Transfer learning enables prediction of myocardial injury from continuous single-lead electrocardiography

Transfer learning in ECG diagnosis: Is it effective?

The adoption of deep learning in ECG diagnosis is often hindered by the scarcity of large, well-labeled datasets in real-world scenarios, leading to the use of transfer learning to leverage features learned from larger datasets. Yet the prevailing assumption that transfer learning consistently outperforms training from scratch has never been systematically validated. In this study, we conduct the first extensive empirical study on the effectiveness of transfer learning in multi-label ECG classification, by investigating comparing the fine-tuning performance with that of training from scratch, covering a variety of ECG datasets and deep neural networks. Firstly, We confirm that fine-tuning is the preferable choice for small downstream datasets; however, it does not necessarily improve performance. Secondly, the improvement from fine-tuning declines when the downstream dataset grows. With a sufficiently large dataset, training from scratch can achieve comparable performance, albeit requiring a longer training time to catch up. Thirdly, fine-tuning can accelerate convergence, resulting in faster training process and lower computing cost. Finally, we find that transfer learning exhibits better compatibility with convolutional neural networks than with recurrent neural networks, which are the two most prevalent architectures for time-series ECG applications. Our results underscore the importance of transfer learning in ECG diagnosis, yet depending on the amount of available data, researchers may opt not to use it, considering the non-negligible cost associated with pre-training.

Nomogram for predicting myocardial injury in pediatric patients undergoing living donor liver transplantation for biliary atresia

BACKGROUND

Myocardial injury is common during liver transplantation and is associated with poor outcomes. The development of a reliable prediction system for this type of injury is crucial for reducing the incidence of cardiac complications in children receiving living donor liver transplantation (LDLT). However, establishing a practical myocardial injury prediction system for children with biliary atresia remains a considerable challenge.

AIM

To create and validate a nomogram model for predicting myocardial injury in children with biliary atresia who received LDLT.

METHODS

Clinical data from pediatric patients who received LDLT for biliary atresia between November, 2019 and January, 2022 were retrospectively analyzed. The complete dataset was randomly partitioned into a training set and a validation set at a ratio of 7:3. Least absolute shrinkage and selection operator regression was used to preliminarily screen out the predictors of myocardial injury. The prediction model was established via multivariable logistic regression and presented in the form of a nomogram.

RESULTS

This study included 321 patients, 150 (46.7%) of whom had myocardial injury. The participants were randomly allocated into two groups: A training group consisting of 225 patients and a validation group comprising 96 patients. The predictors in this nomogram included the preoperative neutrophil-to-lymphocyte ratio, high sensitivity C-reactive protein level, pediatric end-stage liver disease score and postreperfusion syndrome. The area under the curve for predicting myocardial injury was 0.865 in the training set and 0.856 in the validation set. The calibration curve revealed that the predicted values were very close to the actual values in the two sets. Decision curve analysis revealed that the prediction model offered a favorable net benefit.

CONCLUSION

The nomogram developed in this study effectively predicts myocardial injury in pediatric LDLT patients, showing good accuracy and potential for clinical application.

Artificial intelligence in cardiac telemetry

Cardiac telemetry has evolved into a vital tool for continuous cardiac monitoring and early detection of cardiac abnormalities. In recent years, artificial intelligence (AI) has become increasingly integrated into cardiac telemetry, making a shift from traditional statistical machine learning models to more advanced deep neural networks. These modern AI models have demonstrated superior accuracy and the ability to detect complex patterns in telemetry data, enhancing real-time monitoring, predictive analytics and personalised cardiac care. In our review, we examine the current state of AI in cardiac telemetry, focusing on deep learning techniques, their clinical applications, the challenges and limitations faced by these models, and potential future directions in this promising field.

Deep learning and electrocardiography: systematic review of current techniques in cardiovascular disease diagnosis and management

This paper reviews the recent advancements in the application of deep learning combined with electrocardiography (ECG) within the domain of cardiovascular diseases, systematically examining 198 high-quality publications. Through meticulous categorization and hierarchical segmentation, it provides an exhaustive depiction of the current landscape across various cardiovascular ailments. Our study aspires to furnish interested readers with a comprehensive guide, thereby igniting enthusiasm for further, in-depth exploration and research in this realm.

