February 7, 2019Open Access

Effectiveness of LSTMs in Predicting Congestive Heart Failure Onset

Q: What is the clinical evidence from this study?

Study design: Case-Control. Population: Congestive Heart Failure (n=216394). Intervention: LSTM-based Recurrent Neural Network vs. Baseline machine learning models (Logistic Regression, MLP, Random Forest, 1-D CNN). Primary outcome: Area under the receiver operating characteristic curve (AUCROC) for predicting CHF onset.

Key Result

An LSTM-based recurrent neural network achieved an AUCROC of 0.9148 for predicting the onset of congestive heart failure 15 months in advance, outperforming baseline machine learning models.

Study Design

Type

Case-Control (n=216,394)

PICO

Population

Congestive Heart Failure (n=216,394)

Intervention / Comparator

LSTM-based Recurrent Neural Network vs Baseline machine learning models (Logistic Regression, MLP, Random Forest, 1-D CNN)

Primary Outcome

Area under the receiver operating characteristic curve (AUCROC) for predicting CHF onset

Main Result

Absolute Event Rate: 0.9148% vs 0.898%

Limitations

CHF is a heterogeneous syndrome with a range of etiologies and the case definition does not distinguish these.
Significant heterogeneity in CHF definition makes it difficult to compare findings.
The model requires validation in completely independent populations to assess its generalizability.

Abstract

In this paper we present a Recurrent neural networks (RNN) based architecture that achieves an AUCROC of 0.9147 for predicting the onset of Congestive Heart Failure (CHF) 15 months in advance using a 12-month observation window on a large cohort of 216,394 patients. We believe this to be the largest study in CHF onset prediction with respect to the number of CHF case patients in the cohort and the test set (3,332 CHF patients) on which the AUC metrics are reported. We explore the extent to which LSTM (Long Short Term Memory) based model, a variant of RNNs, can accurately predict the onset of CHF when compared to known linear baselines like Logistic Regression, Random Forests and deep learning based models such as Multi-Layer Perceptron and Convolutional Neural Networks. We utilize demographics, medical diagnosis and procedure data from 21,405 CHF and 194,989 control patients to as our features. We describe our feature embedding strategy for medical diagnosis codes that accommodates the sparse, irregular, longitudinal, and high-dimensional characteristics of EHR data. We empirically show that LSTMs can capture the longitudinal aspects of EHR data better than the proposed baselines. As an attempt to interpret the model, we present a temporal data analysis-based technique on false positives to attribute feature importance. A model capable of predicting the onset of congestive heart failure months in the future with this level of accuracy and precision can support efforts of practitioners to implement risk factor reduction strategies and researchers to begin to systematically evaluate interventions to potentially delay or avert development of the disease with high mortality, morbidity and significant costs.

Bookmark

View Full Paper

Bookmark

View Full Paper

Cite This Study

Mallya et al. (Thu,) conducted a case-control in Congestive Heart Failure (n=216,394). LSTM-based Recurrent Neural Network vs. Baseline machine learning models (Logistic Regression, MLP, Random Forest, 1-D CNN) was evaluated on Area under the receiver operating characteristic curve (AUCROC) for predicting CHF onset. An LSTM-based recurrent neural network achieved an AUCROC of 0.9148 for predicting the onset of congestive heart failure 15 months in advance, outperforming baseline machine learning models.

synapsesocial.com/papers/6a17439e4e3ad33a20785f81 https://doi.org/https://doi.org/10.48550/arxiv.1902.02443