Large scale statistical machine learning for the type 1 diabetes management
Interview with Tatjana Pavlenko
Would you like to describe the project in short with a few sentences?
”The proposed PhD project centers around the theory of large-scale statistical machine learning and its applications to the complex data sets arising in Type 1 Diabetes (T1D) management.
Methodological research of this project targets statistical problems in the area of Artificial Intelligence (AI) such as large-scale network modeling and graph structure learning, along with a corresponding rigorous efficiency analysis. Based on methodological results, a user-friendly, decision-support software will be developed. Such software aids the clinician to operate with a wide range of data and compile it into a summarized recommendation list highlighting the most critical outputs that need to be considered when evaluating potential treatment regimens and identifying T1D patients at high risk prior to the onset of complications.”
Would you like to tell us a little about the cross-disciplinary aspects of the project?
”Despite the rapid development of science and technology in healthcare, T1D remains an incurable life-long decease. Statistical machine learning has multiple purposes in studying chronic disease such as diabetes, a condition where continuous support and monitoring are required, making it an ideal candidate for a data-driven approach to management. One of our aims in the project is to establish early clinical markers by combining existing clinical data and novel proteomics data in order to identify individuals with the greatest risk to develop complications in order to be able to tailor their treatment so that future complications can be avoided. This is one example of our cross-disciplinary research topics which promise innovative and beneficial solutions for the problems of efficient T1D management related to the impact of AI techniques in healthcare.”
How was the project and collaboration born?
”The applicants have discussed a number of research questions related to applications of statistical machine learning methods to various problems emerging in T1D management. The idea of a cross-disciplinaly, long-term collaboration comes up when we saw the opportunity of a joint PhD student which could work with development of statistical machine learning methodologies especially designed for studying
T1D. The team members include Associate Professor Ronnie Pingel, Department of Statistics, and Associate Professor Daniel Espes, SciLifeLab, Department of Medical Cell
Biology and Department of Medical Sciences, Uppsala University.”
How can UDC be of help in your research?
”Getting a PhD Grant from UDC shows that the research community appreciates and believes in your research ideas, which is very encouraging. We expect that collaboration with Research Fellows within UDC makes it possible to established a cross-disciplinary network where new efficient statistical learning methodologies will be adopted and further developed to provide useful insights into the practical aspects of T1D management. We believe that sharing experience and ideas with UDC researchers
will provide positive synergies between theoretical and practical aspects of the PhD project, thereby producing results with clear benefits for society.”
What can UDC do to make the PhD students feel like they belong to the centre?
”We believe that belonging to the centre will come from opportunities to share ideas and have scientific discussions with both senior researchers and PhD student working in the intersection between diabetes and several related fields.”