Research Ideas Using CTCA Patient Data
Based on your comprehensive dataset from patients who underwent Coronary Computed Tomography Angiography, here are 10 high-impact research ideas:
1. Machine Learning Model for MACE Prediction Using CTCA-Derived Features
Develop and validate a machine learning algorithm combining clinical risk factors (age, gender, diabetes, hypertension, hyperlipidemia, smoking, family history) with CTCA-specific parameters (calcium score, vessel disease severity, plaque characteristics) to predict major adverse cardiovascular events. Recent evidence demonstrates that ML models based on CCTA data achieve pooled AUROCs of 0.7981 for MACE prediction, with logistic regression models performing particularly well (AUROC 0.8229) 1. Your dataset's inclusion of both traditional risk factors and imaging parameters positions it ideally for developing models that outperform basic feature extraction approaches 1.
2. Comparative Effectiveness of HeartFlow FFR-CT Versus Standard CTCA in Guiding Management Decisions
Investigate whether HeartFlow referral leads to different management recommendations, improved adherence to those recommendations, and superior outcomes on invasive coronary angiography compared to standard CTCA interpretation alone. The 2021 ACC/AHA chest pain guidelines note that CCTA without stenosis has a 3-year CAD event rate of only 0.9%, but functional assessment adds prognostic value 1. Your data on HeartFlow referrals, management recommendations, recommendation adherence, and angiographic outcomes provides a unique opportunity to assess real-world comparative effectiveness 1.
3. Diabetes-Specific Risk Stratification Model Using CTCA Parameters
Create a diabetes-specific risk prediction model, as diabetic patients demonstrate distinct coronary disease patterns with more extensive disease, higher segment involvement scores, and more high-risk plaque features than patients with hypertension or hyperlipidemia alone. Evidence shows DM patients have the highest prevalence of obstructive CAD (34% vs 19% for hypertension vs 15% for dyslipidemia), more extensive disease (SIS 3.1 vs 2.1 vs 1.4), and higher rates of positive remodeling, low attenuation, and spotty calcification 2. The PROMISE trial demonstrated that diabetic patients undergoing CCTA had lower cardiovascular death or MI rates compared to stress testing (adjusted HR 0.38) 1.
4. Gender-Specific Differences in CTCA Findings and Outcomes
Analyze sex-based differences in calcium scores, vessel disease severity, plaque characteristics, management recommendations, and adherence patterns, as women demonstrate different pathophysiology and worse outcomes despite lower prevalence of obstructive CAD. Women are more likely to have myocardial infarction from plaque erosion and nonobstructive coronary arteries, while men more commonly have plaque rupture from epicardial disease 1. Traditional risk prediction models like Diamond-Forrester substantially overestimate CAD likelihood in women 1.
5. Radiation Dose Optimization Study Correlating DLP with Diagnostic Quality
Examine the relationship between DLP dose, heart rate, calcium score, and diagnostic adequacy to identify optimal scanning protocols that minimize radiation while maintaining diagnostic accuracy. This addresses a critical gap in comparative effectiveness research, as the 2010 JACC Cardiovascular Imaging statement emphasized the need for studies examining quality as well as value in CV imaging 1.
6. Atherogenic Dyslipidemia Pattern and CTCA Findings
Investigate whether patients with combined hypertriglyceridemia and low HDL (atherogenic dyslipidemia) demonstrate more extensive coronary disease, higher calcium scores, and different plaque morphology compared to isolated hypercholesterolemia. Research shows 41.3% of CAD patients have atherogenic dyslipidemia, with stronger correlations to inflammatory markers and insulin resistance than hypercholesterolemia alone 3. Hypertriglyceridemia correlates positively with multiple inflammatory markers, while low HDL shows opposite correlations 3.
7. Adherence to Management Recommendations and Predictors of Non-Compliance
Analyze factors predicting whether patients follow CTCA-based management recommendations, including demographic variables, disease severity, type of recommendation (medical therapy vs invasive evaluation), and whether HeartFlow was utilized. The 2012 ACC/AHA SIHD guidelines emphasize shared decision-making, but real-world adherence data is limited 1. Your dataset uniquely captures both recommendations and actual adherence patterns.
8. Calcium Score Reclassification and Downstream Testing Patterns
Evaluate how calcium scores reclassify patients from initial clinical risk estimates and whether this reclassification appropriately guides subsequent management decisions and invasive angiography referrals. A CACS >100 indicates moderate atherosclerosis requiring aggressive risk factor modification and may reclassify patients from low/intermediate to high risk regardless of traditional factors 4. The 2019 ESC diabetes guidelines give calcium scoring a IIb recommendation as a risk modifier in moderate-risk asymptomatic diabetic patients 1.
9. Multi-Risk Factor Clustering Analysis and Coronary Disease Patterns
Perform cluster analysis to identify distinct patient phenotypes based on combinations of diabetes, hypertension, hyperlipidemia, smoking, and family history, then correlate these clusters with vessel disease patterns, calcium scores, and outcomes. Evidence demonstrates meaningful relationships between aging, diabetes, hypertension and 3-vessel disease, and between hyperlipidemia and single-vessel disease 5. Patients with diabetes, hypertension, and hyperlipidemia show greater potential to develop disease at proximal coronary segments 5.
10. Appropriateness of Invasive Angiography Referral Based on CTCA Findings
Assess the concordance between CTCA-identified disease severity, management recommendations for invasive evaluation, and actual angiographic findings to identify rates of appropriate versus inappropriate catheterization. The CE-MARC 2 trial showed that both CMR and SPECT reduced unnecessary invasive angiography (defined as no stenosis ≥70% or normal FFR) to 7.1-7.5% compared to 28.8% with standard testing 1. Your data on vessel disease severity and actual angiographic outcomes allows calculation of similar appropriateness metrics for CTCA-guided strategies 1.