Optical imaging spectroscopy for rapid, primary screening of SARS‑CoV‑2: a proof of concept
Sci Rep. 2021 Aug 10;11(1):16201. doi: 10.1038/s41598-021-95756-3.
Screening and diagnostic tests
Efective testing is essential to control the coronavirus disease 2019 (COVID-19) transmission. Here we report a-proof-of-concept study on hyperspectral image analysis in the visible and near-infrared range for primary screening at the point-of-care of SARS-CoV-2. We apply spectral feature descriptors, partial least square-discriminant analysis, and artifcial intelligence to extract information from optical difuse refectance measurements from 5 µL fuid samples at pixel, droplet, and patient levels. We discern preparations of engineered lentiviral particles pseudotyped with the spike protein of the SARS-CoV-2 from those with the G protein of the vesicular stomatitis virus in saline solution and artifcial saliva. We report a quantitative analysis of 72 samples of nasopharyngeal exudate in a range of SARS-CoV-2 viral loads, and a descriptive study of another 32 fresh human saliva samples. Sensitivity for classifcation of exudates was 100% with peak specifcity of 87.5% for discernment from PCR-negative but symptomatic cases. Proposed technology is reagent-free, fast, and scalable, and could substantially reduce the number of molecular tests currently required for COVID-19 mass screening strategies even in resource-limited settings.
Optical technologies for viral detection.
Optical spectroscopic methods have been developed to detect viruses in vegetal structures21–23 and in human samples, mostly in blood. Tey involve polarimetric and fuorescence spectroscopy, and diferent implementations of Raman spectroscopy for identifcation of Dengue virus24, hepatitis B and C viruses25, and microfuidic devices to recognize avian infuenza A and other respiratory infections26. Further approaches propose the use of nanomaterials targeted to specifc viral antibodies to enhance the potentialities of optical spectroscopy to detect the human immunodefciency virus27. Simpler experimental set-ups pose great interest to develop easy-to-implement testing units suitable for use at the point-of-care, and difuse refectance spectroscopy has been applied to discern mosquitos fed with human blood containing Zika virus28 from controls, although signals from potential alterations in tissues and structures due to the infection remain to be discerned from those arising from the virus itself.
Unmet needs in the COVID‑19 pandemic.
While there are available diagnostic tests to determine if an individual has an active COVID-19 infection, worldwide spread of the virus and its variants—outpacing most public health measures—shows that there remains an urgent need of easily deployable screening tests to perform mass, repetitive (time seriated) checks to detect asymptomatic infectious individuals. Of particular importance is the identifcation of those subjects at initial stages—before the onset of symptoms—and the so-called ‘supercarriers’, asymptomatic individuals with a very high viral load, potential ‘super-spreaders’ of the disease39. From national to local and community levels, many types of COVID-19 screening programs are being implemented, even using tests authorized under emergency approvals (or not approved), with varying levels of success. Criteria for this type of extensive testing include high sensitivity and rapid turnaround time, and the (sometimes difcult) availability of authorized molecular tests for confrmation of positive and ‘concerning negative’ cases40. In addition, virus mutations generate diferent strains which may modify its ability to spread, the severity of related diseases and the performance of public health measures. Unfortunately, by December 2021 there are not any tests authorized (by the United States Food and Drug Administration) to detect specifc SARS-CoV-2 variants14, not even for those categorized by the World Health Organization as ‘variants of concern’ (Alpha, Beta, Gamma, Delta, and Omicron) of the disease41. Te current COVID-19 pandemic has underlined the signifcance of searching for easy-to-implement testing tools potentially useful for real-life applications, particularly at the point-of-care and in constrained resource settings.
See this image and copyright information in PMC
Figure 1 Block diagram of proposed processing platform for classification (positive or negative) of fluid samples with lentiviral particles and quantification of their viral load. Fluid samples are deposited on a supporting plate and imaged (as liquid droplets and dry residues) in the visible and near-infrared ranges using a hyperspectral camera. Optical diffuse reflectance spectra are obtained and processed from individual pixels and averaged. Spectral (wavelength) fringes of interest are determined for the calculation of 29 morphological descriptors of features of individual spectra (F1–F28) and the area ratio (AR) with respect to the spectrum of the background. Two independent classifiers are constructed and evaluated –using the same training and test sample groups– at the pixel and sample (droplet) levels: a partial least square discriminant analysis (PLS-DA) of individual and droplet-averaged spectra and an artificial feed-forward neural network (FFNN) built on the values of the descriptors of features of individual spectra.
