Diagnosis Prediction from Electronic Health Records Using the Binary Diagnosis History Vector Representation

Abstract

Large amounts of rich, heterogeneous information nowadays routinely collected by healthcare providers across the world possess remarkable potential for the extraction of novel medical data and the assessment of different practices in real-world conditions. Specifically in this work, our goal is to use electronic health records (EHRs) to predict progression patterns of future diagnoses of ailments for a particular patient, given the patient's present diagnostic history. Following the highly promising results of a recently proposed approach that introduced the diagnosis history vector representation of a patient's diagnostic record, we introduce a series of improvements to the model and conduct thorough experiments that demonstrate its scalability, accuracy, and practicability in the clinical context. We show that the model is able to capture well the interaction between a large number of ailments that correspond to the most frequent diagnoses, show how the original learning framework can be adapted to increase its prediction specificity, and describe a principled, probabilistic method for incorporating explicit, human clinical knowledge to overcome semantic limitations of the raw EHR data.

Predicting Clinical Diagnosis from Patients Electronic Health Records Using BERT-based Neural Networks

“…In our study we explicitly do not take history into account because at the current stage of our project we focus on isolated visits; however, the history is partially accessible for the model due to the availability of the anamnesis field. The level of granularity for ICD-code in paper [7] is quite similar with ours. We also mainly operate on the second level of ICD-10 classification code hierarchy, but intentionally restrict the number of classes up to 265 (see details in the Data Section 3).…”

“…There have been many approaches to the analysis of the EHR-data and predicting the diagnosis codes (ICDs) proposed in the literature that are related to our work. In particular, papers [7,8] address the task of diagnoses prediction by using the entire patient history. In our study we explicitly do not take history into account because at the current stage of our project we focus on isolated visits; however, the history is partially accessible for the model due to the availability of the anamnesis field.…”

High-risk multimorbidity patterns on the road to cardiovascular mortality

“…The latter fact opens up the possibility to predict future diseases of patients based on their medical history. This has been accomplished with approaches that include collaborative filtering [21,22], frequent itemsets [23,24], learning of transition probabilities between states represented by binary vectors [25,26], deep learning [27,28] and point processes [29].…”

A continual prediction model for inpatient acute kidney injury

“…In past, researchers have done time series analysis of medical data for various applications [25,26,27,28,29] using the aforementioned types of models but they did not do continual prediction. Researchers have also built sequential models to discover longitudinal patterns in patient data, for example, patterns of disease diagnoses in order to predict the diagnosis for the next hospital admission [30,31]. In contrast, our model is not a sequential model but is applied continually over an entire hospital stay in order to predict a particular disease as early as possible using the latest values of patient variables.…”