UniHPF : Universal Healthcare Predictive Framework with Zero Domain Knowledge
This addresses the challenge of utilizing diverse medical data for healthcare predictions, though it appears incremental as a framework for multi-source learning.
The authors tackled the problem of heterogeneous Electronic Healthcare Records (EHR) data limiting predictive model development by proposing UniHPF, a framework that requires no medical domain knowledge and minimal pre-processing for multiple prediction tasks, demonstrating it can build large-scale EHR models processing any form of medical data from distinct EHR systems.
Despite the abundance of Electronic Healthcare Records (EHR), its heterogeneity restricts the utilization of medical data in building predictive models. To address this challenge, we propose Universal Healthcare Predictive Framework (UniHPF), which requires no medical domain knowledge and minimal pre-processing for multiple prediction tasks. Experimental results demonstrate that UniHPF is capable of building large-scale EHR models that can process any form of medical data from distinct EHR systems. We believe that our findings can provide helpful insights for further research on the multi-source learning of EHRs.