AI may detect CVD with the help of gut microbiome bacteria
Click Here to Manage Email Alerts
Researchers have developed a machine learning model using gut microbiome data sets for the diagnostic screening of CVD, according to findings presented at the American Heart Association Hypertension Scientific Sessions.
“Based on our previous research linking gut microbiota to CVD in animal models, we designed this study to test whether it is possible to screen for CVD in humans using artificial intelligence screening of stool samples,” Bina Joe, PhD, FAHA, professor and chairwoman of the department of physiology and pharmacology at the University of Toledo, said in a press release. “Gut microbiota has a profound effect on cardiovascular function, and this could be a potential new strategy for evaluation of cardiovascular health.”
For this research, which was simultaneously published in Hypertension, investigators utilized 16S rRNA reads from stool samples taken for the American Gut Project and identified 478 participants with CVD and 473 without. Researchers evaluated the diagnostic efficacy of a machine learning model using five algorithms that included random forest, support vector machine with radial kernel, decision tree, elastic net and neural networks.
Researchers reported the machine learning model differentiated 39 bacterial taxa among participants with and without CVD.
Utilizing these differential taxonomic features, the AI achieved an area under the receiver operating characteristics curve of approximately 0.58.
The AUC increased to approximately 0.65 for the top 500 high-variance features of operational taxonomic units for training machine learning models, the researchers reported.
Lastly, to limit dimensionality of feature space, investigators narrowed the model to 25 highly contributing operational taxonomic units, which further increased the AUC to approximately 0.7.
“Despite the fact that gut microbiomes are highly variable among individuals, we were surprised by the promising level of accuracy obtained from these preliminary results, which indicate fecal microbiota composition could potentially serve as a convenient diagnostic screening method for CVD,” Joe said in the release. “It is conceivable that one day, maybe without even assessing detailed cardiovascular function, clinicians could analyze the gut microbiome of patients’ stool samples with an artificial machine learning method to screen patients for heart and vascular diseases.”
Reference:
Collapse
Aryal S, et al. Presentation MP10. Presented at: American Heart Association Hypertension Scientific Sessions; Sept. 10-13, 2020 (virtual meeting).
Click Here to Manage Email Alerts