Current Project |

XAI-FinCrime

Mon, 01 Sep 2025 00:00:00 +0000

Funding: Innosuisse · September 2025 – August 2028

The project aims to transform financial crime prevention by leveraging machine learning, generative AI, and explainable AI (XAI) to enhance fraud detection and compliance management. It uses ML models for transactional behavior analysis and multimodal LLMs for unstructured data processing, with counterfactual explainers to improve AI decision transparency.

Objectives

Significantly reduce the typical False Positive rate of 95% found in current rule-based industry standards
Improve F1-scores, accuracy, and True Positive rates compared to existing solutions
Enhance user understanding of AI systems to reduce investigation effort per alert
Validate innovations through pilot studies with financial institutions

EXperiMental: Wearable Technology and EXplainable AI for Mental Health and Inclusivity in Schools

Tue, 01 Jul 2025 00:00:00 +0000

Funding: SNF, grant 230189 · July 2025 – December 2028

Mental health is a vital part of our overall well-being, especially during early adolescence, a time of rapid growth and emotional development. The initiative explores how AI, smartphones, and smartwatches can promote mental well-being in schools through collaboration among psychology, sports science, computer science, and ethics experts across Romania, Croatia, and Switzerland.

The research involves two major studies: (1) observing daily routines using digital tools and mental health check-ins, and (2) applying AI methods to identify patterns signaling mood or mental health changes. The aim is creating personalized, understandable suggestions for young people, teachers, and school psychologists.

Objectives

Understand how physical activity, sleep, and phone use relate to adolescent mental health
Design youth-friendly, privacy-protective monitoring methods via mobile devices
Develop personalised suggestions, such as encouraging a walk or improving sleep habits
Create ethically sound guidelines for technology use supporting adolescent mental well-being in schools

XAI-PAC: Towards Explainable and Private Affective Computing

Mon, 01 Jan 2024 00:00:00 +0000

Funding: SNF · January 2024 – January 2028

Wearable devices combined with AI methods can improve lives through affective computing, which tracks factors related to emotional states and stress for remote mental-health management. However, current AI methods are “black-boxed” and require vast sensitive data. The project addresses the need for explainable and privacy-aware affective computing.

Objectives

Develop novel eXplainable AI (XAI) tools for affective-computing methods based on wearable sensor data with multimodal, interactive capabilities
Develop privacy-aware ML methods for wearable sensor data, including privacy-aware personalization and domain-adaptation
Fuse privacy awareness and explainability into a single integrated approach

TRUST-ME: TRUstworthy enhancement of job SaTisfaction and productivity using Micro-sensing in work Environments

Sun, 01 Oct 2023 00:00:00 +0000

Funding: SNF · October 2023 – October 2026

The TRUST-ME initiative addresses employee well-being in modern workplaces by examining the complex relationship between employee productivity, job satisfaction, and well-being. The endeavor combines sensing technology and AI to create personalized monitoring tools, emphasizing multimodal monitoring and modeling of job satisfaction and productivity, as well as secure and private AI for productivity assessment. The approach relies on workplace sensors and privacy-preserving techniques, such as federated learning, to deliver insights while protecting employee autonomy.

SmartCHANGE

Mon, 01 May 2023 00:00:00 +0000

Funding: Horizon Europe · May 2023 – May 2027

The SmartCHANGE project addresses the shift from communicable to non-communicable chronic diseases (NCDs) in developed countries. According to WHO, more than 70% of deaths worldwide are attributed to NCDs, with projected costs rising from 5.5 trillion € (2010) to over 12 trillion € by 2030. Most NCDs share risk factors including obesity, low physical fitness, poor nutrition, and sedentary behavior. The project aims to create trustworthy, AI-based decision-support tools for assessing NCD risk in children and youth.

Objectives

Build accurate machine learning models for predicting lifetime NCD risk (cardiovascular and metabolic disease) in children and youth
Ensure risk-prediction models and AI tools are trustworthy
Develop tools for health professionals and citizens to improve health outcomes
Engage users in requirements elicitation and participatory design
Conduct proof-of-concept studies across four real-world healthcare scenarios in different countries
Develop implementation recommendations and dissemination strategies
Create an exploitation and sustainability plan