KiMeKo (KI-Med-Kollaborationsplattform) is a publically funded collaborative research project that develops a sustainable AI-Med ecosystem for AI-based medical device development. The project runs from July 2024 to December 2027 and joins seven Northern German research institutions. KiMeKo addresses the complete development trajectory, from concept and data acquisition to validation, regulatory evidence generation, and approval-oriented documentation. The project contributes a practical toolchain and platform capabilities for non-experts and experts, including structured innovation support, uncertaintyaware sensor-data fusion, hybrid expert-system modeling, and workflow-guided data acquisition and anonymization. This paper summarizes project objectives, expected outputs, relevance to IEEE COMPSAC 2026 themes, and current progress. In particular, KiMeKo aligns with Applied AI and Smart & Connected Health by combining AI engineering, privacy-conscious data processing, and regulation-aware medical software development.

In contemporary society people constantly are facing situations that influence appearance of serious diseases. For the development of intelligent decision support systems and services in medical and health domains, it is necessary to collect huge amount of patients’ complex data. Patient’s multimodal data must be properly prepared for intelligent processing and obtained results should be presented in a friendly way to the physicians/caregivers to recommend tailored actions that will improve patients’ quality of life. Advanced artificial intelligence approaches like machine/deep learning, federated learning, explainable artificial intelligence open new paths for more quality use of medical and health data in future. In this paper, we will focus on presentation of a part of a novel Open AI Architecture for cancer patients that is devoted to intelligent medical data management. Essential activities are data collection, proper design and preparation of data to be used for training machine learning predictive models. Another key aspect is oriented towards intelligent interpretation and visualisation of results about patient’s quality of life obtained from machine learning models. The Architecture has been developed as a part of complex project in which 15 institutions from 8 European countries have been participated.

We propose a causal machine learning inference pipeline that combines a given predictive machine learning model with analytical estimations of average treatment effects. It enables to utilize any predictive model for causal inference, which makes it easy to adapt the approach to existing systems. By first estimating the average treatment effect of an intervention on predictors instead of the outcome variable, a causal relationship between an intervention and a wide range of variables is determined. Next, artificial samples are created that are evaluated using the given predictive model to link interventions and outcomes and also allows inferring measurements of uncertainty. Finally, simulations using again the given predictive model are performed to compute measurements of confidence and that allow to compare – according to the given predictive model – the effect of specific treatments. We furthermore demonstrate how this inference engine can be adapted to a privacy-preserving federated learning environment where training data is horizontally distributed across multiple datasets without compromising on our approach’s accuracy. The approach has been evaluated on a use case with a predictive model for the quality of life score of cancer patients, to determine medical interventions to improve their individual quality of life score.

Deploying robots with the ability to autonomously navigate in the environment of a clinic necessitates carefully designing their verbal and non-verbal behaviour in order to avoid irritating, worrying or even endangering users, bystanders and passersby. We describe a user-centred approach used to design, implement and evaluate an interaction concept for mobile robotic assistants for a clinic environment, resulting in two robotic assistant prototypes and a generalized multimodal interaction framework applicable to a variety of autonomously navigating robotic agents. We present the steps of the design process and intermediate findings, the final system components and experiment results showing good acceptance of the interaction concept. A clip of the two robots in action can be found under https://short.dfki.de/ro-man.

Quality of life (QoL) is one of the major issues for cancer patients. With the advent of medical databases containing large amounts of relevant QoL information it becomes possible to train predictive QoL models by machine learning (ML) techniques. However, the training of predictive QoL models poses several challenges mostly due to data privacy concerns and missing values in patient data. In this paper, we analyze several classification and regression ML models predicting QoL indicators for breast and prostate cancer patients. Two different approaches are employed for imputing missing values. The examined ML models are trained on datasets formed from two databases containing a large number of anonymized medical records of cancer patients from Sweden. Two learning scenarios are considered: centralized and federated learning. In the centralized learning scenario all patient data coming from different data sources is collected at a central location prior to model training. On the other hand, federated learning enables collective training of machine learning models without data sharing. The results of our experimental evaluation show that the predictive power of federated models is comparable to that of centrally trained models for short-term QoL predictions, whereas for long-term periods centralized models provide more accurate QoL predictions.

In computer programming education, learning to program tangible objects has become a common way to introduce programming to young students. In an effort to address this intervention, scientific research has been done on the effectiveness of using tangible hardware platforms such as robots and wearable products to teach basic programming concepts to children. However, there is a lack of research on how young students' attitudes and programming skills are influenced over time, when they learn to program tangible objects and make them smart. In this paper, we investigate the impacts of using a tangible everyday object and making it smart on young female students' attitudes towards programming and the acquisition of basic programming skills. During a 4-day non-formal programming workshop with 12 6th grade students, they were introduced to basic programming concepts, and learned how to apply them to turn a houseplant into a smart object. In a pilot study, we employed a block-based programming environment and analyzed the students' trajectories of attitudes towards programming and performance based on repeated open-ended qualitative questionnaires and programming questions throughout the workshop. The results show that all students had high confidence regarding programming skills, regardless of creating smart objects. Furthermore, it indicates that experienced students highly valued the programming of tangible everyday objects compared with inexperienced students. The findings of this work contribute to our understanding of how making tangible everyday objects smart can support the development of a positive attitude and keep up of interest throughout a programming workshop among girls.

In light of the complexity of introductory programming for young learners, visual programming has become more and more popular. In particular, block-based educational programming systems have emerged as an area of active research. This paper introduces an educational block-based programming application, enabling young learners to learn and make programs in the context of smart homes. In this application, smart objects have a set of primitive behaviors which can be integrated in the general features of programming languages like variables, conditionals, loops, and functions. The programming language is shown in a graphical interface to enable young students to program with the application. The development and implementation of this application, along with helping features for the students are described. In a pilot study with 20 7th grade students, the application's effectiveness and ease of use are evaluated. The results show that students can fairly solve programming problems and make real programs in the context of smart homes. Feedback of the learners is presented and discussed.

An overwhelming amount of documents is produced and changed every day in most areas of our everyday life, such as, for instance, business, education, research or administration. The documents are seldom isolated artifacts but are related to other documents. Therefore changing one document possibly requires adaptations to other documents.