Identification of cerebral infarction using bilateral photoplethysmography

Abstract

Background

Cerebral infarction is often associated with underlying cerebral vascular stenosis, such as carotid artery stenosis or cerebral artery stenosis due to arteriosclerosis. Existing imaging techniques, including carotid ultrasound, computed tomography angiography (CTA), and magnetic resonance angiography (MRA), are useful for diagnosis, but have limitations such as radiation exposure, contrast medium use side effects, and high cost. Therefore, the need for a simple, noninvasive, and cost-effective screening tool is emerging.

Objective

In this study, we propose a novel cerebral infarction screening technique using PPG signals measured from both index fingers for 120 s.

Methods

PPG is a noninvasive optical technology that measures pulse waves that appear according to changes in blood volume. The collected waveforms were divided into windows and then normalized. Maximum Positive Amplitude (MPA) and Maximum Negative Amplitude (MNA) were extracted from each section, and the normal group and cerebral infarction patients were classified through linear discriminant analysis.

Results

As a result of analyzing a total of 100 subjects (50 patients with cerebral infarction and 50 normal controls), the recognition rate based on MNA was 84%, MPA was 81%, and when the two indices were combined, it was 80%. Sensitivity was 80% for MNA and 72% for MPA, and specificity was 88% and 90%, respectively, suggesting that amplitude-based PPG indices can effectively reflect the presence or absence of cerebrovascular lesions.

Conclusion

This study suggests the possibility of simply identifying patients with cerebral infarction by analyzing PPG signals of both fingers. The proposed technique can be used as a screening tool to complement existing imaging techniques, and is expected to contribute to reducing the burden of stroke through early diagnosis and preventive intervention in the future.