This research demonstrates how contrastive learning can significantly improve ECG data analysis with limited labeled examples, enhancing cardiovascular diagnostics across expertise levels.

• Successfully applied machine learning to extract valuable patterns from ECG data
• Achieved effective classification with approximately 100 labels
• Particularly optimized for Japanese ECG data analysis
• Assists medical professionals in identifying cases requiring further examination

This advancement matters for healthcare by democratizing accurate cardiac diagnostics, reducing critical errors, and potentially improving early detection of cardiovascular conditions regardless of physician experience level.

Application of Contrastive Learning on ECG Data: Evaluating Performance in Japanese and Classification with Around 100 Labels