Improving Diagnostics with Deep Forest Applied to Electronic Health Records

An electronic health record (EHR) is a vital high-dimensional part of medical concepts. Discovering implicit correlations in the information of this data set and the research and informative aspects can improve the treatment and management process. The challenge of concern is the data sources' limitations in finding a stable model to relate medical concepts and use these existing connections. This paper presents Patient Forest, a novel end-to-end approach for learning patient representations from tree-structured data for readmission and mortality prediction tasks. By leveraging statistical features, the proposed model is able to provide an accurate and reliable classifier for predicting readmission and mortality. Experiments on MIMIC-III and eICU datasets demonstrate Patient Forest outperforms existing machine learning models, especially when the training data are limited. Additionally, a qualitative evaluation of Patient Forest is conducted by visualising the learnt representations in 2D space using the t-SNE, which further confirms the effectiveness of the proposed model in learning EHR representations.

Automated summary

An electronic health record (EHR) is a crucial part of medical concepts, and discovering implicit correlations within this data can improve treatment and management. This paper introduces Patient Forest, an innovative approach for learning patient representations from tree-structured data, which outperforms existing models in predicting readmission and mortality. Experiments on MIMIC-III and eICU datasets demonstrate the accuracy and reliability of Patient Forest, especially when training data is limited. The qualitative evaluation using t-SNE visualization further confirms the effectiveness of this model in learning EHR representations.