Noninvasive imaging of oral neoplasia with a high-resolution fiber-optic microendoscope

Background The purpose of this study was to evaluate the ability of high-resolution microendoscopy to image and quantify changes in cellular and architectural features seen in early oral neoplasia in vivo. Methods A high-resolution microendoscope (HRME) was used to image intact, resected oral squamous carcinoma specimens. HRME images were reviewed and classified as non-neoplastic or neoplastic by expert clinicians. An algorithm based on quantitative morphologic features was also used to classify each image. Results were compared to the histopathologic diagnosis. Results HRME images were obtained from 141 sites in resected specimens from 13 patients. Subjective image interpretation yielded sensitivity and specificity of 85% to 90% and 80% to 85%, respectively, whereas the objective classification algorithm achieved sensitivity and specificity of 81% and 77%, respectively. Conclusion High-resolution microendoscopy of intact oral mucosa can provide images with sufficient detail to classify oral lesions by both subjective image interpretation and objective image analysis.