What is the clinical evidence from this study?

Study design: Other. Population: Tachycardia onset (n=5958). Intervention: TOP-Net (Bidirectional Long Short-term Memory model) vs. Baseline machine learning models (CNN, LSTM, XGBoost, MLP, Random forest). Primary outcome: Area under the receiver operating characteristic curve (AUROC) for predicting tachycardia onset 6 hours in advance (AUROC 0.796, 95% CI 0.768-0.824).

February 21, 2021Open Access

TOP-Net Prediction Model Using Bidirectional Long Short-term Memory and Medical-Grade Wearable Multisensor System for Tachycardia Onset: Algorithm Development Study

Key Result

The TOP-Net deep learning model predicted tachycardia onset 6 hours in advance in ICU patients with an AUROC of 0.796, outperforming baseline machine learning models.

Structured PICO

Does the TOP-Net deep learning model using wearable sensor and EHR data accurately predict tachycardia onset hours in advance compared to baseline machine learning models?

Population

5,958 total patients, comprising an ICU cohort (n=5,699) from the MIMIC-III database and a general ward cohort (n=259) from a real clinical environment. ICU cohort: median age 66.15, 57.2% female, 86.6% with cardiovascular disease history. General ward cohort: median age 61.00, 40.5% female, 90.3% with cardiovascular disease history.

Intervention

TOP-Net, a bidirectional long short-term memory (BiLSTM) deep learning prediction model utilizing 8 types of data from wearable sensors (heart rate, respiratory rate, SpO2) and electronic health records (age, gender, admission type, first care unit, cardiovascular disease history).

Comparator

Baseline machine learning models including Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Extreme Gradient Boosting (XGBoost), Multilayer Perceptron (MLP), and Random Forest.

Outcome

Predictive performance for tachycardia onset 0 to 6 hours in advance, evaluated using area under the receiver operating characteristic curve (AUROC), sensitivity, specificity, accuracy, F1 score, and precision.

The TOP-Net deep learning model, utilizing continuous wearable sensor data and electronic health records, can accurately predict tachycardia onset up to 6 hours in advance, potentially enabling earlier clinical intervention.

Main Result

Effect estimate: AUROC 0.796 (95% CI 0.768-0.824)

Limitations

Small sample size in the general ward cohort requiring transfer learning
Differences in disease severity between ICU and general ward patients may affect prediction performance

Abstract

Background Without timely diagnosis and treatment, tachycardia, also called tachyarrhythmia, can cause serious complications such as heart failure, cardiac arrest, and even death. The predictive performance of conventional clinical diagnostic procedures needs improvement in order to assist physicians in detecting risk early on. Objective We aimed to develop a deep tachycardia onset prediction (TOP-Net) model based on deep learning (ie, bidirectional long short-term memory) for early tachycardia diagnosis with easily accessible data. Methods TOP-Net leverages 2 easily accessible data sources: vital signs, including heart rate, respiratory rate, and blood oxygen saturation (SpO2) acquired continuously by wearable embedded systems, and electronic health records, containing age, gender, admission type, first care unit, and cardiovascular disease history. The model was trained with a large data set from an intensive care unit and then transferred to a real-world scenario in the general ward. In this study, 3 experiments incorporated merging patients’ personal information, temporal memory, and different feature combinations. Six metrics (area under the receiver operating characteristic curve AUROC, sensitivity, specificity, accuracy, F1 score, and precision) were used to evaluate predictive performance. Results TOP-Net outperformed the baseline models on the large critical care data set (AUROC 0.796, 95% CI 0.768-0.824; sensitivity 0.753, 95% CI 0.663-0.793; specificity 0.720, 95% CI 0.645-0.758; accuracy 0.721; F1 score 0.718; precision 0.686) when predicting tachycardia onset 6 hours in advance. When predicting tachycardia onset 2 hours in advance with data acquired from our hospital using the transferred TOP-Net, the 6 metrics were 0.965, 0.955, 0.881, 0.937, 0.793, and 0.680, respectively. The best performance was achieved using comprehensive vital signs (heart rate, respiratory rate, and SpO2) statistical information. Conclusions TOP-Net is an early tachycardia prediction model that uses 8 types of data from wearable sensors and electronic health records. When validated in clinical scenarios, the model achieved a prediction performance that outperformed baseline models 0 to 6 hours before tachycardia onset in the intensive care unit and 2 hours before tachycardia onset in the general ward. Because of the model’s implementation and use of easily accessible data from wearable sensors, the model can assist physicians with early discovery of patients at risk in general wards and houses.

Mark Helpful

Bookmark

Relay

View Full Paper