Machine Learning Applications in Acute Coronary Syndrome: Diagnosis, Outcomes and Management

Acute coronary syndrome (ACS) is a leading cause of death worldwide. Prompt and accurate diagnosis of acute myocardial infarction (AMI) or ACS is crucial for improved management and prognosis of patients. The rapid growth of machine learning (ML) research has significantly enhanced our understanding of ACS. Most studies have focused on applying ML to detect ACS, predict prognosis, manage treatment, identify risk factors, and discover potential biomarkers, particularly using data from electrocardiograms (ECGs), electronic medical records (EMRs), imaging, and omics as the main data modality. Additionally, integrating ML with smart devices such as wearables, smartphones, and sensor technology enables real-time dynamic assessments, enhancing clinical care for patients with ACS. This review provided an overview of the workflow and key concepts of ML as they relate to ACS. It then provides an overview of current ML algorithms used for ACS diagnosis, prognosis, identification of potential risk biomarkers, and management. Furthermore, we discuss the current challenges faced by ML algorithms in this field and how they might be addressed in the future, especially in the context of medicine.

Artificial neural networks for ECG interpretation in acute coronary syndrome: A scoping review

INTRODUCTION

The electrocardiogram (ECG) is a crucial diagnostic tool in the Emergency Department (ED) for assessing patients with Acute Coronary Syndrome (ACS). Despite its widespread use, the ECG has limitations, including low sensitivity of the STEMI criteria to detect Acute Coronary Occlusion (ACO) and poor inter-rater reliability. Emerging ECG features beyond the traditional STEMI criteria show promise in improving early ACO diagnosis, but complexity hinders widespread adoption. The potential integration of Artificial Neural Networks (ANN) holds promise for enhancing diagnostic accuracy and addressing reliability issues in ECG interpretation for ACO symptoms.

METHODS

Ovid MEDLINE, CINAHL, EMBASE, Cochrane, PubMed and Scopus were searched from inception through to 8th of December 2023. A thorough search of the grey literature and reference lists of relevant articles was also performed to identify additional studies. Articles were included if they reported the use of ANN for ECG interpretation of Acute Coronary Syndrome in the Emergency Department patients.

RESULTS

The search yielded a total of 244 articles. After removing duplicates and excluding non-relevant articles, 14 remained for analysis. There was significant heterogeneity in the types of ANN models used and the outcomes assessed, making direct comparisons challenging. Nevertheless, ANN appeared to demonstrate higher accuracy than physician interpreters for the evaluated outcomes and this proved independent of both specialty and years of experience.

CONCLUSIONS

The interpretation of ECGs in patients with suspected ACS using ANN appears to be accurate and potentially superior when compared to human interpreters and computerised algorithms. This appears consistent across various ANN models and outcome variables. Future investigations should emphasise ANN interpretation of ECGs in patients with ACO, where rapid and accurate diagnosis can significantly benefit patients through timely access to reperfusion therapies.

Trustworthy deep learning framework for the detection of abnormalities in X-ray shoulder images

Musculoskeletal conditions affect an estimated 1.7 billion people worldwide, causing intense pain and disability. These conditions lead to 30 million emergency room visits yearly, and the numbers are only increasing. However, diagnosing musculoskeletal issues can be challenging, especially in emergencies where quick decisions are necessary. Deep learning (DL) has shown promise in various medical applications. However, previous methods had poor performance and a lack of transparency in detecting shoulder abnormalities on X-ray images due to a lack of training data and better representation of features. This often resulted in overfitting, poor generalisation, and potential bias in decision-making. To address these issues, a new trustworthy DL framework has been proposed to detect shoulder abnormalities (such as fractures, deformities, and arthritis) using X-ray images. The framework consists of two parts: same-domain transfer learning (TL) to mitigate imageNet mismatch and feature fusion to reduce error rates and improve trust in the final result. Same-domain TL involves training pre-trained models on a large number of labelled X-ray images from various body parts and fine-tuning them on the target dataset of shoulder X-ray images. Feature fusion combines the extracted features with seven DL models to train several ML classifiers. The proposed framework achieved an excellent accuracy rate of 99.2%, F1Score of 99.2%, and Cohen's kappa of 98.5%. Furthermore, the accuracy of the results was validated using three visualisation tools, including gradient-based class activation heat map (Grad CAM), activation visualisation, and locally interpretable model-independent explanations (LIME). The proposed framework outperformed previous DL methods and three orthopaedic surgeons invited to classify the test set, who obtained an average accuracy of 79.1%. The proposed framework has proven effective and robust, improving generalisation and increasing trust in the final results.

