will speed up the diagnosis and treatment of cancer. Trustworthy Data combines the open source community’s ideals with data for a wide range of AI applications. Thinking Assistants showed up in several of the presentations but is also ensuring on-time delivery of luggage. Responsible Mobility will promote healthy lifestyles by changing our environment for the better. 

We are looking forward to welcome you all on-site next year!



Impression of the day

View the EAISI Summit 2021 photobook for an impression of the day!

recordings & slides


Liselotte Graas (Vanderlande)    |    Watch on YouTube


EAISI connection by Carlo van de Weijer - General manager at EAISI, TU/e
Watch on YouTube    |    Download slides


The future impact of AI by Jim Stolze - Tech/AI educator, founder at Aigency
Abstract    |    Watch on YouTube    |    Download slides

nextgen industry

Advancements in behavioral autonomy for medical robots by Marco Alonso (Philips) & Elena Torta (TU/e)
Abstract    |    Watch on YouTube    |    Download slides


Advancing cancer care through interpretable AI by Igor Jacobs & Jon Pluyter (Philips) and Fons van der Sommen (TU/e)
Abstract    |    Watch on YouTube


Short research pitches on various topics by PhD's and Postdocs
Watch on YouTube    |    Download abstracts and contact details


Prevention of overcrowding by incentivizing people to take a stroll on attractive paths; an AI approach by Matthijs Bootsman (Lynxx) and Soora Rasouli (TU/e)
Abstract    |    Watch on YouTube    |    Download slides


Open machine learning, automation, and shared digital twins by Joaquin Vanschoren (TU/e) in collaboration with Matthijs Punter (TNO)
Abstract    |    Watch on YouTube    |    Download slides


Thinking Assistants for Automated Material Handling Solutions by Dirk Fahland (TU/e) in collaboration with Joost van Montfort (Vanderlande)
Abstract    |    Watch on YouTube    |    Download slides


AI for a Digital Intelligent Operator by Wouter de Vries (KPN) and Jaron Sanders (TU/e)
Abstract    |    Watch on YouTube    |    Download slides


Liselotte Graas (Vanderlande) and Carlo van de Weijer (TU/e - EAISI)   |   Watch on YouTube

Abstracts & biographies

WELCOME | EAISI connection

Carlo van de Weijer | General Manager of EAISI


Carlo van de Weijer (1966) has a master’s degree in mechanical engineering from the TU Eindhoven and a PhD degree with honors from TU in Graz. He carries a broad experience in the automotive industry with a.o. executive positions at Siemens and TomTom. Currently he is General Manager of the Eindhoven AI System Institute (EAISI) at Eindhoven University of Technology. He advises ministries and industries around the world on the future of mobility and is member of the supervisory board of several high-tech companies and start-ups. He is an international speaker on exponential technology and the future of high-tech, amongst others as faculty member of Silicon Valley based Singularity University. Furthermore, he is a weekly columnist in the leading Dutch financial newspaper. Back to program ↑

KEYNOTE | The future impact of AI

Jim Stolze | Tech/AI educator & entrepreneur


We are living in extraordinary times. In the last decade we have digitized nearly every process, every aspect of business. We are literally data-rich, but information poor. The next phase will be to use artificial intelligence to derive new insights and value from that data. In his keynote Jim Stolze will paint a picture of what that looks like, what skills are needed and how we can make sure to innovate in a responsible way. → Watch the presentation on YouTube


Jim Stolze became known as the face of TED.com and as the founder of TEDxAmsterdam. Since 2017 he is co-founder of Aigency that implements artificial intelligence at companies and governments. Previously, he wrote books like "Sold out - Welcome to the attention economy" which was nominated as management book of the year in 2012 and pulled the strings at the largest websites in the Netherlands (Startpagina.nl and NU.nl) between 2003 - 2008. In 2018, he chronicled his experiences in the field of artificial intelligence in the book “Algorithmisation, get used to it!” He also launched the National AI Course together with Elephant Road and ICAI. In 2020 and 2021 he presented both the Dutch National AI course as well as the Brainport AI Course, a free online course for everybody who wants to learn more about artificial intelligence. Back to program ↑

