Integrating untargeted metabolomics and deep learning approaches to identify specific metabolic signatures and new mechanisms in unstable plaques - Report - MDSpire
Advertisement
Integrating untargeted metabolomics and deep learning approaches to identify specific metabolic signatures and new mechanisms in unstable plaques
Clinical Report: Combining Untargeted Metabolomics with Deep Learning to Uncover Metabolic Profiles and Mechanisms in Unstable Carotid Plaques
Overview
This study identifies 98 metabolites significantly associated with unstable carotid plaques, which are critical risk factors for ischemic stroke. Machine learning algorithms were employed to predict metabolic biomarkers that may enhance stroke risk assessment.
Background
Unstable carotid artery plaques pose a significant risk for ischemic stroke and are often difficult to detect early. Understanding the metabolic changes associated with these plaques can provide insights into their instability and potential for rupture. This research aims to leverage metabolomics and machine learning to improve the identification of biomarkers for stroke risk associated with unstable plaques.
Data Highlights
Metabolite
Association
cGMP-PKG signaling pathway
Significantly associated with unstable plaques
Glucagon signaling pathway
Significantly associated with unstable plaques
Central carbon metabolism in cancer
Significantly associated with unstable plaques
Lipolysis regulation in adipocytes
Significantly associated with unstable plaques
Key Findings
Identified 98 metabolites significantly associated with unstable carotid plaques.
Utilized four machine learning algorithms (RF, SVM, LASSO, LR) for feature analysis.
Highlighted metabolic pathways including cGMP-PKG and glucagon signaling pathways.
Developed potential metabolic biomarkers for predicting stroke risk.
Demonstrated the utility of metabolomics in understanding plaque instability.
Clinical Implications
The identification of specific metabolic biomarkers associated with unstable carotid plaques can enhance the early detection of stroke risk. Clinicians may consider integrating metabolomic profiling into routine assessments for patients at risk of cerebrovascular events.
Conclusion
This study underscores the importance of metabolic profiling in understanding unstable carotid plaques and offers potential biomarkers that could improve stroke risk prediction. Further research is needed to validate these findings in clinical settings.
