MediTab: Scaling Medical Tabular Data Predictors via Data Consolidation, Enrichment, and Refinement

arXiv:2305.12081v3 Announce Type: replace
Abstract: Tabular data prediction has been employed in medical applications such as patient health risk prediction. However, existing methods usually revolve around the algorithm design while overlooking the significance of data engineering. Medical tabular datasets frequently exhibit significant heterogeneity across different sources, with limited sample sizes per source. As such, previous predictors are often trained on manually curated small datasets that struggle to generalize across different tabular datasets during inference. This paper proposes to scale medical …

abstract algorithm algorithm design applications arxiv consolidation cs.ai cs.lg data data engineering datasets design engineering health however medical patient prediction risk scaling significance tabular tabular data the algorithm type via while