Ambient-Light-Immune NIR Photodetector Based on WS₂/SiHMs Heterojunction for Hypertension Taxonomy Via CNN Integration.

Abstract

Hypertension is a leading cause of death globally and a critical risk factor for serious diseases. However, continuous monitoring and accurate acquisition of blood pressure (BP) values using photoplethysmography (PPG) techniques face significant challenges in daily life. High-fidelity PPG waveform acquisition is particularly susceptible to ambient light interference. Herein, an in situ fabricated ambient-light-immune WS₂/Si hexagonal micro-holes (WS₂/SiHM) heterojunction photodetector is developed by pulsed laser deposition (PLD) method. This device leverages the unique metallic phase characteristics of PLD-1T-WS₂. The heterojunction exhibits exceptional performance, including a broad near-infrared (NIR) spectral response (700-1550 nm) and remarkable visible-light rejection, achieving a spectral rejection ratio > 10³. These properties, coupled with the custom-designed circuit, enable robust acquisition of PPG waveforms even under challenging ambient light conditions. Comparative evaluations against commercial broadband silicon photodiodes demonstrate superior performance in ambient light suppression and heart rate estimation accuracy. Furthermore, when integrated with a convolutional neural network, the system achieves hypertension classification accuracy exceeding 90%, highlighting its promising application in hypertension risk assessment. This work presents a new paradigm for ambient-light-immune NIR photodetector technology in wearable health monitoring, offering a viable solution to the longstanding challenges in continuous, non-invasive cardiovascular assessment.