NEXTGEN INDUSTRY | Advancements in behavioral autonomy for medical robot

Marco Alonso | Senior Architect at the Image Guided Therapy (IGT) Systems Mechatronics department of Philips IGT
Elena Torta | Control Systems Technology group, Mechanical Engineering at TU/e

→ Watch the presentation on YouTube


Image Guided Therapy robots are designed to consistently and efficiently help the user navigate equipment (such as catheters) inside arteries or veins, as well as to help visualize in three dimensions the anatomic regions of interest. These robots are manipulated manually to reach desired positions during clinical procedures in crowded and dynamic settings. While easy to operate, the user is required to always monitor the movements of the robot and to reposition it manually after every procedure. Moving from manual manipulation to robot autonomous decision making could enhance the user experience and increase usage efficiency. In this talk we are going to present the ongoing collaboration between Philips IGT (MDC cluster) and the CST group of TU/e to advance the behavioral autonomy of Image Guided Therapy robots. 


Marco Alonso is a Senior Architect at the Image Guided Therapy (IGT) Systems Mechatronics department of Philips. He obtained his M.Sc. degree in Embedded Systems from the TU/e in 2010. He has experience in the areas of software and mechatronics development together model-driven techniques for system development. He is responsible for the Mechatronics development and roadmap of robotic positioning systems for X-ray imaging. He is a Technical Leader and Architect working on the Mechatronics roadmap of the next generation of IGT systems. He is the work package leader for the data analytics within the IMOCO4.E. He is actively collaborating with the TU/e to enable the future generation of IGT robots.

Elena Torta is an assistant professor at the Control Systems Technology group (TU/e). She obtained her M.Sc. degree in Industrial Automation Engineering from the Università Politecnica delle Marche (IT) in 2009. In 2010 she joined TU/e as PhD student and performed research in the domain of cognitive autonomous robots and human-robot interaction for care applications for the EU-FP7 KSERA project. After obtaining her PhD degree (cum laude) in 2014, she joined ASML, where she worked as software architect and Matlab competence leader. In 2020 she joined the mechanical engineering department of TU/e as assistant professor. Her research is focused on autonomous robots and systems with applications in human-robot collaboration, mobile robots and digital twins and medical robotics. She is part of the EAISI digital twin lab. Back to program ↑

HEALTHCARE ANYWHERE | Advancing cancer care through interpretable AI

Igor Jacobs | Clinical Scientist Oncology at Philips Research
Jon Pluyter | Senior Usability Engineer at Philips Experience Design
Fons van der Sommen | Video Coding & Architecture Group, Electrical Engineering at TU/e


Early detection, accurate diagnosis, and optimal treatment selection are critical factors to improve outcomes for cancer patients. Radiological imaging has a central role in oncology care, but considerable challenges are to be overcome regarding its accuracy and efficacy. While AI holds great promise to support radiologic imaging in oncology, AI is rarely successfully adopted in clinical practice and therefore its impact on patient outcomes remains limited. Within the Eindhoven MedTech Innovation Center (e/MTIC) innovation ecosystem, a close multidisciplinary collaboration on oncology has been established between Catharina Hospital Eindhoven, Eindhoven University of Technology, and Philips Research and Design.
We aim to develop robust AI for oncology, shape clinician-AI collaboration, and evaluate adoption and use of AI solutions in the field. In our presentation, we will highlight how AI could improve cancer care and discuss what challenges need to be overcome to make AI solutions “ready” for real-world conditions in clinical practice. 


Igor Jacobs is working as a clinical scientist at the Hospital Services and Informatics department of Philips Research. He has studied Biomedical Engineering at the Eindhoven University of Technology, after which he did a PhD on MRI techniques for cancer imaging and therapy evaluation and a post-doc on cancer nanomedicine. At Philips, he currently focuses on development of healthcare information management solutions aimed at improving clinical workflows, multidisciplinary decision-making and above all, patient outcomes. He is responsible for translating unmet clinical needs into innovative solutions and validation of novel propositions, in co-creation with clinical partners. He has a passion for oncology research and gets energy from close multidisciplinary collaboration with clinicians and forging partnerships such as the Eindhoven MedTech Innovation Center (e/MTIC) oncology collaboration.

