High spatial resolution imaging of melanoma tissue by femtosecond laser-induced breakdown spectroscopy

The classification and imaging of melanoma embedded in murine skin tissue by femtosecond laser-induced breakdown spectroscopy (fs-LIBS) is reported. The tissue samples were collected from hairless mice with induced melanoma, cryosectioned to 50 mu m thickness, and analyzed with no other sample treatment. LIBS measurement was carried out using an ultraviolet (UV) fs-laser (lambda = 343 nm, tau = 550 fs) and intensified charge coupled device detector. The support vector machine (SVM) classifier is customized to distinguish melanoma and dermis with dominant peaks and their intensity ratios, of which normal distribution are obtained by maximum likelihood estimation. Based on the classifiers, a classification accuracy of about 97.9% was achieved by using the SVM algorithm adopting 3rd order polynomial kernel. LIBS mapping was carried out at the spatial resolution of about 15 mu m. The LIBS mapping image to which a posterior probabilistic interpretation of the SVM results was applied agreed closely to the hematoxylin and eosin (H&E) staining image. It was shown that LIBS imaging can identify the tissue area of cancer and also inflammatory cells. The results showed that UV-fs-LIBS could be a useful alternative or supporting tool for melanoma screening.

Automated summary

This study reports the classification and imaging of melanoma in murine skin tissue using femtosecond laser-induced breakdown spectroscopy (fs-LIBS). The use of an ultraviolet (UV) fs-laser and support vector machine (SVM) classifier achieved an accurate classification accuracy of about 97.9%, making UV-fs-LIBS a potentially valuable tool for melanoma screening.