{"id":1828,"date":"2026-09-15T09:24:58","date_gmt":"2026-09-15T09:24:58","guid":{"rendered":"https:\/\/prg.inf.unibe.ch\/?page_id=1828"},"modified":"2026-09-15T09:24:58","modified_gmt":"2026-09-15T09:24:58","slug":"thesis-fusing-icu-time-series-with-tabpfn-based-clinical-context","status":"publish","type":"page","link":"https:\/\/prg.inf.unibe.ch\/index.php\/education\/thesis-fusing-icu-time-series-with-tabpfn-based-clinical-context\/","title":{"rendered":"thesis-Fusing ICU Time Series with TabPFN-Based Clinical Context"},"content":{"rendered":"\n
<\/div>\n\n\n\n
<\/div>
\n
<\/div>\n

Fusing ICU Time Series with TabPFN-Based Clinical Context<\/h1>\n\n\n\n

Co-Supervised by:<\/strong> Nico Bigler<\/p>\n\n\n\n

If you are interested in this topic or have further questions, do not hesitate to contact daniel.bigler@students.unibe.ch<\/u><\/a>.<\/p>\n\n\n\n

Background \/ Context<\/h3>\n\n\n\n

ICU risk prediction can combine evolving physiology with tabular clinical context. This project investigates whether a TabPFN-derived representation adds predictive value beyond a conventional context branch. HiRID contains more than 33,000 ICU admissions and longitudinal physiological, laboratory and treatment data. Its static context is limited: age and sex can be supplemented with explicitly defined early measurements, but a comprehensive comorbidity profile should not be assumed. The project builds on the HiRID-ICU-Benchmark and requires credentialed PhysioNet access.<\/p>\n\n\n\n

Research Question(s) \/ Goals<\/h3>\n\n\n\n

The research aims to evaluate the incremental value of TabPFN-based context by:<\/p>\n\n\n\n