Artificial Intelligence-enabled Screening and Diagnosis of Cardiomyopathies Using Coronary Computer Tomography Angiography
Overview
- Phase
- Not Applicable
- Intervention
- Not specified
- Conditions
- Cardiovascular Diseases
- Sponsor
- Shanghai Zhongshan Hospital
- Enrollment
- 5000
- Primary Endpoint
- Diagnostic performance
- Status
- Not yet recruiting
- Last Updated
- last year
Overview
Brief Summary
The goal of this observational and diagnostic study is to develop and validate an artificial intelligence assisted approach for coronary computer tomography angiography-(CCTA)-based screening and diagnosis of cardiomyopathies in patients with suspected coronary artery diseases. This study aims to develop a computerized CCTA interpretation using artificial intelligence for multi-label classification task to assist cardiomyopathy diagnosis in the clinical workflow.
Detailed Description
Cardiovascular diseases (CVD) are the leading causes of death and disability worldwide. With coronary artery disease accounting for a large proportion of CVD disease burden, coronary computer tomography angiography (CCTA) has become widely used for a comprehensive assessment of the total coronary atherosclerotic burden. In contrast, cardiac magnetic resonance (CMR) remains the gold standard for evaluating and diagnosing cardiomyopathies. However, clinical application of CMR has been hindered by the time and cost of examination and shortage of qualified doctors and staff. Consequently, the value of CCTA in screening and diagnosis in cardiomyopathies warrants further investigation. The ability of artificial intelligence to learn distinctive features and to recognize characteristic patterns on big data without extensive manual labor makes it highly effective for interpreting CCTA data. Although very few studies investigated the diagnostic value of CCTA for myocardiopathies, which is by far not established or applied in clinical practice by radiologists, automated image analysis has a clear advantage compared to humans by offering objective and uniform solutions. Further, whether a comprehensive, end-to-end, artificial intelligent approach can be used to analyse CCTA for diagnosis multi-classifications of cardiomyopathies remains unknown. Therefore, this study aims to develop and validate an artificial intelligence assisted approach on CCTA for screening and diagnosis of cardiomyopathies in patients with suspected coronary artery diseases.
Investigators
Junbo Ge
Prof
Shanghai Zhongshan Hospital
Eligibility Criteria
Inclusion Criteria
- •A clinical diagnosis of cardiomyopathies, including hypertrophic cardiomyopathy, dilated cardiomyopathy, restrictive cardiomyopathy, cardiac amyloidosis, myocarditis, arrhythmogenic right ventricular cardiomyopathy, and coronary artery disease/ischemic heart disease.
- •At least one CCTA before surgery or implantable device treatment.
Exclusion Criteria
- •No recorded diagnosis of cardiomyopathy or undetermined type of cardiomyopathy.
- •A clinical diagnosis of secondary cardiac abnormalities due to other organic or systemic diseases.
- •Surgery or implantable device treatment before CCTA examination.
- •Control cohort:
- •Inclusion Criteria: participants with at least one CCTA examination.
- •Exclusion Criteria: clinical diagnosis of cardiovascular diseases (including cardiomyopathy, history of myocardial infarction, history of cardiac surgery, stent implantation, ICD implantation and so on) or secondary cardiac abnormalities due to systemic diseases.
Outcomes
Primary Outcomes
Diagnostic performance
Time Frame: CCTA examination before surgical or interventional treatments.
The performance of the AI models is evaluated by assessing their sensitivity, specificity, precision and F1 score (harmonic mean of the predictive positive value and sensitivity), with two-sided 95% CIs, as well as the AUC of the ROC with two-sided CIs. The F1 score is complementary to the AUC, which is particularly useful in the setting of multiclass prediction and less sensitive than the AUC in settings of class imbalance. For an aggregate measure of model performance, the investigators compute the class frequency-weighted mean for the F1 score and the AUC. Other diagnostic performance assessing metrics include true-positive rate, true-negative rate, false-positive rate, false-negative rate, precision, sensitivity (recall), specificity, positive predictive value, and negative predictive value.