Instrumentation & Measurement Magazine 25-1 - 67

that respond to the hydrogen peroxide dosage. Two peaks of
the same sign (i.e., positive and positive, or negative and negative)
are positively correlated, while two peaks of opposite
sign (i.e., one positive and the other negative) are anticorrelated.
For example, the 1212 cm−1
peak and 1224 cm−1
peaks are
anticorrelated, meaning that when dosed, one of the peaks will
increase in intensity while the other will decrease. The prominent
peaks in the loadings vector at 1212, 1224, 1375, 1549, 1604
and 1637 cm−1
are consistent with known Raman biomarkers
for oxidation [13], demonstrating the feature extraction capabilities
of PLS-DA. The model was also used to predict the
classes of the test set, achieving an accuracy of about 97%.
Challenges and Future Work
The inherently small differences between the Raman spectra
of pathological and control blood are confounded by many
sources of interference as discussed above. Therefore, in order
for RS of blood to become a viable diagnostic technique,
it is important to develop standardized protocols including:
pre-analytical steps that include sample collection, processing
and storage conditions; measurement steps that optimize
the Raman SNR and SBR, and the spectral dataset size; and analytical
steps that include data preprocessing and application
of chemometrics. In addition, when building a training model,
appropriate methods need to account for the problem of interdonor
variability that arises due to differing concentrations of
blood components such as erythrocytes and leukocytes from
donor to donor.
Most of the current research involving RS of blood utilizes
commercially available confocal Raman microscopes that are
prohibitively expensive, limiting their use to only a few research
labs. Compact or portable Raman setups [18] with a
cost-effective design and ease of use are needed so that RS
of blood will be more accessible to a wider community of researchers
and may open up opportunities for blood-based RS
detection of biomarkers of new diseases or infections. Further,
current techniques of RS of blood are both labor-intensive and
time-consuming, so the Raman data is typically collected from
a small number (~25-30) of independent donors with a total
of ~200-300 spectra. This hampers the accuracy of machine
learning classification methods since they require large data
sets. Automated sampling and increasing data throughput
will permit the collection of large data sets. In addition, techniques
for standardization of spectral datasets need to be more
widely adopted by researchers so that data from multiple research
labs could be combined to conduct large-scale studies.
These important steps are essential for the clinical translation
of RS of blood.
Acknowledgment
We would like to acknowledge the support of the Natural Sciences
and Engineering Research Council of Canada (NSERC)
Discovery Grant (SM) and Undergraduate Summer Research
Assistant funding (CC). We would like to thank Ben Hansson,
Harry Allen, and Connor McNairn at Carleton University, and
Nadine Adam at Health Canada.
February 2022
References
[1] M. J. Baker et al., " Developing and understanding biofluid
vibrational spectroscopy: a critical review, " Chemical Soc. Rev., vol.
45, no. 7, pp. 1803-1818, Nov. 2016.
[2] S. Perakis and M. R. Speicher, " Emerging concepts in liquid
biopsies, " BMC Med., vol. 15, no. 1, pp. 1-12, Apr. 2017.
[3] J. L. Gehrich et al., " Optical fluorescence and its application to an
Intravascular blood gas monitoring system " IEEE Trans. Biomed.
Eng., vol. BME-33, no. 2, pp. 117-132, 1986.
[4] H. Butler et al., " Using Raman spectroscopy to characterize
biological materials, " Nature Protocols, vol. 11, no. 4, pp. 664-687,
Mar. 2016.
[5] K. Kong, C. Kendall, N. Stone, and I. Notingher, " Raman
spectroscopy for medical diagnostics-from in-vitro biofluid
assays to in-vivo cancer detection, " Advanced Drug Delivery
Reviews, vol. 89, pp. 121-134, Jul. 2015.
[6] C. H. Allen et al., " Raman micro-spectroscopy analysis of human
lens epithelial cells exposed to a low-dose-range of ionizing
radiation, " Physics in Med. Biol., vol. 63, no. 2, pp. 025002, Jan. 2018.
[7] D. Mason, S. Murugkar, and A. D. Speirs, " Measurement of
cartilage sub-component distributions through the surface by
Raman spectroscopy-based multivariate analysis, " J. Biophotonics,
vol. 14, no. 1, pp. e202000289, Aug. 2020.
[8] M. Jermyn et al., " Intraoperative brain cancer detection with
Raman spectroscopy in humans, " Sci. Translational Med., vol. 7,
no. 274, pp. 274ra19-274ra19, Feb. 2015.
[9] C. G. Atkins, K. Buckley, M. W. Blades, and R. F. Turner,
" Raman spectroscopy of blood and blood components, " Applied
Spectroscopy, vol. 71, no. 5, pp. 767-793, 2017.
[10] C. R. Kong et al., " A novel non-imaging optics based
Raman spectroscopy device for transdermal blood analyte
measurement, " AIP Advances, vol. 1, no. 3, pp. 032175, Sep. 2011.
[11] R. L. McCreery, " Magnitude of Raman scattering, " in Raman
Spectroscopy for Chemical Analysis, vol. 225, ch. 2, pp. 15-30.
Hoboken, New Jersey, USA: John Wiley and Sons, 2005.
[12] T. Talala et al., " Time-resolved Raman spectrometer with high
fluorescence rejection based on a CMOS SPAD line sensor and a
573-nm pulsed laser, " IEEE Trans. Instrum. Meas., vol. 70, pp. 1-10,
2021.
[13] B. Hansson et al., " Development of a flow cell based Raman
spectroscopy technique to overcome photodegradation in human
blood, " Biomedical Optics Express, vol. 10, no. 5, pp. 2275-2288,
2019.
[14] Y. Zhang, X. Mi, X. Tan, and R. Xiang, " Recent progress on liquid
biopsy analysis using surface-enhanced Raman spectroscopy, "
Theranostics, vol. 9, no. 2, pp. 491, Jan. 2019.
[15] C. L. Morais et al., " Standardization of complex biologically
derived spectrochemical datasets, " Nature Protocols, vol. 14, no. 5,
pp. 1546-1577, Apr. 2019.
[16] J. M. Smulko, N. C. Dingari, J. S. Soares, and I. Barman, " Anatomy
of noise in quantitative biological Raman spectroscopy, "
Bioanalysis, vol. 6, no. 3, pp. 411-421, Jan. 2014.
[17] F. Dallaire et al., " Quantitative spectral quality assessment
technique validated using intraoperative in vivo Raman
spectroscopy measurements, " J. Biomedical Optics, vol 25, no. 4,
pp. 040501, Apr. 2020.
IEEE Instrumentation & Measurement Magazine
67
Instrumentation & Measurement Magazine 25-1

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 25-1
