@inproceedings{sun2025RDD-WPW,
 abstract = {Machine Learning for Detection of Wolff-Parkinson-White Syndrome Using ECG Data Background Wolff-Parkinson-White (WPW) syndrome is a rare but clinically significant cardiac condition requiring timely diagnosis to prevent life-threatening complications such as cardiac arrest. Early and accurate detection remains a challenge in clinical practice. Objective This study aimed to develop machine learning (ML) models to improve the detection of WPW syndrome using electrocardiograms (ECGs). Methods ML models were trained using a development dataset comprising 132,045 patient records (805,938 ECGs) with WPW diagnoses identified through the International Classification of Diseases, 10th Revision (ICD-10) code I456. Model performance was evaluated on an independent cohort of 83,304 patients, using their first ECGs recorded during hospital admissions (227,129 ECGs). Gradient boosting (XGBoost), leveraging ECG measurements, and deep learning methods, analyzing ECG traces, were employed. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC). Results The prevalence of WPW syndrome was approximately 0.13% in both the development and test datasets. XGBoost demonstrated the highest performance, achieving an AUROC of 84.69% (95% CI: 81.10%–88.15%). Notably, the model also identified 21 out of 51 ECGs flagged as normal by ICD-10 coding but exhibiting delta wave signatures in machine-analyzed ECG readings, demonstrating its capability to identify potential WPW cases beyond the limitations of ICD-10 coding. Conclusion This study highlights the potential of ML models to replicate and enhance traditional diagnostic pathways, particularly in the early detection of WPW syndrome. Future research could investigate how these models might complement existing clinical workflows by identifying cases overlooked by conventional diagnostic methods, thereby improving diagnostic accuracy and patient outcomes. },
 accepted = {2025-01-01},
 author = {Weijie Sun and Sunil Vasu Kalmady and Abram Hindle and Russell Greiner and Padma Kaul},
 authors = {Weijie Sun and Sunil Vasu Kalmady and Abram Hindle and Russell Greiner and Padma Kaul},
 booktitle = {Rare Disease Day},
 code = {sun2025RDD-WPW},
 date = {2025-02-01},
 funding = {NSERC Discovery},
 location = {Edmonton, Canada},
 pages = {1--2},
 rate = {Unknown},
 role = {Co-Author},
 title = {Machine Learning for Detection of Wolff-Parkinson-White Syndrome Using ECG Data},
 type = {inproceedings},
 url = {http://softwareprocess.ca/pubs/sun2025RDD-WPW.pdf},
 venue = {Rare Disease Day},
 year = {2025}
}