Coronary disease & ACSDiagnostic or prognostic study
A Machine Learning Driven Approach to Quantifying Coronary Artery Tortuosity
- development and validation of an automated right coronary artery (RCA) tortuosity measure using machine learning (ML)-based vessel segmentation
Authors' conclusion
We derived a scalable ML-enabled measure of RCA tortuosity from coronary angiography and found associations with sex, hypertension, diabetes, and presence and severity of CAD.
Study context
- Population
- Patients with RCA angiogramspatients, 38,691 RCA angiograms, concordance subset of 300 angiograms
- Participants
- 22,334 patients
- Comparison
- ML-enabled RCA tortuosity measure against blinded interventional cardiologist review; association of tortuosity with age, sex, hypertension, diabetes, hypercholesterolemia, smoking, and presence of coronary artery disease (CAD)
Performance
As reported in the abstract
Tortuosity range
Tortuosity ranged
0.007 to 0.289
Further findings
| Outcome | Estimate | P value |
|---|---|---|
| Tortuosity by sex, hypertension, diabetes — women · higher in women | β per SD 0.17 | P < 0.001 |
| Tortuosity by sex, hypertension, diabetes — hypertension · higher with hypertension | β per SD 0.06 | P = 0.002 |
| Tortuosity by sex, hypertension, diabetes — diabetes · lower with diabetes | β per SD −0.11 | P < 0.001 |
| CAD, severe CAD, Gensini score by tortuosity — Higher tortuosity · CAD · even after adjustment | OR per SD 1.05 | P < 0.001 |
| CAD, severe CAD, Gensini score by tortuosity — Higher tortuosity · severe CAD · even after adjustment | OR per SD 1.09 | P < 0.001 |
| CAD, severe CAD, Gensini score by tortuosity — Higher tortuosity · higher Gensini score · even after adjustment | β per SD 0.05 | P < 0.001 |
Blinded interventional cardiologist classification of high tortuosity vs not
- precision
- 85.6%
Age, hypercholesterolemia, smoking
- After adjustment for risk factors, age was not independently associated with tortuosity
- Hypercholesterolemia and smoking were not associated
Original publication
A Machine Learning Driven Approach to Quantifying Coronary Artery Tortuosity. JACC Adv. doi:10.1016/j.jacadv.2026.102829
Summarised from the published abstract; every number is checked against it