Abstract
Hyperspectral imaging (HSI) is a noninvasive optical modality that holds promise for early detection of tongue lesions. Spectral signatures generated by HSI contain important diagnostic information that can be used to predict the disease status of the examined biological tissue. However, the underlying pathophysiology for the spectral difference between normal and neoplastic tissue is not well understood. Here, we propose to leverage digital pathology and predictive modeling to select the most discriminative features from digitized histological images to differentiate tongue neoplasia from normal tissue, and then correlate these discriminative pathological features with corresponding spectral signatures of the neoplasia. We demonstrated the association between the histological features quantifying the architectural features of neoplasia on a microscopic scale, with the spectral signature of the corresponding tissue measured by HSI on a macroscopic level. This study may provide insight into the pathophysiology underlying the hyperspectral dataset.
Highlights
Hyperspectral imaging (HSI) is a noninvasive optical modality that holds promise for early detection of tongue lesions
The major contribution of this study was the successful validation of the hypothesis that the spectral signature has significant association with histologic features that reflect the tissue architectural changes during malignant transformation
Multiple quantitative histologic features were extracted and selected from the epithelium and its constituent components-nuclei and cytoplasm, which best distinguished normal from neoplastic tissue for both mouse tongue and human tongue
Summary
Hyperspectral imaging (HSI) is a noninvasive optical modality that holds promise for early detection of tongue lesions. Digital pathology, which leverages the power of whole slide imaging and computer-aided diagnosis, holds great promise to providing rapid, consistent, and quantitative cancer diagnosis from histopathology images Noninvasive alternatives, such as various kinds of optical imaging techniques, have been sought to avoid the pain and discomfort of the biopsy procedures. Palmer et al.[10] developed a fast Monte Carlo-based inverse model of diffuse reflectance to extract the concentration of absorbers and the size and density of scatters present in human breast tissue samples, this model is limited in that it requires a priori knowledge of the absorbers and scatterers present in the tissue of interest These modeling methods provide a way to connect the spectral features with the underlying biochemistry and morphology. They generally rely on the assumption of simplified tissue composition and structure, and specific source-detector settings
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
