Development and application of an optimised Bayesian shrinkage prior for spectroscopic biomedical diagnostics.

BACKGROUND AND OBJECTIVE: Classification of vibrational spectra is often challenging for biological substances containing similar molecular bonds, interfering with spectral outputs. To address this, various approaches are widely studied. However, whilst providing powerful estimations, these techniques are computationally extensive and frequently overfit the data. Shrinkage priors, which favour models with relatively few predictor variables, are often applied in Bayesian penalisation techniques to avoid overfitting. METHODS: Using the logit-normal continuous analogue of the spike-and-slab (LN-CASS) as the shrinkage prior and modelling, we have established classification for accurate analysis, with the established system found to be faster than conventional least absolute shrinkage and selection operator, horseshoe or spike-and-slab. These were examined versus coefficient data based on a linear regression model and vibrational spectra produced via density functional theory calculations. Then applied to Raman spectra from saliva to classify the sample sex. RESULTS: Subsequently applied to the acquired spectra from saliva, the evaluated models exhibited high accuracy (AUC>90 %) even when number of parameters was higher than the number of observations. Analyses of spectra for all Bayesian models yielded high-classification accuracy upon cross-validation. Further, for saliva sensing, LN-CASS was found to be the only classifier with 100 %-accuracy in predicting the output based on a leave-one-out cross validation. CONCLUSIONS: With potential applications in aiding diagnosis from small spectroscopic datasets and are compatible with a range of spectroscopic data formats. As seen with the classification of IR and Raman spectra. These results are highly promising for emerging developments of spectroscopic platforms for biomedical diagnostic sensing systems.

Low-cost hyper-spectral imaging system using a linear variable bandpass filter for agritech applications.

Hyperspectral imaging for agricultural applications provides a solution for non-destructive, large-area crop monitoring. However, current products are bulky and expensive due to complicated optics and electronics. A linear variable filter was developed for implementation into a prototype hyperspectral imaging camera that demonstrates good spectral performance between 450 and 900 nm. Equipped with a feature extraction and classification algorithm, the proposed system can be used to determine potato plant health with ∼88% accuracy. This algorithm was also capable of species identification and is demonstrated as being capable of differentiating between rocket, lettuce, and spinach. Results are promising for an entry-level, low-cost hyperspectral imaging solution for agriculture applications.

Durable infrared optical coatings based on pulsed DC-sputtering of hydrogenated amorphous carbon (a-C:H).

Optical properties of low-temperature pulsed DC-sputter deposited ($ {\le} {70° {\rm C}}$≤70°C) hydrogenated carbon are presented. Increasing hydrogen incorporation into the sputter deposited carbon significantly decreases infrared optical absorption due to a decrease in deep absorptive states associated with dangling bonds. Hydrogen flow is optimized (hydrogen flow 3 sccm), achieving the best compromise between increased infrared transmittance and hardness for durable coating performance. Optical, environmental, and durability performance of pulsed DC-sputtered carbon incorporated in multilayer (a-C:H/Ge) infrared antireflective coatings indicates suitability as a durable infrared optical coating for commonly used infrared substrates, including temperature sensitive chalcogenide glass.

Modeling and validation of uniform large-area optical coating deposition on a rotating drum using microwave plasma reactive sputtering.

Magnetron sputter deposition onto a rotating drum is a method applied to high-throughput large-area optical coating deposition, where film physical thickness uniformity is an important parameter. Techniques have been developed, such as masking/substrate movement, in order to improve sputtered film uniformity. In this study, a model is described and validated for predicting film uniformity. Experimental data show excellent agreement with modeled simulations, with and without a modified sputtering mask. Practical application is demonstrated in maximizing uniformity over an individual substrate size of 100 cm2 for a high-optical-density visible/near-infrared dual-band laser protection filter.

Reactive dynamics analysis of critical Nb2O5 sputtering rate for drum-based metal-like deposition.

Drum-based metal-like film deposition for oxide was investigated using single wavelength in situ monitoring. The data were used to investigate the oxidation mechanism using combined second-order kinetic and parabolic models. A critical Nb2O5 deposition rate of 0.507 nm/s was found at drum rotation of 1 rev/s. However, Nb2O5 samples prepared at varying deposition rates showed that the deposition rate must be much lower than the critical deposition rate to achieve reasonable absorption. Thus simulation for the volume-fraction of metal in the oxide layer was done using effective medium approximation and a distribution function. Simulation gave high agreement with experimental results and allows the prediction of extinction coefficients at various deposition rates.