Projects

Coming soon! 

Coming soon! 

BOCEMON is an VLAIO imec.ICON project. The consortium consists of IDLab-Ghent University-imec & MOSAIC- UAntwerp - imec, Quares, Dyamand & DAIKIN. 

Sepsis is a syndrome caused by an infection and leads to organ failure and death. In the US, sepsis is responsible for almost half of all in-hospital deaths. Mortality also increases with the degree of severity of sepsis. In addition, patients with sepsis spend an average of 75% longer in hospital compared to other conditions. Importantly, the timing of sepsis diagnosis is strongly related to survival. It is therefore a serious condition that requires a rapid and multidisciplinary approach, both within hospitals and in the wider community.

However, within hospitals it is observed that vital parameters that may indicate incipient sepsis are not consistently and adequately recorded in the electronic patient file (EPD). Systemic Inflammatory Response Syndrome (SIRS) is a clinical condition that may precede sepsis. Many hospitals have an Early Warning Scoring (EWS) system in which certain of these SIRS criteria are systematically measured. A critical aspect of this current situation is the manual completion of parameters to determine SIRS, e.g. respiratory rate, which poses a significant hurdle to the accurate recording of vital parameters in the EPD. This aspect highlights the urgent need for automated and standardized systems for recording vital parameters, including respiratory rate, to improve the quality of care and reduce the impact of SEPSIS. 

The proposed project aims to address sepsis with a holistic and multidisciplinary approach that supports healthcare providers and integrates new technologies to improve the detection and treatment of SEPSIS. In concrete terms, S.E.P.S.I.S. Connect tackles the 4 major challenges needed to detect SEPSIS more quickly in patients in Belgium:

1. Equipping patients with biosensors with the aim of monitoring them more accurately and continuously regarding the risk of sepsis.

2. The development of smart alarms, specifically tailored to the sepsis case, with the aim of supporting nursing staff in the large-scale follow-up of patients.

3. Setting up a safe and reliable connectivity solution that enables hospitals to continuously monitor patients both in the hospital and outside.

4. Better awareness of SEPSIS among general practitioners and home nurses in order to register and/or monitor patients more quickly via the developed monitoring solution.

This project is funded by the FOD and is a collaboration of the following partners: AZ Groeninge Kortrijk, Ziekenhuis Oost-Limburg, Sint-Jozefskliniek Izegem, Proximus NXT, imec. 

Coming Soon

PACSOI is an VLAIO imec.ICON project. The consortium consists of IDLab, FAQIR Foundation, FAQIR Institute, MoveUP, Byteflies, and AContrario. 

Flanders is a region where the water stress level is high. At the same time water companies estimate they lose 60 million cubic meter of water a year due to leaks in the network, representing almost a fifth of the annual water consumption. More than half of the water loss is caused by small or historic leaks in the network. Another third could be attributed to leaks in the connection between the consumers’ water meters and the network pipes.

Small or historic leaks and leaks at the connection points are difficult to detect, and therefore remain largely unsolved nowadays. To improve this situation, the Flemish government set forth an ambitious target to reduce this water loss to half of what is internationally accepted as unavoidable loss in a water distribution network.

The SmartWaterConnect 5.0 project aims to optimize the localization of leaks, both small and historic leaks in the network, as well as leaks in the connection between the network and the consumers’ smart meters. The goal is to reduce the loss, often referred to as non-revenue water, for De Watergroep down to 8%. De Watergroep is the largest water company in Flanders, supplying drinking water for three million customers.

To do so, the project will build the missing link in water leak localization. We will bring together data from various sources and develop robust algorithms using hybrid AI for a fine-grained leak localization. The data will stem from a fine-grained deployment of fixed sensors (i.e., the smart meters) at the consumer’s location, together with dynamically positioned mobile sensors to pinpoint leaks in the public network.

SmartWaterConnect 5.0 is an VLAIO imec.ICON project. The consortium consists of De Watergroep, Hydroscan, Hydroko, Eunoia Studios & IDLab-Ghent University-imec 

RE-ENNOVATE is an ambitious 3-year project to make home renovation smarter, cheaper, faster and better and re-innovate the process of energetic renovation. As the need for more energy efficient homes is only amplified by the current energy crisis, new tools are necessary for homer owners and professionals to understand which renovation measures work for them. With novel data-driven methods, RE-ENNOVATE will make the benefits and cost of renovations more transparent to individuals, propose personalized renovation packages, and increase automation, while continuously improving all steps by a closed feedback loop. The research consortium includes WTCB, imec, Ghent University, Adaptive, Mosard, Scone, Leuven2030 and Vandereyt. The project runs in cooperation with Flux50 and VLAIO.

