Stress Inference from Abdominal Sounds using Machine Learning.

Stress is often considered the 21st century's epidemic, affecting more than a third of the globe's population. Long-term exposure to stress has significant side effects on physical and mental health. In this work we propose a methodology for detecting stress using abdominal sounds. For this study, eight participants were either exposed to a stressful (Stroop test) or a relaxing (guided meditation) stimulus for ten days. In total, we collected 104 hours of abdominal sounds using a custom wearable device in a belt form-factor. We explored the effect of various features on the binary stress classification accuracy using traditional machine learning methods. Namely, we observed the impact of using acoustic features on their own, as well as in combination with features representing current mood state, and hand-crafted domain-specific features. After feature extraction and reduction, by utilising a multilayer perceptron classifier model we achieved 77% accuracy in detecting abdominal sounds under stress exposure. Clinical relevance- This feasibility study confirms the link between the gastrointestinal system and stress and uncovers a novel approach for stress inference via abdominal sounds using machine learning.

Microparticle-Based Biochemical Sensing Using Optical Coherence Tomography and Deep Learning.

Advancing continuous health monitoring beyond vital signs to biochemistry will revolutionize personalized medicine. Herein, we report a biosensing platform to achieve remote biochemical monitoring using microparticle-based biosensors and optical coherence tomography (OCT). Stimuli-responsive, polymeric microparticles were designed to serve as freely dispersible biorecognition units, wherein binding with a target biochemical induces volumetric changes of the microparticle. Analytical approaches to detect these submicron changes in 3D using OCT were devised by modeling the microparticle as an optical cavity, enabling estimations far below the resolution of the OCT system. As a proof of concept, we demonstrated the 3D spatiotemporal monitoring of glucose-responsive microparticles distributed throughout a tissue mimic in response to dynamically fluctuating levels of glucose. Deep learning was further implemented using 3D convolutional neural networks to automate the vast processing of the continuous stream of three-dimensional time series data, resulting in a robust end-to-end pipeline with immense potential for continuous in vivo biochemical monitoring.

Efficient Coupling of an Antenna-Enhanced nanoLED into an Integrated InP Waveguide.

Increasing power consumption in traditional on-chip metal interconnects has made optical links an attractive alternative. However, such a link is currently missing a fast, efficient, nanoscale light-source. Coupling nanoscale optical emitters to optical antennas has been shown to greatly increase their spontaneous emission rate and efficiency. Such a structure would be an ideal emitter for an on-chip optical link. However, there has never been a demonstration of an antenna-enhanced emitter coupled to a low-loss integrated waveguide. In this Letter we demonstrate an optical antenna-enhanced nanoLED coupled to an integrated InP waveguide. The nanoLEDs are comprised of a nanoridge of InGaAsP coupled to a gold antenna that exhibits a 36× enhanced rate of spontaneous emission. Coupling efficiencies as large as 70% are demonstrated into an integrated waveguide. Directional antennas also demonstrate direction emission down one direction of a waveguide with observed front-to-back ratios as high as 3:1.

Engineering light outcoupling in 2D materials.

When light is incident on 2D transition metal dichalcogenides (TMDCs), it engages in multiple reflections within underlying substrates, producing interferences that lead to enhancement or attenuation of the incoming and outgoing strength of light. Here, we report a simple method to engineer the light outcoupling in semiconducting TMDCs by modulating their dielectric surroundings. We show that by modulating the thicknesses of underlying substrates and capping layers, the interference caused by substrate can significantly enhance the light absorption and emission of WSe2, resulting in a ∼11 times increase in Raman signal and a ∼30 times increase in the photoluminescence (PL) intensity of WSe2. On the basis of the interference model, we also propose a strategy to control the photonic and optoelectronic properties of thin-layer WSe2. This work demonstrates the utilization of outcoupling engineering in 2D materials and offers a new route toward the realization of novel optoelectronic devices, such as 2D LEDs and solar cells.

Optical antenna enhanced spontaneous emission.

Atoms and molecules are too small to act as efficient antennas for their own emission wavelengths. By providing an external optical antenna, the balance can be shifted; spontaneous emission could become faster than stimulated emission, which is handicapped by practically achievable pump intensities. In our experiments, InGaAsP nanorods emitting at ∼ 200 THz optical frequency show a spontaneous emission intensity enhancement of 35 × corresponding to a spontaneous emission rate speedup ∼ 115 ×, for antenna gap spacing, d = 40 nm. Classical antenna theory predicts ∼ 2,500 × spontaneous emission speedup at d ∼ 10 nm, proportional to 1/d(2). Unfortunately, at d < 10 nm, antenna efficiency drops below 50%, owing to optical spreading resistance, exacerbated by the anomalous skin effect (electron surface collisions). Quantum dipole oscillations in the emitter excited state produce an optical ac equivalent circuit current, I(o) = qω|x(o)|/d, feeding the antenna-enhanced spontaneous emission, where q|x(o)| is the dipole matrix element. Despite the quantum-mechanical origin of the drive current, antenna theory makes no reference to the Purcell effect nor to local density of states models. Moreover, plasmonic effects are minor at 200 THz, producing only a small shift of antenna resonance frequency.