SAT-535 - Quantum Dot-Encoded Filter Array for Wearable Non-Invasive Blood Glucose Monitoring

Continuous glucose monitoring is essential for diabetes management; however, current systems based on interstitial fluid are limited by physiological lag and reduced real-time accuracy. Near-infrared spectroscopy enables direct, non-invasive glucose sensing in blood, but conventional systems typically require bulky instrumentation, limiting wearable applications. We aimed to evaluate whether a quantum dot-encoded narrow-band filter array could enable miniaturized, wearable-compatible, non-invasive glucose monitoring. This study was designed as an exploratory proof-of-concept investigation. A multi-channel spectral encoding matrix was constructed using lead sulfide quantum dots with tunable narrow-band absorption (full width at half maximum < 25 nm) to target glucose-associated near-infrared absorption features around 930 nm. The quantum dot filter array transformed tissue diffuse reflectance into a discrete intensity matrix for compact optical acquisition. Spectral reconstruction and glucose signal extraction were performed using a hybrid machine learning framework integrating partial least squares regression and stacked autoencoders. Preliminary in vitro validation was conducted in a simulated complex biological environment. To support analytical rigor, standardized experimental conditions, model-based signal reconstruction, and quantitative performance metrics were applied throughout system evaluation. A prototype 5-channel lead sulfide quantum dot narrow-band filter array was successfully fabricated and validated. In preliminary in vitro experiments simulating biologically complex conditions, the system demonstrated strong predictive performance for glucose quantification, achieving an R-squared of 0.973 and a root mean square error of 0.57 mmol/L. These findings support the feasibility of accurate glucose estimation using discrete spectral encoding and data-driven signal decoding in a miniaturized sensing format. This exploratory study demonstrates that a quantum dot-encoded filter array can support compact, high-accuracy spectral sensing for non-invasive glucose monitoring. By integrating nanomaterial-based optical encoding with machine-learning-based decoding, this platform may provide a promising pathway toward next-generation wearable continuous glucose monitoring systems with real-time capability.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*