Jon Pluyter works as a Sr. Usability Designer and Experience Lead Design Innovation at Philips Experience Design. He designs for the human factor, ‘de menselijke maat’ in technology. He focuses on research and development of smart data- and artificial intelligence-driven medical applications for improvement of complex clinical decision-making; for example, to support teams of clinicians in diagnosis and treatment of cancer. Besides, he builds bridges between industry, academic and clinical partners to innovate in a multidisciplinary way, not for clinicians but with clinicians.

Fons van der Sommen is an assistant professor at the research group Video Coding & Architectures of the department of Electrical Engineering. He is a passionate scientist and educator, holding a MSc. in Electrical Engineering (2012, cum laude) and a PhD in Computer Vision (2017, cum laude). Driven by a personal tragedy, a passion for science and a desire to contribute to this world, he applies his understanding of machine learning and computer vision in healthcare, where he aims to develop novel assistive technologies for medical doctors, to support them with early diagnosis and effective treatment of disease. In particular, in the field of Medical Image Analysis, Van der Sommen's research focuses on Computer-Aided Detection and Diagnosis (CADe/CADx) for oncology, striving to increase the detection rates and early diagnosis of developing cancer -- thereby considerably increasing the chances of survival for patients.

Back to program ↑


→ Download all abstracts and contact details

1. Özge Tüncel | Lower Tier Sustainability Framework
2. Shervin Azadi | Urban Digital Twin
3. Robbert Reijnen | Understandable and Generalizable AI Hybridizations
4. Vahideh Reshadat | Knowledge Acquisition from Textual documents 
5. Dirk Aarts | Aristotle Cognitive Training
6. Lasitha Chamari Rathnayaka | Data Integration for Smart Demand Side Management Operations 
7. Ya Song | Improving heuristic algorithms with machine learning for sequential decision-making problems
8. Larissa Capobianco Shimomura | Graph Generating Dependencies for Data Quality
9. Andrii Kompanets | Intelligent Bridge Infrastructure Maintenance
10. Reza Refaei Afshar | Machine Learning for Ad Publishers in Real Time Bidding
11. Mariia Turchina | Extending human senses with AI
12. Debargha (Dave) Dey | Human Factors of Automated Vehicles: Communication between self-driving cars and other road users

Back to program ↑

TRUSTWORTHY DATA | Open machine learning, automation, and shared digital twins

Joaquin Vanschoren | Machine Learning, Mathematics & Computer Science at TU/e

in collaboration with Matthijs Punter | Senior Researcher ICT Data Ecosystems at TNO

→ Watch the presentation on YouTube


Machine learning, an important tool in modern scientific research and smart industries, can lead to breakthrough improvements in productivity and knowledge, but crucially relies on the availability of data and technical expertise. New tools are being developed a TU/e and TNO to ensure that data and expertise is available. OpenML is an open-source platform that aims to organize the world’s machine learning information. It allows anyone to share data, retrieve it in a way that directly feeds into machine learning workflows, and share the resulting models and evaluations in a reproducible way. It learns what works well and uses this to automate machine learning (AutoML). However, data is not always universally available, due to technical constraints (technical lock-in), or commercial or IP-considerations (vendor lock-in). TNO and TU/e are collaborating on shared digital twins - based on new Industry 4.0 standards – that can be used to collect and share asset data, and are collaborating with industrial partners to scale-up these solutions.


Matthijs Punter is a senior researcher TNO ICT Data Ecosystems department. His research focuses on large scale data sharing in communities of organizations. A key concept is the principle of data sovereignty: retaining control over the sharing and usage one’s data. He is a member of the International Data Spaces Association, which develops standards in this field. In collaboration with Brainport Industries and TU/e he has been working for many years in the adoption of these approaches in high-tech manufacturing supply chains. 

