Background: When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. Objective: We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead–based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. Methods: We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. Results: The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads—ideally more than 4 leads—is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. Conclusions: By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.
Background: Successful sepsis treatment depends on early diagnosis. We aimed to develop and validate a system to predict sepsis and septic shock in real time using deep learning. Methods: Clinical data were retrospectively collected from electronic medical records (EMRs). Data from 2010 to 2019 were used as development data, and data from 2020 to 2021 were used as validation data. The collected EMRs consisted of eight vital signs, 13 laboratory data points, and three demographic information items. We validated the deep-learning-based sepsis and septic shock early prediction system (DeepSEPS) using the validation datasets and compared our system with other traditional early warning scoring systems, such as the national early warning score, sequential organ failure assessment (SOFA), and quick sequential organ failure assessment. Results: DeepSEPS achieved even higher area under receiver operating characteristic curve (AUROC) values (0.7888 and 0.8494 for sepsis and septic shock, respectively) than SOFA. The prediction performance of traditional scoring systems was enhanced because the early prediction time point was close to the onset time of sepsis; however, the DeepSEPS scoring system consistently outperformed all conventional scoring systems at all time points. Furthermore, at the time of onset of sepsis and septic shock, DeepSEPS showed the highest AUROC (0.9346). Conclusions: The sepsis and septic shock early warning system developed in this study exhibited a performance that is worth considering when predicting sepsis and septic shock compared to other traditional early warning scoring systems. DeepSEPS showed better performance than existing sepsis prediction programs. This novel real-time system that simultaneously predicts sepsis and septic shock requires further validation.
Electrocardiograms (ECG) are widely employed as a diagnostic tool for monitoring electrical signals originating from a heart. Recent machine learning research efforts have focused on the application of screening various diseases using ECG signals. However, adapting to the application of screening disease is challenging in that labeled ECG data are limited. Achieving general representation through self-supervised learning (SSL) is a well-known approach to overcome the scarcity of labeled data; however, a naive application of SSL to ECG data, without considering the spatial-temporal relationships inherent in ECG signals, may yield suboptimal results. In this paper, we introduce ST-MEM (Spatio-Temporal Masked Electrocardiogram Modeling), designed to learn spatio-temporal features by reconstructing masked 12-lead ECG data. ST-MEM outperforms other SSL baseline methods in various experimental settings for arrhythmia classification tasks. Moreover, we demonstrate that ST-MEM is adaptable to various lead combinations. Through quantitative and qualitative analysis, we show a spatio-temporal relationship within ECG data. Our code is available at https://github.com/bakqui/ST-MEM.
Early detection of mortality in intensive care units (ICUs) is significant to improve patient survival. Since heterogeneous data can be collected from the intensive care unit, there are meaningful static features (e.g., ICU type, gender, and ethnicity) that can enhance the performance of early detection of mortality. Although the characteristics of physiological representation for each patient can be different by static features, recently proposed mortality prediction models overlook these features and only rely on a one-size-fits-all model. In this paper, we propose a simple yet effective domain encoder named G-DAM (Graph-Domain Aggregation Module) that relies on a relational graph between each static feature to adapt to patient groups and capture the relationship. We show the importance of utilizing static features to predict ICU mortality through extensive experimental results. The experimental results demonstrate that our proposed G-DAM outperforms existing baseline methods for encoding static features not only in the major domains with dense populations but also in the minor domains with sparse populations. The ablative study also shows different sequential models can perform better combined with G-DAM.
Background: Despite advances in central nervous system (CNS)-protective anesthetic and surgical strategies, perioperative stroke remains a significant concern in high-risk cardiothoracic and vascular surgery (CTVS). Early detection, facilitating timely and prompt intervention, is often hindered by sedation and mechanical ventilation (MV) in the immediate postoperative period. This study aimed to develop and validate a deep learning (DL)-based artificial intelligence (AI) program for early detection of severe, surgery-related major adverse cerebral injury (sMACI), encompassing fatal CNS and systemic insults in high-risk CTVS patients. Methods: We retrospectively analyzed data from 4,455 patients who underwent seven types of CTVS (2010–2021), requiring postoperative ICU admission and ongoing MV. Continuous vital signs (heart rate, blood pressures, respiratory rate, pulse oximetry saturation, temperature) were extracted from the operating room (OR) and intensive care unit (ICU), along with demographic and laboratory data. sMACI was defined as significant postoperative CNS injury (modified Rankin Scale ≥3 at 1 month) or 1-month mortality. Two-tier DL models were constructed: Model 1 using ICU data alone, and Model 2 integrating pre-ICU and ICU data. Performance in detecting sMACI within 24 hours of ICU admission was assessed using the area under the receiver operating characteristic curve (AUROC) and precision-recall curve (AUPRC). Results: Among 4,455 patients, 5% experienced sMACI. Model 1 achieved an AUROC of 0.809 (95% CI: 0.759–0.858) and an AUPRC of 0.275 (0.195–0.375). Model 2 showed improved detection (AUROC 0.826 [0.781–0.871]; AUPRC 0.322 [0.233–0.423]). Both models outperformed conventional early warning scores and other machine learning algorithms, demonstrating robust performance as early as 4 hours after ICU admission. Key contributors included systolic blood pressure, heart rate, diastolic blood pressure, mean arterial pressure, and pulse oximetry saturation. Conclusions: A DL-based AI program leveraging continuous vital signs enables effective early detection of severe surgery-related CNS and systemic injury in high-risk CTVS patients, outperforming established scoring systems and other machine learning approaches.
BACKGROUND When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. OBJECTIVE We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead–based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. METHODS We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. RESULTS The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads—ideally more than 4 leads—is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. CONCLUSIONS By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.