Objective: The objective of this study was to develop a non-invasive, dual-mode sensing system for the early diagnosis and progression tracking of Parkinson’s disease (PD) by using α- synuclein molecules in saliva as biomarkers and to explore the potential of surface-enhanced Raman scattering (SERS) spectrum technology in the diagnosis of PD.

Background: PD is a common neurodegenerative disorder. Current diagnostic methods for Parkinson’s disease (PD) are deficient in sensitivity and specificity, particularly during the early stages. Additionally, PD symptoms overlap with those of MSA and PSP, leading to misdiagnosis. While α-synuclein is a recognized biomarker for PD, its levels do not correlate with disease severity, limiting its clinical utility. Wearable sensing technologies offer a promising avenue for continuous monitoring of PD-relevant biomarkers. Existing detection methods mostly require invasive procedures, which are not well-tolerated and carry potential risks. Thus, a non-invasive detection method for α-synuclein is urgently needed.

Method: Methods: We developed a dual-mode sensing system integrating surface-enhanced Raman scattering (SERS) for α-synuclein detection in saliva and piezoelectric sensors for monitoring eye movements. The system was assembled in a 3D stacking configuration to enable simultaneous detection of both biomarkers. Saliva samples from 34 PD patients, 7 MSA patients, and 8 PSP patients were analyzed using SERS to identify distinct spectral patterns. Eye movements, including blink rate, horizontal eye scans, and vertical eye scans, were monitored using piezoelectric sensors to quantify PD severity. The multimodal data were analyzed using an artificial intelligence (AI) model to achieve high diagnostic accuracy. Sensor Preparation: A general approach was developed to prepare an SERS spectrum sensor. Chemical etching was used to create silver fibers with extremely high-surface-roughness as the SERS substrate.

Sample Collection and Analysis: Saliva samples were collected in the morning from 30 PD patients, 7 patients with multiple system atrophy (MSA), and 7 patients with progressive supranuclear palsy (PSP) after they had fasted overnight. The samples were centrifuged to remove impurities, and the supernatant was used for SERS analysis.

Data Analysis: An artificial intelligence model was employed to analyze the SERS spectral data. The model was trained using a large dataset of SERS spectra from different disease groups, with machine-learning algorithms such as support vector machines (SVM) and deep neural networks (DNN) used for model construction.

Results: The SERS analysis revealed distinct spectral patterns for α-synuclein in saliva samples from PD, MSA, and PSP patients, enabling differentiation among these conditions. Eye movement metrics, particularly blink rate and eye scan patterns, were strongly correlated with PD severity, providing a quantitative measure of disease progression. The AI model integrating SERS and eye movement data achieved an accuracy of 98.0% in diagnosing PD and distinguishing it from MSA and PSP. This demonstrates the system’s potential as a reliable tool for early diagnosis and progression tracking.

Conclusion: This study presents a groundbreaking, non-invasive approach to PD diagnosis and monitoring by combining α-synuclein detection in saliva with eye movement analysis. This innovative approach has the potential to transform the diagnosis and monitoring of PD, improving patient outcomes and advancing neurodegenerative disease research.

« Back to 2025 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/multimodal-sensing-enabled-diagnosis-and-assessment-of-parkinsons-disease-via-biomolecule-and-biomechanics-signals/