A. Zhan, N. Yellai, A. Majumdar
Tunoptix, Inc.,
United States
Keywords: meta-optics, computational imaging, spectral imaging, spectroscopy, low-SWaP, dual-use
Summary:
Spectral sensing provides access to rich, material-specific information far beyond what conventional RGB imaging can offer. This enables breakthrough applications across healthcare, environmental monitoring, industrial inspection, and defense. However, traditional hyperspectral and multispectral systems are often bulky, costly, and power-hungry. These systems typically rely on moving parts, tunable filters, or dispersive optics, which limit their utility in compact or embedded platforms. Tunoptix has developed an ultra-compact spectral sensing platform using a fully passive, all-meta-optic architecture. This novel system integrates arrays of meta-lenses and meta-filters – nanoscale optical structures engineered to manipulate light with extreme precision – directly onto standard CMOS sensors. Each optical channel samples a different spectral band, enabling true snapshot acquisition of narrowband spectral data without scanning, filter wheels, or mosaics. With a system volume under 1 cm³, Tunoptix’s platform captures a broad range of spectral channels across the visible to near-infrared (VNIR) range with narrowband resolution. The system is passive and contains no moving parts, supporting real-time imaging and low-power operation. These features make it highly suitable for edge-constrained environments such as handheld diagnostics, mobile sensors, small UAVs, wearables, and smart devices. Unlike conventional systems requiring complex assemblies or tiled filters, Tunoptix employs wafer-scale fabrication to enable scalable, cost-efficient manufacturing. The architecture is modular and compatible with standard semiconductor workflows, allowing rapid customization for various wavelength ranges and application-specific requirements. In addition to hardware innovation, Tunoptix has developed lightweight computational imaging and machine learning pipelines tailored to operate directly on captured spectral data. These pipelines enable real-time material classification, anomaly detection, and health diagnostics at the edge. Key application areas include: • Consumer Electronics: Integrate compact modules into smartphones, wearables, and IoT devices. Applications include chemical signature detection, enhanced color rendering, and health safety diagnostics. • Personalized Healthcare: Use spectral sensing to monitor skin hydration, melanin, and blood oxygen levels for skincare; track oral health with smart toothbrushes; and monitor cardiovascular health via photoplethysmography (PPG). • Agriculture & Food Quality: Real-time detection of ripeness and freshness (e.g., avocados), and early-stage crop monitoring and pest detection using spectral analysis. • Industrial & Defense: Perform compact on-site material analysis for QA/QC, integrate into AR/VR and machine vision systems, and support low-SWaP defense tools for chemical detection or camouflaged object recognition. Tunoptix is also advancing this platform into compact spectroscopy domain by adapting its meta-optic architecture for high-resolution spectral analysis. Drawing on principles from Raman and NIR spectroscopy, the system can capture noninvasive, real-time spectral data in a miniaturized package suitable for wearables, portable diagnostics, and environmental sensing – where traditional lab instruments are too large or expensive. This submission to the TechConnect World Innovation Conference showcases a scalable, commercially viable sensing platform with strong dual-use potential. Tunoptix welcomes collaboration with industry and government partners to drive pilot demonstrations and next-generation product development. By eliminating the barriers of size, cost, and complexity, Tunoptix enables the broad adoption of spectral sensing in consumer and mission-critical technologies.