460
H. M. Heise
ethanol concentration, attributable to both assay accuracy and pharmacokinetics in
the elimination phase of alcohol. With error analysis, a significant fraction of the
concentration variance could be explained by alcohol pharmacokinetics using a first
order kinetic model. It is interesting to note that the PLS calibration models gave
different prediction errors when using either data from the entire monitoring session,
or data from the alcohol elimination phase only; minimum SEP values were for
regressions of finger tissue spectra versus venous blood concentration with a value
of 6.6 mg/dl. As summary, the statement is valid that their work provided a first
investigation of the relationships between breath alcohol, venous blood alcohol and
interstitial tissue alcohol concentration, as measured at multiple skin sites [45].
20.4 NIR Spectroscopy for Tissue Analysis
20.4.1 Applications for Spectral Histopathology
A few tissue NIR spectra have already been presented in previous sections, in particular when skin spectroscopy for non-invasive assays is involved. Here, different skin
areas have been studied such as oral mucosa of the inner lip, finger and forearm skin
sites and most spectra, apart from a few transmission measurements, were presented
as measured with fiber-optic and mirror-based accessories in diffuse reflection mode.
Muscle and fatty tissue spectra have also been introduced and it is shown for all that
water features are dominating the spectra. NIR microscopic studies are rare; see,
for example, Ref. [32], where the authors were studying the heterogeneity of skin
with its impact on non-invasive glucose measurements by skin spectroscopy. The
histopathology of biopsy tissues on the microscopic level is a domain of mid-infrared
and Raman spectroscopy, where imaging applications are supporting the pathologist
on tissue assessment after, for example, a colonoscopy or bronchoscopy using endoscopic techniques. However, at the end of this section, such a NIR spectroscopic
application with an inverted microscope is discussed. Larger tissue volumes can be
easily accessed by fiber-optic probes attached to FT-NIR or dispersive spectrometers as described in one of the previous subchapters. For spectroscopy, the tissues or
organs must be accessible, although also endoscopic techniques have been described
in the past with more sophisticated instrumentation.
There have been different tissues studied for cancer, which comprise breast
tumors, cervical dysplasia and cancer, glioma, human melanoma xenografts, head
and neck tumors or cancerous bronchial mucosa; for the literature collection, see our
review on NIR spectroscopy in cancer diagnosis and therapy [46]. Further studies
have been performed earlier on pancreatic and colorectal tissues by my group, which
has also been covered by our review. For dermatological studies on different skin
cancers, the paper by McIntosh et al. should be consulted [47]. When looking at the
physiological relevance of NIR spectroscopy for cancer studies, one has to look at
H. M. Heise
ethanol concentration, attributable to both assay accuracy and pharmacokinetics in
the elimination phase of alcohol. With error analysis, a significant fraction of the
concentration variance could be explained by alcohol pharmacokinetics using a first
order kinetic model. It is interesting to note that the PLS calibration models gave
different prediction errors when using either data from the entire monitoring session,
or data from the alcohol elimination phase only; minimum SEP values were for
regressions of finger tissue spectra versus venous blood concentration with a value
of 6.6 mg/dl. As summary, the statement is valid that their work provided a first
investigation of the relationships between breath alcohol, venous blood alcohol and
interstitial tissue alcohol concentration, as measured at multiple skin sites [45].
20.4 NIR Spectroscopy for Tissue Analysis
20.4.1 Applications for Spectral Histopathology
A few tissue NIR spectra have already been presented in previous sections, in particular when skin spectroscopy for non-invasive assays is involved. Here, different skin
areas have been studied such as oral mucosa of the inner lip, finger and forearm skin
sites and most spectra, apart from a few transmission measurements, were presented
as measured with fiber-optic and mirror-based accessories in diffuse reflection mode.
Muscle and fatty tissue spectra have also been introduced and it is shown for all that
water features are dominating the spectra. NIR microscopic studies are rare; see,
for example, Ref. [32], where the authors were studying the heterogeneity of skin
with its impact on non-invasive glucose measurements by skin spectroscopy. The
histopathology of biopsy tissues on the microscopic level is a domain of mid-infrared
and Raman spectroscopy, where imaging applications are supporting the pathologist
on tissue assessment after, for example, a colonoscopy or bronchoscopy using endoscopic techniques. However, at the end of this section, such a NIR spectroscopic
application with an inverted microscope is discussed. Larger tissue volumes can be
easily accessed by fiber-optic probes attached to FT-NIR or dispersive spectrometers as described in one of the previous subchapters. For spectroscopy, the tissues or
organs must be accessible, although also endoscopic techniques have been described
in the past with more sophisticated instrumentation.
There have been different tissues studied for cancer, which comprise breast
tumors, cervical dysplasia and cancer, glioma, human melanoma xenografts, head
and neck tumors or cancerous bronchial mucosa; for the literature collection, see our
review on NIR spectroscopy in cancer diagnosis and therapy [46]. Further studies
have been performed earlier on pancreatic and colorectal tissues by my group, which
has also been covered by our review. For dermatological studies on different skin
cancers, the paper by McIntosh et al. should be consulted [47]. When looking at the
physiological relevance of NIR spectroscopy for cancer studies, one has to look at