Today, centralized stores govern the storage and processing of Big Data. Due to regulation (e.g. GDPR) and increasing awareness of people about data sensitivity, a paradigm shift towards decentralization is imminent. This allows people to control their own personal data, by guarding all public and private data they or others create about them in a vault, and selectively granting access to people and organizations of their choice. Future uses of Big Data are thus bound to shift from a small number of large datasets to a large number of small datasets. As such, the fundamental assumptions on which the current approaches are built to deal with the characteristics of Big Data (Volume, Variety & Velocity) are no longer valid, i.e. volume cannot be tackled by centralizing data in a single location, high velocity data streams cannot be sent to a centralized data center and the variety problem cannot be resolved by imposing a single data format.

FRACTION supports the shift to a decentralized approach by leveraging Semantic Web technologies. It will investigate 1) algorithms that autonomously distribute the analytics across the decentralized network, while hiding its complexity to the user, 2) decentralized and user-friendly data access control policies, and 3) methods to exploit the heterogeneity of the decentralized network to improve scalability and performance of the analytics.

FRACTION is a fundamental BOF & FWO junior research project of Ghent University IDLab.

Digital health behavior change interventions are being developed and

tested at a growing pace. Indeed, digital communication technologies

offer a vast array of opportunities to facilitate and support behavior change at a large scale including large uptake, constant availability and computational capabilities for data analysis. Compared to real-life personal coaches, they can be cost-effective and more scalable. However, they fail to retain engagement and show limited long-term effects due to lack of personalization of coaching interactions, an undisputed asset of real-life coaching.

IMPERIO aims to develop and evaluate an effective smoking cessation program, delivered by a conversational agent or ‘chatbot’ serving as virtual coach that guides the user through a personalized & contextualized coaching program. Therefore, a strong research focus lies on the development of a trustworthy and effective virtual coach, particularly in terms of personalized conversations. Through self-learning algorithms, the chatbot coach will determine the optimal point to engage with the user. It will automatically construct personalized conversation flows to achieve the optimal coaching strategy tailored to the context, profile, personality traits, behavior & motivation of the user. Evaluation of the effectiveness of the system for smoking cessation is the final outcome.

IMPERIO is a fundamental FWO senior research project of Ghent University IDLab & MICT. 

The internet has become part of our daily life. Nevertheless, we often worry about the lack of transparency companies show when they collect our data. According to a recent study conducted by Flanders’ research center imec, 67% of Flanders’ citizens feel the same. Modern-day companies built their own way of storing and collecting their users’ data. In these so-called ‘data silos’ each company protects its data from others, giving them a competitive advantage. The alternative proposed the SOLID project is a decentralized way of storing data, a personal data pod. If, for example, you migrate to another service, your data pod will move along with you. In that way, you can keep control over your own data and decide what you share, and with whom. Data pods are best seen as a standardized way of storing personal data, just like JPEG files are for pictures, or PDFs are for documents. As a result, the playing field between small and big developers becomes leveled, since everyone uses the same way of handling data.

To support this development, the government of Flanders provided a major influx of financial support for research on Solid. Flanders’ Minister of Economy, Innovation, Work, Social Economy and Agriculture, Hilde Crevits, earmarked EUR 7 million for the first two years of a study on how Solid might be implemented practically. The study, a research project called SolidLab Vlaanderen, will be conducted by a consortium of three of Flanders’ universities (UGent, KU Leuven and VUB), and led by strategic research center imec. 

The goal of ADAM is to design AI methods that are able to predict how a particular patient will react to a particular treatment. This makes it possible to tailor the treatment to each patient. ADAM will focus on two pathologies: heart diseases (coronary artery disease) & lung diseases (lung cancer). 

For this ADAM will use three forms of input. First, the structured medical records of the patients will be used. Second, ADAM will develop technology to unlock the patient data that is hidden in clinical notes (text) according to international standards. Third, ADAM will also collect data outside the hospital. Patients, who are willing to participate, will wear wearables (e.g. smart wristwatches) to measure heart rate, oxygen saturation, etc. Instead of snapshots (the doctor's check), data can be recorded continuously and give a better real-life picture. A computer will structure this health data in an international code to be able to process this data in real time. In this way, the doctor can remotely monitor the patient's health and proactively schedule an appointment with the patient if necessary.

This is done with respect for patient privacy and in accordance with data protection law and clinical trial law.

ADAM is a VLAIO O&O Project. It is a collaboration between the RADAR innovation and learning center of AZ Delta & IDLab. 

