Prediction of Coronary Artery Disease Based on Multimodal, Non-contact Information With Artificial Intelligence
Study Details
Study Description
Brief Summary
The goal of this observational study are 1) to assess the effectiveness of modalities and/or their combination of multimodal non-contact information in predicting coronary artery disease; 2) to prospectively validate the performance of the developed artificial Intelligence models in predicting coronary artery disease.
| Condition or Disease | Intervention/Treatment | Phase |
|---|---|---|
|
Detailed Description
This observational study aims to assess the effectiveness and potential mechanism of modalities of non-contact captured bio-physiological information, including facial RGB information, infrared thermography temperature information, gait information, and wearable device information, individually and/or in combination, in predicting coronary artery disease (CAD) with artificial intelligence technology.
Individuals suspected of CAD and referred for evaluation will be invited to participate in the current study for analyzing the non-contact information and association with underlying CAD status, in order to establish the most efficient artificial Intelligence modeling strategy, and prospectively validate the predictive performance of the developed artificial Intelligence models for CAD prediction.
Study Design
Arms and Interventions
| Arm | Intervention/Treatment |
|---|---|
| Individuals suspected of coronary artery disease Individuals suspected of coronary artery disease and referred for evaluation |
Other: No intervention
No intervention
|
Outcome Measures
Primary Outcome Measures
- Sensitivity of algorithm [At the end of enrollment (1 mouth)]
Sensitivity of algorithm in predicting coronary artery disease assessed in test group
- Specificity of algorithm [At the end of enrollment (1 mouth)]
Sensitivity of algorithm in predicting coronary artery disease assessed in test group
Secondary Outcome Measures
- Area under receiver operating curve (AUC) [At the end of enrollment (1 mouth)]
Area under receiver operating curve of algorithm in predicting coronary artery disease assessed in test group
Other Outcome Measures
- Positive predictive value (PPV) of algorithm [At the end of enrollment (1 mouth)]
Positive predictive value (PPV) of algorithm in predicting coronary artery disease assessed in test group
- Negative predictive value (NPV) [At the end of enrollment (1 mouth)]
Negative predictive value (NPV) of algorithm in predicting coronary artery disease assessed in test group
- Diagnostic accuracy rate [At the end of enrollment (1 mouth)]
Diagnostic accuracy rate of algorithm in predicting coronary artery disease assessed in test group
Eligibility Criteria
Criteria
Inclusion Criteria:
- Suspected individuals referred to for coronary angiography, coronary computer tomography angiography, or functional stress tests.
Exclusion Criteria:
-
Prior percutaneous coronary intervention (PCI)
-
Prior coronary artery bypass graft (CABG)
-
Undergoing confirmatory coronary evaluation as pre-operation routines for other cardiac diseases
-
With artificial body alteration (e.g. cosmetic surgery, facial trauma, or make-up) that may affect the non-contact information of study interest
-
Age less than 18 years old
-
Other circumstances that prevent participants from cooperating with the study process
-
Decline to consent for study participation
Contacts and Locations
Locations
No locations specified.Sponsors and Collaborators
- China National Center for Cardiovascular Diseases
Investigators
- Principal Investigator: Shen Lin, M.D., Ph.D., Fuwai Hospital, CAMS & PUMC
Study Documents (Full-Text)
None provided.More Information
Publications
- 2022-GSP-QN-10