The following is a summary of “Detection of late gadolinium enhancement in patients with hypertrophic cardiomyopathy using machine learning,” published in the December 2024 issue of Cardiology by Akita et al.
Late gadolinium enhancement (LGE) on cardiac magnetic resonance (CMR) imaging in individuals with hypertrophic cardiomyopathy (HCM) often indicates myocardial fibrosis, which can lead to fatal arrhythmias. However, CMR is resource-intensive and may be contraindicated in some individuals.
Researchers conducted a retrospective study to develop a machine learning (ML) model to detect LGE in individuals with HCM using clinical parameters.
They used a ridge classification method to build an ML model from 22 clinical parameters, including 9 echocardiographic data points. Data from 742 individuals, training set: n=554, test set: n=188 were analyzed, with LGE detected in 299 individuals (54%) in the training set and 76 individuals (40%) in the test set.
The results showed that the area under the receiver-operating-characteristic curve (AUC) of the ML model in the test set was 0.77 (95% CI 0.70–0.84). The ML model outperformed a reference model constructed with 3 conventional risk factors for LGE (AUC 0.69 [95% CI 0.61–0.77]) with a DeLong’s test P value of 0.01.
They concluded that the ML model can effectively detect LGE in individuals with HCM and may help healthcare providers identify individuals with a high pre-test probability for LGE, improving the utility of CMR.
Source: internationaljournalofcardiology.com/article/S0167-5273(24)01533-X/abstra
The post Machine Learning Model for Detecting LGE in Hypertrophic Cardiomyopathy first appeared on Physician's Weekly.