Imagine a world where chronic diseases could be targeted directly by electrical signals within and through the body, targeting directly the bodily processes while reducing side effects, and improving outcomes.

By capturing and analyzing a person’s bioelectronic signals, these technologies can detect disease and, through therapeutic devices, intervene to address the underlying causes.

IMEC are innovating to make this a reality by creating solutions for Bio-Electronic Medicine utilizing nano-technology to create technologies for advanced power reduction, miniaturization, encapsulation, increased monitoring capabilities and closing the loop for specific medical applications.

As IDLab.PreDiCT we collaborate on this research with contributing machine learning technologies that can interpret the bioelectronic signals to personalize the interventions (target neuromodulation) or quickly optimize the device to the person.

More than 50% of Intensive Care Unit (ICU) patients have a serious infection and 70% of ICU patients are treated with at least one antibiotic. However, antibiotics are underdosed in up to 40% of them. This can lead to therapy failure, longer hospital stays and higher mortality. Moreover, this leads to long-term exposure to (possibly more powerful) antibiotics, which in turn promotes antibiotic resistance. The pharmacokinetics of antibiotics in these critically ill patients differ greatly from healthy individuals and are difficult to predict. It is therefore desirable to measure the antibiotic concentration in the blood. However, these blood tests are not routinely available, very labor intensive, and therefore expensive. HEROI2C aims to develop models that accurately predict the antibiotic concentration in the blood of a patient, using artificial intelligence, for the most commonly used antibiotics. Based on these predictions, dosing for the individual patient can be optimized by the doctors. 

HEROI2C is collaboration between the ICU of UZ Ghent and IDLab of UGent-imec

The goal of these projects is to gain more insight into the context, profile and lifestyle of people in order to identify triggers and behaviors linked to stress, depression and migraines. Identifying these triggers allows the people to make adaptations to their lifestyle to improve their health.  

We use machine learning algorithm to assess the activities (e.g. sedentary, walking, running, lying, commuting), sleep (e.g. duration, quality) and emotional state of the user (e.g. stress) based on physiological collected data through wearables, i.e. the imec Chillband+ and Empatica E4. 

These projects are a collaboration between UZ Ghent, IDLab Ghent University-imec, and imec-WHS. 

The Flemish Ministry of Economy, Science and Innovation has made 30 million euros available to get Flanders to the head of the pack on AI. In the coming years, this AI-impulse program will be centered around three main pillars: Strategic basic research, Technology transfer and industrial applications, Supporting activities (awareness, training, ethics…).

Specifically within this initiative, Prof. Van Hoecke and Prof. Ongenae are investigating novel methods towards the realization of context-aware machine learning. 

The EMILIO Project enhances self-reliance and counteracts social isolation of elderly clients who live in an assisted facility.

it will manage a comprehensive set of webservices, supporting various use cases to increase their Comfort, Vitality and Safety. Web services are delivered in an intuitive way, fit for use by the target audience. To this end, an IoT (Internet of Things) infrastructure is deployed in the premise that observes the client’s Daily Activities using UWB and interacts vocally with the client when needed.

Within EMILIO the Predict team researches the design of machine learning models and LLMs for the detection of activities of daily living. 

EMILIO is an AAL (Ambient-Assisted Living) project, funded by the EU and VLAIO. The consortium consists of IDLab, MagicView, ePoint, Vulpia, INRCA, Solving Team, ICT Factory, ERDMANN Solutions, ISS and UNITBV. 

Process engineers interpret a high volume of data from sensors to steer chemical processes. However, before acquiring the necessary experience, they need to supervise actual production environments for years. This makes finding the right people a real bottleneck when scaling up production plants.

Machine learning (ML) could drastically increase efficiency by supporting less experienced process operators and more experienced engineers. The former could get actionable insights and expertise, and the latter would have more time to focus on crucial tasks. This will be done by creating smart hybrid AI solutions that learn from data (i.e., historical examples) and knowledge elicited from highly experienced engineers.

These are some of the challenges that should be solved before ML can be adopted to optimize chemical processes. Today’s ML models are purely data-driven; they cannot take into account the expertise and know-how of the chemical process engineers. Moreover, black-box models are hard to understand whereas process engineers need to be able to trust the predictions. Only by understanding why these predictions were made can they take the appropriate control actions knowing they are safe. The control of a chemical process needs to be improved by co-optimizing productivity and safety rather than giving precedence to one over the other. Finally, feedback is needed to continuously tune the ML to process engineers’ ever-changing experience and knowledge. The ML needs to be able to deal with conflicting feedback.

