Differentiating between normal and cancerous thyroid tissue using quantitative Second Harmonic Generation microscopy

Significant challenges such as a high proportion of indeterminate results, inadequate sampling, and false positives remain prevalent in the diagnosis of thyroid cancer when using fine needle aspiration biopsy or core needle biopsy. Second harmonic generation (SHG) imaging is a label-free, nonlinear optical imaging method that selectively uses fibrillar collagen arrangement to provide endogenous contrast. We investigated applications of SHG-specific feature extraction methods such as anisotropy, and more broadly applied gray-level co-occurrence matrices to generate quantitative metrics that describe either the structural or physical properties of the imaged thyroid samples. The features differed significantly between thyroid cancer and normal thyroid samples, thus independently possessing the capability to differentiate between cancerous and normal thyroid tissue.