Study Governance and Limitations

Risks & contraindications

Other

Other

Other

Highlighted claims

• Patient-level data will not be publicly released because of agreements and privacy requirements. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol
• The analyses will use deidentified or anonymised data in secure institutional environments and report only aggregate results. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol
• Ethics approval has been obtained from review boards at the University of Calgary, Ottawa Health Science Network, and University Health Network. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol
• Use of retrospective routinely collected records may require extensive imputation and introduce site-specific measurement bias. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol
• Transportability to other health systems is uncertain because participating centres may share referral pathways and care protocols. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol
• Dataset shift and calibration drift after deployment may require monitoring and model updates. — Development and cross-site validation of machine-learning models for diagnosis and prognosis of stable angina with and without obstructive coronary artery disease: a study protocol