With these challenges in mind, the CHAI consortium will design explainable hybrid AI algorithms. These will incorporate expert knowledge into the machine learning. As a result, data streams will automatically be translated into contextualized insights, outcome predictions, and suggested control actions. The insights, predictions, and suggested actions will be brought to both operators and experienced process engineers through contextualized and dynamic visualization in responsive dashboards. These will be enriched with hybrid AI insights and explanations of how the AI arrived at its conclusions. A feedback loop will allow the operators and experts to optimize the hybrid AI decision-making intuitively and continuously. This way, improvements are made in the standardization of processes, the amount of expert attention needed for low-level routine monitoring, and the speed and quality of the process outcome.

The CHAI project is an imec.ICON project and the consortium consists of P&G, Allnex, Dotdash and IDLab UAntwerp & Ghent University-imec. 

With this project, Volvo Car Gent aims to transform its paint shop by leveraging on all the data that can be gathered from the machines used in the painting process and combining it with the latest developments in data analytics through partnership with Imec. Together, they will develop, train and implement new artificial intelligence based (AI) algorithms to enable Volvo Car Gent to predict possible downtime of its machines, quality issues during the painting process and optimize the maintenance schedule. This new cloud based IoT ecosystem will provide Volvo Car Gent unique insights into its operations, both on an operational and business perspective and will lead to a significant cost reduction of the production process.

This project is funded by VLAIO. 

HVAC & building management systems are very complex and often contain configuration errors. There is often little budget for extensive testing during start-up phase nor for operational monitoring and optimization. Current systems also contain no personalization, nor incorporation of building dweller preferences/needs. 

The goal of this project is: 

• to bring all relevant building data (e.g. BIM models, environmental sensor data, user preferences & feedback, etc.) together in a seamless fashion through semantic technologies. 

• apply and combine data-driven and knowledge-driven AI techniques to extract as much knowledge as possible from the data to provide insights for facility management and dwellers and 

• optimize building management combining generic objectives (e.g. general comfort & energy efficiency) and individual dweller objectives (e.g. personal preferences regarding temperature)

This project is a collaboration between IDLab Ghent University-imec, IDLab UAntwerp-imec, and imec. 

The OPAL project wants to realize an operational efficiency increase of an offshore windfarm by enlarging the possible working time window for maintenance, by reducing waste costs of not executed or failed transfers of goods and personnel and by optimizing the overall complex planning process. We also want to increase predictability and scalability by reducing dependency on human expertise.

The goal is to investigate and de-risk the necessary components for the next generation operational support systems, by PREDICTING ACCESSIBILITY for offshore assets taking context, weather, vessel characteristics, captain profile, park maps, boat landing locations and vessel routes into account. Using these accessibility predictions, the COMPLEX PLANNING PROCESS of offshore maintenance operations will be optimized, and transport vessels and work force will become smart and designed for more predictable outcome.

OPAL is a VLAIO Cluster O&O project. It is a collaboration between e-BO Enterprises, dotOcean, Geo.xyz, Multi and Ghent University. 

The needs of the care industry are quickly changing. An aging population is causing an increase in chronic diseases and longer hospitalization times. This requires more complex care, more beds in residential care centers and more employees. Geriatrics records the longest average hospital stay. Elderly people who are eligible for discharge cannot be sent home due to the lack of a social support system. To minimize the effects of this lack on the health care system, care delivery should become more transmural and be offered at home. Residential care can then be reserved for those with intensive care needs.

Today, personal alarm systems and monitoring devices need to be assessed and handled by caregivers, who are often unable to quickly assess the priority and validity of the alarm due to a lack of contextual and sensor data. As a result, one third of all interventions are false alarms. The nature of the required intervention is also hard to estimate, leading to inefficient dispatching, care workflows and use of resources.

PROTEGO aims to increase the efficiency of care organizations by optimizing the handling of acute, unplanned events. PROTEGO will achieve the following innovations:

• Offer caregivers and call centers the required context by accurate assessment and prioritization of incoming unplanned events through AI services based on sensor and profile information;

• proactive alarms from home by detecting deviations in lifestyle patterns;

•  Accurately assess, manage and dispatch unplanned requests to caregivers through adaptive and interactive care workflows based on recommenders;

• comprehensive patient and call overviews in hyperpersonalized dynamic dashboards 

The PROTEGO consortium consists of Televic Healthcare, Amaron, ML2Grow, Z-Plus, and IDLab Ghent University-imec. 

As the climate changes, water availability comes under stress, making efficient water provision all the more critical. However, undiscovered water grid leaks cause over 60 million m3 of drinking water to be lost every year in Flanders alone. To help turn the tide, SmartWaterGrid will create hybrid digital twins of real-time water flow and pressure. This will be done using IoT sensors augmented with GIS data, hydraulic models, expert knowledge and human feedback to accelerate and fine-tune the process of localizing leaks.

