Non-Invasive Diagnosis of Moyamoya Disease Using Serum Metabolic Fingerprints and Machine Learning

Abstract

Moyamoya disease (MMD) is a progressive cerebrovascular disorder that increases the risk of intracranial ischemia and hemorrhage. Timely diagnosis and intervention can significantly reduce the risk of new-onset stroke in patients with MMD. However, the current diagnostic methods are invasive and expensive, and non-invasive diagnosis using biomarkers of MMD is rarely reported. To address this issue, nanoparticle-enhanced laser desorption/ionization mass spectrometry (LDI MS) was employed to record serum metabolic fingerprints (SMFs) with the aim of establishing a non-invasive diagnosis method for MMD. Subsequently, a diagnostic model was developed based on deep learning algorithms, which exhibited high accuracy in differentiating the MMD group from the HC group (AUC = 0.958, 95% CI of 0.911 to 1.000). Additionally, hierarchical clustering analysis revealed a significant association between SMFs across different groups and vascular cognitive impairment in MMD. This approach holds promise as a novel and intuitive diagnostic method for MMD. Furthermore, the study may have broader implications for the diagnosis of other neurological disorders.

Keywords: biomarkers; fingerprints; mass spectrometry; moyamoya disease diagnosis.