Abstract:The label scarcity problem is the main challenge that hinders the wide application of deep learning systems in automatic cardiovascular diseases (CVDs) detection using electrocardiography (ECG). Tuning pre-trained models alleviates this problem by transferring knowledge learned from large datasets to downstream small datasets. However, bottlenecks in computational efficiency and detection performance limit its clinical applications. It is difficult to improve the detection performance without significantly sacrificing the computational efficiency during model training. Here, we propose a computation-efficient semi-supervised learning paradigm (CE-SSL) for robust and computation-efficient CVDs detection using ECG. It enables a robust adaptation of pre-trained models on downstream datasets with limited supervision and high computational efficiency. First, a random-deactivation technique is developed to achieve robust and fast low-rank adaptation of pre-trained weights. Subsequently, we propose a one-shot rank allocation module to determine the optimal ranks for the update matrices of the pre-trained weights. Finally, a lightweight semi-supervised learning pipeline is introduced to enhance model performance by leveraging labeled and unlabeled data with high computational efficiency. Extensive experiments on four downstream datasets demonstrate that CE-SSL not only outperforms the state-of-the-art methods in multi-label CVDs detection but also consumes fewer GPU footprints, training time, and parameter storage space. As such, this paradigm provides an effective solution for achieving high computational efficiency and robust detection performance in the clinical applications of pre-trained models under limited supervision. Code and Supplementary Materials are available at <a class="link-external link-https" href="https://github.com/KAZABANA/CE-SSL" rel="external noopener nofollow">this https URL</a>

Concurrent Classifier Error Detection (CCED) in Large Scale Machine Learning Systems

Concurrent Classifier Error Detection (CCED) in Large Scale Machine Learning Systems

Concurrent Linguistic Error Detection (CLED) for Large Language Models

Error-Resilient Machine Learning in Near Threshold Voltage Via Classifier Ensemble.

Context-aware Feature Reconstruction for Class-Incremental Anomaly Detection and Localization

Error-correcting Ability Based Collaborative Multi-Layer Selective Classifier Ensemble Model for Intrusion Detection

CE-SSL: Computation-Efficient Semi-Supervised Learning for ECG-based Cardiovascular Diseases Detection

Mitigating Catastrophic Forgetting in Task-Incremental Continual Learning with Adaptive Classification Criterion

Improved Continually Evolved Classifiers for Few-Shot Class-Incremental Learning

Continual learning classification method with constant-sized memory cells based on the artificial immune system

Efficient Synchronous Real-Time CADe for Multicategory Lesions in Gastroscopy by Using Multiclass Detection Model

Adaptive Rentention & Correction for Continual Learning

Corrective Classification: Classifier Ensembling with Corrective and Diverse Base Learners

Modelling Concurrency Bugs Using Machine Learning

Corrective Classification: Learning From Data Imperfections With Aggressive And Diverse Classifier Ensembling

MCES: Multi-classifier Ensemble System for Malware Detection and Identification

CECT: Controllable Ensemble CNN and Transformer for COVID-19 Image Classification

Confident Learning: Estimating Uncertainty in Dataset Labels

Error Detection and Constraint Recovery in Hierarchical Multi-Label Classification without Prior Knowledge

DECIDER: Leveraging Foundation Model Priors for Improved Model Failure Detection and Explanation

Combining data augmentation and domain information with TENER model for Clinical Event Detection