The SmartWaterGrid consortium consists of De Watergroep, Hydroscan, Itineris, Aloxy, IDLab-Ghent University-imec & MOSAIC- UAntwerp - imec 

Each year, 1 million babies die as a result of preterm birth complications. Yet, many of them could be saved — provided timely and well-chosen medical interventions. What if we could help caregivers in their treatments, by predicting whether and when patients will deliver preterm?

By combining clinical observations and freely written doctors’ notes, the machine learning models designed by IDLab and Ghent University Hospital predict the chance of giving birth within the next seven days. This way, the therapy can be adjusted accordingly to optimize outcome for mother and child.

Machine learning systems can detect anomalies in environments such as IT and telecommunication networks. But to teach them how to do that, a lot of data is required. Consequently, machine learning systems for anomaly detection are often trained to be accurate in only one environment or situation. Because anomalies manifest differently in different contexts, false positives often result when the solution is deployed in different situations and configurations. To overcome this hurdle, RADIANCE will extend machine learning algorithms, enabling them to detect anomalies in real time under resource constraints and adapt to changing contexts. This will cut training time and reduce the need for human involvement.

The RADIANCE Consortium consists of Barco, Skyline Communications, ML6, DistriNet-KU Leuven-imec, IDLab-Ghent University-imec.

This was a bilateral project between Harmoney and IDlab Ghent University - imec.

Harmoney provides a next-generation digital platform of solutions that automates and streamlines complex on-boarding and end-to-end, continuous compliance processes while delivering frictionless user experiences. By maintaining a complete, high-quality, reusable financial passport for each individual and corporate customer, Harmoney remotely orchestrates all interactions between the front office (relationship management), the back office (compliance) and end users. Harmoney is the end-to-end customer due diligence partner of banking and insurance organisations. 

IDLab investigated how the seamless integration of questionnaire information could be better supported by Semantic Web Technologies. 

Mining sensor data can give companies valuable insights into their assets and activities. But as technology advances, analyzing and visualizing the data becomes increasingly time- and resource-intensive. The DyVerSIFy project developed software components and methodologies in the domains of dynamic visualization, adaptive anomaly detection and scalability to drive dynamic, adaptive and scalable sensor analytics.

With the collected data, you can monitor the states of equipment, environmental conditions, infrastructure and more. This allows companies to detect errors and anomalies to improve the efficacy of maintenance and design. However, state-of-the-art data visualization and analytics systems need to be manually configured when new sensors, visualizations or algorithms are added. Even more, if system configuration is suboptimal, this leads to an inability to process and analyze the ever-growing amount of sensor data.

The DyVerSIFy project enabled scalable sensor analytics that respond and adapt to changing needs. To that aim, it realized:

1) Dynamic visualization of data, which automatically responds to users’ needs to present relevant data;

2) Adaptive anomaly detection (AD) and root cause analysis (RCA), bringing machine learning and semantics together and introducing automated tuning based on implicit user feedback;

3) Scalable services and processing, to enable new sensors or changes to sensor systems to be added dynamically without manual intervention.

The DyVerSIFy consortium consisted of Renson, Televic Rail, Cumul.io and IDLab Ghent University-imec.

Anxiety disorders lead to feelings of incapacitating fear and represent a cost of EUR 74.4 billion in Europe alone. Exposure therapy involves gradually confronting the patient with the object or situation of their fear. As such, exposure therapy is difficult to manage outside of holistic clinical settings, resulting in a treatment rate of 33%. Virtual reality exposure therapy (VRET) is a safe, controlled, cost-effective and potent alternative. Using a computer-generated environment, VRET immerses the patient gradually into an emotional experience. 

The PATRONUS consortium offered an effective, lightweight and personalized blended care solution consisting of IT tools that support the therapist and patient throughout the anxiety treatment process. It is capable of:

• personalized and objective anxiety assessment based on physiological data from wearables;

• a comprehensive therapist dashboard allowing seamless patient follow-up and (VR) therapy management;

• a personalized and low-cost VR experience that allows for automatic adaptation to the profile of the patient;

• individualized longitudinal follow-up and coaching through VR homework exercises and a mobile application. 

The consortium consisted of The Human Link, PreviewLabs, Van Roey, Bazookas, MICT Ghent University-imec, IDLab Ghent University-imec, imec-WHS.

