Biomedical Engineering Reference
In-Depth Information
overlying nontumour-bearing skin in 98.0% of cases. Spectra of basal cell
carcinoma, squamous cell carcinoma, nevi, and malignant melanoma were
qualitatively similar. Distinction of basal cell carcinoma, squamous cell car-
cinoma, and melanocytic lesions by linear discriminant analyses, however,
was 93.5% accurate. Therefore, spectral separation of abnormal versus nor-
mal tissue was achieved with high sensitivity and specificity [92].
Barry et al. recorded FT Raman and infrared spectra of the outermost layer of
human skin, the stratum corneum. Assignments consistent with the FT Raman
vibrations were made for the first time and compared with assignments from
the FTIR spectrum. The results demonstrated that FT Raman spectroscopy
holds several advantages over FTIR in studies of human skin. The molecular
and conformational nature of human skin, and modifications induced by drug
or chemical treatments, may be assessed by FT Raman spectroscopy [93].
Huang et al. used a combination of Raman spectroscopy with near-infrared
autofluorescence for assessment of skin fibrosarcoma in mice. A murine tumour
was implanted into the subcutaneous region of the lower back. Diagnostic
algorithms were developed for differentiating tumours from normal tissue
based on the spectral features. Thirty-two in vivo Raman, NIR fluorescence,
and composite Raman and NIR fluorescence were analysed (16 normal, 16
tumours). Classification results showed diagnostic sensitivities of 81.3%, 93.8%,
and 93.8%; specificities of 100%, 87.5%, and 100%; and overall diagnostic accu-
racies of 90.6%, 90.6%, and 96.9%, respectively for tumour identification [94].
A study on confocal Raman microspectroscopy as a noninvasive in vivo
optical method to measure molecular concentration profiles in the skin was
carried out by Caspers et al. It was shown that the technique can be applied
to determine the water concentration in the stratum corneum as a function
of distance to the skin surface, with a depth resolution of 5 µm. The result-
ing in vivo concentration profiles were in qualitative and quantitative agree-
ment with published data. No other noninvasive in vivo technique exists
that analyses skin molecular compositions as a function of distance to the
skin surface with similar detail and spatial resolution [95].
Ly et al. applied FTIR spectral imaging on formalin-fixed paraffin-
embedded biopsies from colon and skin cancerous lesions [96]. The samples
were fixed in formalin and then embedded in paraffin. This technique can
preserve molecular structures and is the gold standard in tissue storage.
However, paraffin absorption bands are significant in the mid-infrared
region and can mask some molecular vibrations of the tissue. Then direct
data processing was applied on spectral images without any chemical
dewaxing of the tissues. The signal of paraffin was modelled and paraffin
spectra were removed from the raw images based on an outlier detection.
Afterwards, pseudo-colour images were computed by K-means clustering
in order to highlight histological structures of interest. Using this method,
tumour areas were successfully demarcated in both types of tissues.
Lieber et al. reported on near-infrared Raman microspectroscopy for
in vitro detection of skin cancer [97]. Thirty-nine skin tissue samples
Search WWH ::
Custom Search