Towards federated transfer learning in electrocardiogram signal analysis

Modern methods in artificial intelligence perform very well on many healthcare datasets, at times outperforming trained doctors. However, many assumptions made in model training are not justifiable in clinical settings. In this work, we propose a method to train classifiers for electrocardiograms, able to deal with data of disparate input dimensions, distributed across different institutions, and able to protect patient privacy. In addition, we propose a simple method for creating federated datasets from any centralized dataset. We use autoencoders in conjunction with federated learning to model a highly heterogeneous modeling problem using the Massachusetts Institute of Technology Beth Israel Hospital Arrhythmia dataset, the Computing in Cardiology 2017 challenge dataset, and the PTB-XL dataset. For an encoding dimension of 1000, our federated classifier achieves an accuracy, precision, recall, and F1 score of 73.0%, 66.6%, 73.0%, and 69.7%, respectively. Our results suggest that dropping commonly made assumptions significantly complicate training and that as a result, estimates of performance of many machine learning models may overestimate performance when adopted for clinical settings.

Evaluating progress in automatic chest X-ray radiology report generation

Artificial intelligence (AI) models for automatic generation of narrative radiology reports from images have the potential to enhance efficiency and reduce the workload of radiologists. However, evaluating the correctness of these reports requires metrics that can capture clinically pertinent differences. In this study, we investigate the alignment between automated metrics and radiologists' scoring of errors in report generation. We address the limitations of existing metrics by proposing new metrics, RadGraph F1 and RadCliQ, which demonstrate stronger correlation with radiologists' evaluations. In addition, we analyze the failure modes of the metrics to understand their limitations and provide guidance for metric selection and interpretation. This study establishes RadGraph F1 and RadCliQ as meaningful metrics for guiding future research in radiology report generation.

Deep learning augmented ECG analysis to identify biomarker-defined myocardial injury

Chest pain is a common clinical complaint for which myocardial injury is the primary concern and is associated with significant morbidity and mortality. To aid providers' decision-making, we aimed to analyze the electrocardiogram (ECG) using a deep convolutional neural network (CNN) to predict serum troponin I (TnI) from ECGs. We developed a CNN using 64,728 ECGs from 32,479 patients who underwent ECG within 2 h prior to a serum TnI laboratory result at the University of California, San Francisco (UCSF). In our primary analysis, we classified patients into groups of TnI < 0.02 or ≥ 0.02 µg/L using 12-lead ECGs. This was repeated with an alternative threshold of 1.0 µg/L and with single-lead ECG inputs. We also performed multiclass prediction for a set of serum troponin ranges. Finally, we tested the CNN in a cohort of patients selected for coronary angiography, including 3038 ECGs from 672 patients. Cohort patients were 49.0% female, 42.8% white, and 59.3% (19,283) never had a positive TnI value (≥ 0.02 µg/L). CNNs accurately predicted elevated TnI, both at a threshold of 0.02 µg/L (AUC = 0.783, 95% CI 0.780-0.786) and at a threshold of 1.0 µg/L (AUC = 0.802, 0.795-0.809). Models using single-lead ECG data achieved significantly lower accuracy, with AUCs ranging from 0.740 to 0.773 with variation by lead. Accuracy of the multi-class model was lower for intermediate TnI value-ranges. Our models performed similarly on the cohort of patients who underwent coronary angiography. Biomarker-defined myocardial injury can be predicted by CNNs from 12-lead and single-lead ECGs.