DiSSeCt was set up because of a very clear demand of various companies to research distributed semantic algorithms, software techniques and solutions enabling the exchange of large streams of data and knowledge between different services to intelligently compose reliable, dynamic and secure workflows that provide personalized and context-aware solutions to end-users. As such DiSSeCt performed the necessary research into the required algorithms, software solutions and end-user guidelines to make the leap from “Big Data” to “Big Service” solutions. Key research challenges of DiSSeCt were:

Study, design and inception of scalable, performant & robust algorithms, techniques, models, methods and guidelines to

• expose and compose raw data semantically as intelligent services,

• (semi-)automatically generate dynamic workflows based on these services,

• cope with the rapid & timely processing of high volumes of data (stream reasoning) by these services and workflows, and

• enable self-learning capabilities for these services and workflows.

Ensuring that these research results are:

• secure and trustworthy systems in terms of both data processing and data access in order to guarantee reliable outcomes for the end-users, and

• easily manageable by end-user developers.

This resulted in two research prototypes in smart health and smart mobility.

The consortium consisted of IDLab Ghent University-imec, SMIT VUB-imec, DistriNet KU Leuven-imec. 

This was a bilateral project between Amaron and IDLab Ghent University-imec.

In this project, IDLab investigated a machine learning algorithm that took as input the non-clinical profile of a person (e.g. elderly) and outputs a suggestion on which care support services this person needs, e.g. help with cleaning, meal service, etc. 

This was a bilateral project between IDLab Ghent University-imec and Klaas Vandevyvere, rheumatologist. 

Rheumatologists in Flanders, Belgium have to cope with an ever-increasing workload. This is due to the growing number of patients per expert, as a result of an aging population and a decrease in active rheumatologists, as the older generation is retiring and new doctors specializing in the field are scarce. The rheumatologist had strong beliefs that a digital triage system, supporting General Practitioners (GPs) in their diagnosis, could remove some of the burden from the medical specialist, if able to reduce the number of misguided patient referrals. 

IDLab designed a decision support system, which allowed to:

• Give guidance to GPs on which questions should be asked to the patient to gather information on his/her symptoms, and

• Give a suggestion to a GP on whether the patient should be referred to a rheumatologist (with high priority). 

The collection and analysis of athlete data in cycling offers a lot of potential to improve cycling experience, training regimes and cycling races strategies. Popular cycling apps gather data, but report and analyze it offline. That is why CONAMO enabled the real-time monitoring of athletes. The project realized a data-driven cycling experience through the development of a mobile network layer for data transmission and an analysis layer that interprets the data gathered in real-time – for instant updates on position and performance that can be vital to training approaches and strategic decision-making during events.

The topic of athlete data in cycling is especially relevant in Flanders, the birthplace of many cycling innovations. Real-time data doesn’t just enable cyclists to get an instant overview of their performance for improved training and strategic decision-making; it also makes cycling a social experience, opening up areas of a market in which novelty and added value are key differentiators.

By introducing new technologies in the areas of long-range networks and data analysis, CONAMO addresses several innovation goals, including:

- reliable long-range sensor connectivity in high-density networks;

- a semantic reasoning platform that analyses cyclist data to implement individualized and adaptive training models;

- increased user experience during training and collective cycling events and the identification of business opportunities for cyclists, fans and media professionals.

The CONAMO consortium consisted of Energy Lab, Rombit, VRT, IDLab Ghent University-imec, IDLab UAntwerpen-imec, MICT Ghent University-imec, SMIT VUB-imec, imec Living Labs and the Ghent University research  group on "Inspanningsfysiologie en Trainingsleer".

WONDER designed an automated and interactive 24/7 care service with Zora — a humanoid robot, successfully used for entertainment and therapy in about 130 Belgian nursing homes. Its goal is to use Zora to trigger positively charged personal memories in persons with dementia, and thus reassure and calm them. On the one hand, the project enabled to automatically detect behavioral disturbances in nursing home residents, e.g. wandering, yelling, aggressive behavior. On the other hand, it explored how Zora can walk around semi-autonomously and interact personally with different residents, as a useful support tool for nurses and caregivers. 

The WONDER Consortium consisted of ZoraBots, Xetal, WZC Weverbos, WZC De Vijvers, IDLab Ghent University-imec, SMIT VUB-imec. 

SMILE-IT extended and improved upon the state of the art in multi-agent reinforcement learning and network management and implemented and validated a generic, stable, and robust multi-agent reinforcement learning framework, capable of (semi-)automatically managing modern networked systems (e.g., telecommunications networks, smart grids, air traffic routing, traffic control) through software. 

The consortium consisted of AI Lab VUB, SOFT VUB, ELECTA KU Leuven, VITO, IDLab Ghent University-imec, IDLab UAntwerpen-imec.

