Hazardous liquid detection by active hyperspectral imaging

Howle, Christopher R.; Frisby, Alexander; McIntosh, Alastair J. S.; Stothard, David J. M.; Dunn, Malcolm H.; Robertson, Gordon; Miller, W. David; Malcolm, Graeme P. A.; Maker, Gareth T.

doi:10.1117/12.898516

Cited by 3 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hyperspectral images of the 12-well plates for the trauma study were collected using a prototype negative contrast imaging device (NCI) [ 28 ]. This NCI device, developed by M Squared Lasers (Glasgow, UK), is a reflectance imaging device that collects hyperspectral images from illumination in both the short and mid-wave infrared regions (SWIR/MWIR) and has been used to identify spectral differences between wound biopsies, with seven human samples being imaged to predict their wound healing outcome [ 29 ].…”

Section: Methodsmentioning

confidence: 99%

Machine learning utilising spectral derivative data improves cellular health classification through hyperspectral infra-red spectroscopy

Mellors¹,

Spear²,

Howle³

et al. 2020

PLoS ONE

View full text Add to dashboard Cite

The objective differentiation of facets of cellular metabolism is important for several clinical applications, including accurate definition of tumour boundaries and targeted wound debridement. To this end, spectral biomarkers to differentiate live and necrotic/apoptotic cells have been defined using in vitro methods. The delineation of different cellular states using spectroscopic methods is difficult due to the complex nature of these biological processes. Sophisticated, objective classification methods will therefore be important for such differentiation. In this study, spectral data from healthy/traumatised cell samples using hyperspectral imaging between 2500-3500 nm were collected using a portable prototype device. Machine learning algorithms, in the form of clustering, have been performed on a variety of pre-processing data types including 'raw' unprocessed, smoothed resampling, background subtracted and spectral derivative. The resulting clusters were utilised as a diagnostic tool for the assessment of cellular health and quantified using both sensitivity and specificity to compare the different analysis methods. The raw data exhibited differences for one of the three different trauma types applied, although unable to accurately cluster all the traumatised samples due to signal contamination from the chemical insult. The background subtracted and smoothed data sets reduced the accuracy further, due to the apparent removal of key spectral features which exhibit cellular health. However, the spectral derivative data-types significantly improved the accuracy of clustering compared to other data types, with both sensitivity and specificity for the background subtracted data set being >94% highlighting its utility to account for unknown signal contamination while maintaining important cellular spectral features.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine learning utilising spectral derivative data improves cellular health classification through hyperspectral infra-red spectroscopy

Mellors¹,

Spear²,

Howle³

et al. 2020

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Two principal subsystems make up the NCI: a broadly tunable intracavity optical parametric oscillator (ICOPO) that serves as the active illumination source and a galvanometer-based raster scanning system for image acquisition. [6][7][8][9] The intracavity approach confers two key advantages. First, the circulating power inside the cavity is much higher than outside, thereby reducing the pump-laser power requirements and those of the overall system.…”

Section: Figure 2 Thiodiglycol (Tdg) a Chemical Warfare Agent Simulmentioning

confidence: 99%

Imager locates toxic liquids at stand-off distances

Howle¹,

Clewes²,

Guicheteau³

et al. 2013

SPIE Newsroom

Self Cite

View full text Add to dashboard Cite

show abstract

“…Spectral imaging is a technique used for in-situ material classification for a variety of applications [1][2][3]. Traditional hyperspectral analysis requires recording a vast amount of data, acquired over the course of many measurements, generated by scanning through one or two dimensions of the spectral datacube-the optical data consisting of two spatial dimensions and one spectral dimension.…”

Section: Introductionmentioning

confidence: 99%

Experimental Validation of the Adaptive Feature-Specific Spectral Imaging Classifier (AFSSI-C)

Dunlop-Gray

Poon

Vera

et al. 2014

Frontiers in Optics 2014

View full text Add to dashboard Cite

The AFSSI-C is a spectral imager that generates spectral classification directly, in fewer measurements than are required by traditional systems that measure the spectral datacube (which is later interpreted to make material classification). By utilizing adaptive features to constantly update conditional probabilities for the different hypotheses, the AFSSI-C avoids the overhead of directly measuring every element in the spectral datacube. The system architecture, feature design methodology, simulation results, and preliminary experimental results are given.

show abstract

Hazardous liquid detection by active hyperspectral imaging

Cited by 3 publications

References 15 publications

Machine learning utilising spectral derivative data improves cellular health classification through hyperspectral infra-red spectroscopy

Machine learning utilising spectral derivative data improves cellular health classification through hyperspectral infra-red spectroscopy

Imager locates toxic liquids at stand-off distances

Experimental Validation of the Adaptive Feature-Specific Spectral Imaging Classifier (AFSSI-C)

Contact Info

Product

Resources

About