Optical nerve identification in head and neck surgery after Er:YAG laser ablation

Facial nerve function may be hampered by iatrogenic damage during head and neck laser surgery procedures. Optical techniques can serve as a basis for feedback-controlled tissue-specific laser surgery on the jaw bone and the parotid gland. In order to preserve nerve tissue during laser surgery, the alteration of optical tissue properties through laser-tissue interactions have to be taken into account. It was the aim of this study to evaluate the viability of optical tissue differentiation through diffuse reflectance spectroscopy after exposure to laser light as a basis for a feedback system for tissue-specific laser surgery. Spectra of diffuse reflectance (wavelength, 350-650 nm) of nerves, salivary glands, and cortical and cancellous bone of the midfacial region (ex vivo domestic pig heads) were acquired before/after Er:YAG laser (wavelength, 2.94 μm) ablation (each 16,800 spectra). Principal component analysis was computed followed by quadratic discriminant analysis. The tissue classification performance as well as area under the curve (AUC) sensitivity and specificity for tissue differentiation was assessed before and after laser-tissue exposure. A high classification performance was observed before laser ablation (total error, 7.74%). Nerve tissue was differentiated from bone and salivary glands with results greater than 0.96 in AUC, sensitivity and specificity. After laser exposure, a total classification error of 18.61% was observed. The differentiation of nerve tissue was reduced with an AUC of >0.94, sensitivity of >0.95, and specificity >0.87. Er:YAG laser ablation only slightly reduces the differentiation performance through diffuse reflectance in the investigated tissue types. The results show the general viability of diffuse reflectance spectroscopy in identifying neural structures in the vicinity of salivary glands and bone as a basis for nerve preservation during feedback-controlled laser surgery.