Machine learning identification of risk factors for heart failure in patients with diabetes mellitus with metabolic dysfunction associated steatotic liver disease (MASLD): the Silesia Diabetes-Heart Project
Authors' conclusion
A ML approach demonstrated high performance in identifying HF in patients with DM independently of their MASLD status, as well as both in patients with and without MASLD based on easy-to-obtain patient parameters.
Study context
- Population
- Patients with diabetes mellitus (DM) with and without MASLD, Silesia Diabetes-Heart Project; mean age 58.85 ± SD 17.37 years; 48% women
- Participants
- 2,000 patients
- Outcome
- Identification of HF risk (sensitivity, specificity, % correctly classified high- and low-risk patients)
Methodological detailsmachine learning, multiple logistic regression classifier, χ2 test feature selection with Monte Carlo strategy
- Machine learning
- Multiple logistic regression classifier
- χ2 test feature selection with Monte Carlo strategy
Performance
As reported in the abstract
HF identification by machine learning (ML) model, all patients
- Sensitivity
- 0.81
- Specificity
- 0.70
- AUC
- 0.84 (95% CI 0.82-0.86)
- Selected features, all patients: age, type of DM, atrial fibrillation (AF), hyperuricemia, estimated glomerular filtration rate (eGFR)
- 5
- Selected features, MASLD(+): AF, hyperuricemia, eGFR
- 3
- Selected features, MASLD(-): age, eGFR
- 2
Further findings
Selected heart failure (HF) risk factors by MASLD status
- MASLD(+): AF, hyperuricemia and eGFR
- 3 features
- MASLD(-): age and eGFR
- 2 features
Original publication
Machine learning identification of risk factors for heart failure in patients with diabetes mellitus with metabolic dysfunction associated steatotic liver disease (MASLD): the Silesia Diabetes-Heart Project. Cardiovasc Diabetol. Published 20 November 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →