Continuous Predictive Modeling of Clinical Notes and ICD Codes in Patient Health Records

TL;DR

This paper explores early prediction of ICD codes from electronic health records during patient stays, aiming to enable predictive medicine by forecasting diagnoses before clinicians assign codes.

Contribution

It introduces a novel approach for predicting final ICD codes early during hospital stays, outperforming existing methods and enabling timely medical interventions.

Findings

Predictions can be made two days after admission.

The proposed model improves early ICD code prediction accuracy.

Early predictions can support proactive clinical decision-making.

Abstract

Electronic Health Records (EHR) serve as a valuable source of patient information, offering insights into medical histories, treatments, and outcomes. Previous research has developed systems for detecting applicable ICD codes that should be assigned while writing a given EHR document, mainly focusing on discharge summaries written at the end of a hospital stay. In this work, we investigate the potential of predicting these codes for the whole patient stay at different time points during their stay, even before they are officially assigned by clinicians. The development of methods to predict diagnoses and treatments earlier in advance could open opportunities for predictive medicine, such as identifying disease risks sooner, suggesting treatments, and optimizing resource allocation. Our experiments show that predictions regarding final ICD codes can be made already two days after admission and we propose a custom model that improves performance on this early prediction task.