Companies that want to deliver sophisticated services still struggle to take into account and to learn from all relevant context information. They may have extensive information about their customers including history, behavior and present situation, but cannot always leverage those data to take intelligent decisions. 

Some of the challenges they face are:

• Which context data can influence and improve decisions?

• Is this information readily available or does it require additional processing?

• With fast-changing contexts, can a system be made self-adapting?

• Can we process information offline or in-stream, and how fast?

• What are acceptable processing latencies for service delivery, and how can we minimize them?

• Can our context-based system be scaled to account for future growth?

CAPRADS was set up to develop tools for service providers to make their offer context-aware. The middleware that resulted is self-learning and scalable. It allows fast and valuable decision making and can be customized to support a wide variety of use cases. 

Three use cases were tackled. At JForce, they need to scour context to make lightning-fast decisions about financial transactions, decisions that we can really trust on in a forever-changing context. Luciad, in contrast, needs to take into account huge and growing amounts of geographical data to offer context-aware information to users, in some cases requiring fast and ultraprecise responses, for example for emergency response systems. And Televic Education wanted to customize and enrich its customers’ learning experience by taking into account their learning context.

The consortium consisted of JForce, Luciad, Televic Education, DistriNet KU Leuven-imec, IDLab Ghent University-imec. 

AORTA investigated ways hospitals can achieve efficiencies through the dynamic allocation of tasks and resources to logistics and nursing staff. The project seeks to support the different logistics processes and integrate them with new systems that continuously adapt to a hospital’s context. AORTA hopes this will, among other things, help reduce waiting times for patients, relieve nursing staff of non care related tasks and make sure consumables are stocked more efficiently. 

As such, AORTA designed a context-aware system to accurately get a view on the current context of the hospital and uses this information to dynamically assign transports to staff. By using AI, AORTA also investigates the reasons why transport delays occur and uses these to further optimize the allocation process. These delays can also be reported to management for follow-up. 

The AORTA consortium consisted of Televic Healthcare, Xperthis, AZ Maria Middelares, Ziekenhuis Netwerk Antwerpen (ZNA), Mintlab KU Leuven, ITEC KU Leuven, and IDLab Ghent University-imec. 

The environmental sensing project was a bilateral project between CNH Industrial and IDLab Ghent University-imec. 

It researched the design of image processing algorithms to enable the design of autonomous driving agriculture vehicles. 

ORCA was a bilateral project between IDLab Ghent University-imec and Televic Healthcare.

It researched the design of a mobile, person-oriented and context-aware nurse call system. 

Both broadcasting systems and conferencing systems see a demand for high-quality blended data overlays within end-user video streams. Hence the need for an automatic play-out system which allows to capture and visualize events and which is able to integrate additional data in real time, only requiring minor guidance by a 'director' (be it the DJ or a conference chairman). Not only do we want to control play-out systems, we will investigate the control of physical objects in a studio by automated triggers, such as automatic adjusting of live cameras and microphone arrays based on audio, video and contextual information.

A robust semantic understanding of the stream, based on the fusion of contextual information and the analysis of media content is mandatory. In the end, we designed a semantic play-out decision-support system centralized around the currently active radio- show or conference. This project will contribute to 1) the creation of a real new medium, 2) efficient and cost-reducing post-production, and 3) the opening up of TV Graphics to the ecosystem of Flemish web agencies. 

The consortium consisted of WMMa, Televic Conference, Small Town Heroes, MIX iMinds, IDLab Ghent University-imec, Mintlab KU Leuven. 

For elderly people fall incidents are often a life-changing event that might lead to degradation or even loss of autonomy. More than half of the elderly living in a nursing home and about one third of the elderly living at home fall at least once a year, denoting the high prevalence of falling. Of those who fall, 10 to 15% suffer severe injuries. The lack of timely aid can lead to further complications. Although not all fall incidents lead to physical injuries, psychological consequences are equally important. Due to the high impact of falling, both fall prevention and reliable fall detecting are necessities. 

The FallRisk project designed non-stigmatizing, holistic services in view of the automated follow up of fall risk and the multi-sensor and contextual detection of fall incidents. The project developed a multi-sensor approach in the home network environment, exploring the active inclusion of microphone arrays and the TV set. Optimum design of a multi-sensor system focuses on the assurance that every real fall incident can be detected (100% sensitivity), while at the same time capturing valuable context information.

Secondly a context-aware and social-aware selection algorithm will was developed to be able to support the dynamic and optimum forwarding of FallRisk events or alarms towards the (in)formal caregiver network. This back-end system will handle the event forwarding in view of the context of the event and in view of the social-network of the elderly.

