Prosjektkatalog

The SCoDA project is a collaboration between Norway (University of Stavanger, Stavanger University Hospital (SUS), University of Tromsø), South Korea (Seoul National University), and Japan (University of Tsukuba). The project is strongly linked to ongoing research activities in complex data at partner institutions and seeks to establish an educational platform in complex data analysis to enable students and researchers to fully exploit the entire potential of such data. Complex data includes non-linear data (such as directional data and shape data) and High-Dimension Low-Sample Size (HDLSS) data (such as genetic data). For example, data blocks of medical data include medical images, parameterized shapes of human organs, genetic data, and clinical data, all of which play vital roles in the diagnostics of diseases. Such data blocks can be found for example in the ParkWest study (a longitudinal study of patients with Parkinson’s disease) and several studies within dementia at SUS. Conventional statistical methods are designed for linear data while shapes live on curved non-linear spaces, and their properties have been studied for low-dimensional large-sample size situations, the opposite of HDLSS. The surge in data availability leads to an increasing amount of complex data which presents both an opportunity and challenge, as conventional statistical methods are not designed for complex data and often struggle to uncover hidden patterns, correlations, or might lead to false conclusions from these complex datasets. For medical data, the full potential of complex data by applying suitable statistical methods is of utmost importance to unravel the whole data information and to attain a full understanding of serious diseases in order to predict disease progression, and to discover new treatment pathways.

The SCoDa project seeks to overcome the challenges with complex data through several activities including (1) joint course development in the form of thematic online lecture modules in shape analysis, HDLSS statistics, manifold statistics, geometric statistics, and topological data analysis; (2) exchange of students and co-supervising MSc / Ph.D. students with hands-on projects in collaboration with SUS in the field of complex data analysis in their thesis work; (3) arranging of training schools that are accomplished by workshops, and a symposium; (4) seminars with guest lectures and staff exchange; (5) existing master courses will be revised. In addition, (6) the project includes several communication and dissemination activities of project results on different target groups. Thus, the project will acquaint students and researchers with state-of-the-art methods in complex data analysis.

SCoDa will intensify existing research activities and foster new collaborations between Norway, South Korea, and Japan to address the growing demand for advanced data analysis techniques to extract meaningful insights from complex datasets. Ultimately, the project seeks to empower students, researchers, and organizations to make data-driven decisions with clarity and confidence.