Joaquin Vanschoren is an assistant professor at the Eindhoven University of Technology (TU/e). His research focuses on the automation of machine learning (AutoML) and Meta-Learning. He co-authored the books 'Automatic Machine: Methods, Systems, Challenges', published over 100 articles on these topics, and received an Amazon Research Award, Azure Research Award, the Dutch Data Prize, and ECML PKDD demonstration award. He co-founded OpenML.org, an open science platform for machine learning, and is a founding member of the European AI associations ELLIS and CLAIRE. He has been tutorial speaker at NeurIPS and AAAI, and has given more than 30 invited talks, including the UN Global Summit on AI for Good, VLDB, IDEAL, and workshops at NeurIPS, ICML, and SIGMOD. Back to program ↑

RESPONSIBLE MOBILITY | Prevention of overcrowding by incentivizing people to take a stroll on attractive paths; an AI approach

Matthijs Bootsman | Data Scientist at Lynxx
Soora Rasouli | Urban Planning & Transportation, The Built Environment at TU/e


As it has many things, the COVID-19 pandemic has severely impacted the way we look at crowding. As a result, crowding levels that we used to be comfortable with, were actively managed and – in most cases – successfully prevented. It was in these times that the Municipality of Amsterdam has started a trial to try and avoid over-crowdedness at hotspots in the city center. In partnership with HERE and 9292, data scientists at Lynxx are currently using AI to predict current and future crowdedness on many such hotspots, using data from smart cameras, probes, roads, public transport and 9292-travel requests. Using these results, HERE and 9292 have implemented several measures in their navigation and trip-planning applications to notify their users of future crowdedness. Directing crowds away from key locations, however, remains a challenge how to incentivize the travelers to move away from their original destinations and possibly choose alternative, less crowded places. One of the incentives could be encouraging them to walk a short distance in a pleasant and safe environment to enjoy less crowded visiting areas. Semantic segmentation of images from Google Street View, using neural network algorithms, can be used to provide a measure of safety, security, greenery and comfort of the walking path to the travelers and as such serves as a trigger for them to reach alternative (nearby) locations by means of a healthy transport mode.


Matthijs Bootsman is a seasoned Data Scientist working at Lynxx, a Netherlands/Australia based Data Science consultancy firm operating mainly in the transport and mobility domain. Matthijs has a background in econometrics, statistics and machine learning. He has worked on several projects that attempt to predict and influence crowding in both public transport and the public domain. In his current role as Lead Data Scientist at 9292 - the Netherlands’ leading public transport trip planning application - he aims to provide users with the clear and real-time impacts of each choice (to be) made. This refers to possible delays and sustainability, as well as to crowding. The latter goal ignited a partnership with the Municipality of Amsterdam, where the aim lies at monitoring, informing and influencing visitor flows through the city.

Soora Rasouli has background in transportation engineering and is specialized in developing models on citizens’ travel behavior with special attention to the recent advances in transport technology including EV, autonomous driving and Mobility and the interaction between mobility demand and supply. Decision making under uncertainty and systematic incorporation of uncertain events into modeling of travel behavior are other aspects of her research focus. She is the chair of Urban planning and Transportation group in Eindhoven university of Technology (TU/e) She has published about 60 papers in leading journals in transportation. She is editorial board member of Transportation Letter, Journal of Urban Planning and Development,  International Journal of Urban Science, Travel Behaviour and Society and Journal of Traffic and Transportation Engineering. Back to program ↑

THINKING ASSISTANTS | Thinking Assistants for Automated Material Handling Solutions

Joost van Montfort | Service Design Lead at Vanderlande
Dirk Fahland | Process Analytics on Multi-Dimensional Data, Mathematics & Computer Science at TU/e


Vanderlande provides automated Material Handling Solutions for major Airports, Parcel and Warehouse companies around the world. With the ongoing trends such as labor scarcity and increasing customer expectations on pro-activeness and predictability, we identified the need to enhance our systems with AI techniques that augment our system operators. Our aim is to support operators to make better decisions in their daily work resulting in enhanced system performance for our customers.