The consortium consisted of Televic Healthcare, Verhaert, TP Vision Belgium, COMmeto, WGK Limburg, EDM UHasselt, SMIT VUB-imec, IDLab Ghent University-imec, STADIUS KU Leuven, AdvISe KU Leuven. 

The O’CareCloudS team pursued the creation of a platform that can capture, aggregate and enrich loads of data about the medical condition and (real-time) assistance requirements of people in a home care context, and share those insights in an intelligent way with formal and informal caregivers alike. The team’s ultimate objective: allowing people to live as autonomously as possible, for as long as possible, while offloading – and maximizing the efficiency of – caregivers. 

This consortium consisted of Televic Healthcare, Boone NV, Familiehulp VZW, OCMW Gent - RVT De Vijvers, OCMW Kortrijk, Philips, Telecom IT, IDLab Ghent University-imec, DistriNet KU Leuven-imec, SMIT VUB-imec, MintLab KU Leuven. 

The Astute Project aims at the development of an advanced and innovative pro-active HMI interface and reasoning engine system for improving the way the human being deals with complex and huge information quantities, during real operations that without any type of assistance would saturate his performance and decision-making capabilities in different operative conditions and contexts. To reach this goal, semantic technologies such as ontologies, DL-based and rule-based Reasoners are used. ASTUTE will enable users of embedded systems to perform fewer errors and improve their working efficiency, in particular when taking decisions in complex and critical situations. This not only has a huge impact on the users, but also on the public when considering safety critical systems (e.g. industrial plants, airplanes, road transportation), thus also addressing key societal needs (e.g. less accidents, but also less pollution). In all addressed domains, by focusing on the user state and context Astute intends to extend the level of task automation or even introduce task automation in applications today only relying on human control. 

In this project IDLab Ghent University-imec was subcontracted by SIRRIS to carry out research into the semantic reasoning engine. 

Accio stands for Ambient aware provisioning of Continuous Care for Intra-muros Organizations. This project investigated how the newest insights and developments in ICT can support the continuous care provision process within intra-muros care organizations. Ontologies and rule-based algorithms were researched and developed for two real life contexts, nursing homes and hospitals, in view of optimizing existing care logistics tools such as nurse call systems.

A participatory ontology engineering methodology was investigated and developed within the project which involves the stakeholders, e.g., nurses, patients and doctors, in every step of the ontology development cycle without overburdening them. The ACCIO Continuous Care Ontology was developed using this methodology. Finally, a demonstrator was built using this ontology. The demonstrator consisted of an intelligent nurse call system, which finds the most appropriate caregivers to assign to a call based on the staff & patient profiles and the available context information. A mobile application was also developed on which caregivers can receive, assess and forward nurse calls. 

The consortium consisted of Televic Healthcare, Dienstencentrum Gidts, Dominiek Savio Instituut, Boone, In-HAM vzw,  IDLab Ghent University - imec, SMIT VUB-imec, MintLab KU Leuven. 

Cosara is a computerized system for the surveillance and alerting of nosocomial infections, antimicrobial resistance and antibiotic consumption in the Intensive Care Unit. It was designed by IDLab Ghent University - imec in collaboration with Ghent University Hospital. COSARA is still daily used in the ICU of Ghent University Hospital.

Cosara provides a dashboard, containing the patient's infections and antibiotic therapies along with basic inflammatory parameters. This dashboard, along with an integrated thorax viewer and microbiology-overview, assist the physician in his daily workflow. Some of its unique features include: 

• Providing an overview of all the patient´s infections 

• Easy visible indication of which antibiotics are used to treat the infections 

• Easy visualization of the responsible microorganisms for a specific infection 

• Very fast access to relevant thorax images 

• A quick and comprehensive overview of all antibiotics and microbiology information 

• A simple tool to link infections, microorganisms and antibiotics 

• ck access to detailed information regarding the patient´s admission and comorbidity information 

• Desktop integration for patient selection and workflow optimization

The FP6 European project InteGRail project created a holistic, coherent information system, integrating the major railway sub-systems, in order to achieve higher levels of performance of the railway system in terms of capacity, average speed and punctuality, safety and the optimized usage of resources. Building on results achieved by previous projects, InteGRail proposed new intelligent procedures and contributed to the definition of new standards, in accordance with EC directives and TSI’s.

IDLab Ghent University-imec was active in the subproject Intelligent Monitoring and has contributed to the definition of the ontology-based Distributed Intelligent Monitoring Architecture (IMON), development of ontology-based data conversion and information integration adapters and modelling the ontology for several demonstrators.