Established machine learning matches tabular foundation models in clinical predictions.

Lawrence A Shaktah¹, Marco Gustav¹, Tim Lenz¹, Junhao Liang¹, Lars Hilgers¹, Zunamys I Carrero¹, Jakob Nikolas Kather^2,3,4,5

¹Else Kroener Fresenius Center for Digital Health, Faculty of Medicine, TUD Dresden University of Technology, Dresden, Germany.

²Else Kroener Fresenius Center for Digital Health, Faculty of Medicine, TUD Dresden University of Technology, Dresden, Germany. kather.jn@tu-dresden.de.

³Department of Medicine I, Faculty of Medicine, TUD Dresden University of Technology, Dresden, Germany. kather.jn@tu-dresden.de.

⁴Medical Oncology, National Center for Tumor Diseases (NCT), University Hospital Heidelberg, Heidelberg, Germany. kather.jn@tu-dresden.de.

⁵Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK. kather.jn@tu-dresden.de.

PubMed DOI

Abstract

Foundation models (FMs) promise to standardise predictive modeling across domains, yet their clinical value for tabular data remains unproven. To test this, we performed a large, fully reproducible benchmark of TabPFN, a leading FM for tabular prediction, against twelve established machine learning (ML) methods across twelve binary clinical tasks. Cohorts spanned 788 - 139,528 patients across dive…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.