Laying the foundations for a “Thinking Assistant” that augments our operators in system-level decision making needed a paradigm shift in how we analyze and predict behaviors from event data. Taking a different perspective when analysing tens of thousands of bags at once allowed us to detect and analyse complex patterns in system-level dynamics. Operationalizing these patterns enhances our operators with the intelligence to let an individual bag make it to its flight.


Joost van Montfort is the Service Design Lead of Vanderlande’s Digital Service Factory. With a background in Industrial Engineering & Management Science his passion for data arouse during a multi-year assignment at London Heathrow. After this experience he established the Data Science & Data Service development team, which is now accelerated as Digital Service Factory within Vanderlande’s Technology department. It aims to connect Vanderlande's global Installed Base and leverage the obtained data to create new insights and digital solutions. He builds bridges between Technology and Business, accelerating Vanderlande’s Servitization journey by developing new digital solutions. These focus on continuously improving the operational performance of Vanderlande’s systems and enhancing the operational, tactical and strategic decision-making at customer sites. Given the complex nature of Vanderlande’s systems and the hundreds of decision made per day by a multitude of operators in various domains, "Thinking Assistants" are considered necessary to realize Vanderlande’s ambition.

Dirk Fahland is an Associate Professor in Process Analytics on Multi-Dimensional Event Data at Eindhoven University of Technology (TU/e). He is responsible for the Master Data Science and Artificial Intelligence at TU/e. His research area is the analysis and improvement of complex, distributed systems through event data, process mining, and explainable models. He specifically studies cause-effect relations and emergent behavior in networks and dynamic systems as a whole. Insights gained in numerous industrial projects led to the idea of encoding behavioral information in knowledge graphs, a cornerstone of "Thinking Assistants". Back to program ↑

NEXTGEN INDUSTRY | AI for a Digital Intelligent Operator

Wouter de Vries | Architect at KPN
Jaron Sanders | Stochastic Operations Research, Mathematics & Computer Science at TU/e


The telecommunication sector is constantly innovating to enable the digital society of tomorrow. Shaping the future of our real and digital lives by seamlessly connecting all devices and unlocking new experiences such as virtual and augmented reality. Enabling the secure and reliable exchange of data and information for businesses and providing them with new technologies like 5G, Internet of Things and fiber-to-the-home give them a cutting-edge advantage. Artificial Intelligence is at the heart of the future proof infrastructure to shape this digital transition. In a well-balanced innovation ecosystem together with knowledge institutes like EAISI, ingenious AI applications are being developed. Two EAISI-projects will be highlighted in the presentation:
•    Learning algorithms for matching queues in call centers
•    Stochastic clustering for faster reinforcement learning


Wouter de Vries has a master’s degree in applied physics from the state university Groningen and PD. Eng. in Design and Technology of Instrumentation from the TU Eindhoven. After joining KPN Research in 1994 he gained a wide experience in the field of telecommunications. Currently, he works as an architect in KPN’s Technology Office in the field of artificial intelligence and machine learning and as a visiting lecturer at the TU Delft. His focus is mainly on bridging the gap between theoretical and conceptual AI models and creating value in practice for KPN and its customers. He was recently granted the VNO NCW’s Costa Award for outstanding collaboration between a corporate and a start-up.

Jaron Sanders received in 2012 M.Sc. degrees in Mathematics and Physics from the Eindhoven University of Technology, The Netherlands, as well as a PhD degree in Mathematics in 2016. After he obtained his PhD degree, he worked as a post-doctoral researcher at the KTH Royal Institute of Technology in Stockholm, Sweden. Jaron now works as an assistant professor at the Eindhoven University of Technology. His research expertise is on optimization algorithms for large complex systems. His focus is on techniques on the intersection of probability theory / statistics / operations research, and on applications in machine learning / data science. Back to program ↑


Liselotte Graas | Vanderlande, Product manager | Creative Strategist

In 2015, Liselotte was part of the TU/e team developing the world's first energy-positive family car: Stella. They became world champion in Australia and all proceeded their careers in technology. Recently Liselotte created a documentary in which she tracks down her peers, which led her to Taiwan, India, Boston, and Manchester. As productmanager at Vanderlande she is responsible for the innovative portfolio of 'Robotics & Autonomous Vehicles'.