Machine learning-enabled cancer diagnostics with widefield polarimetric second-harmonic generation microscopy

Abstract

The extracellular matrix (ECM) collagen undergoes major remodeling during tumorigenesis. However, alterations to the ECM are not widely considered in cancer diagnostics, due to mostly uniform appearance of collagen fibers in white light images of hematoxylin and eosin-stained tissue sections. Polarimetric second-harmonic generation (P-SHG) microscopy enables label-free visualization and ultrastructural investigation of non-centrosymmetric molecules, which, when combined with texture analysis, provides multiparameter characterization of tissue collagen. This paper demonstrates whole slide imaging of breast tissue microarrays using high-throughput widefield P-SHG microscopy. The resulting P-SHG parameters are used in classification to differentiate tumor tissue from normal with 94.2% accuracy and F1-score, and 6.3% false discovery rate. Subsequently, the trained classifier is employed to predict tumor tissue with 91.3% accuracy, 90.7% F1-score, and 13.8% false omission rate. As such, we show that widefield P-SHG microscopy reveals collagen ultrastructure over large tissue regions and can be utilized as a sensitive biomarker for cancer diagnostics and prognostics studies.

Wide-Field Multicontrast Nonlinear Microscopy for Histopathology

“…The SHG LD distribution is also influenced by the R-and C-ratios and the relative phase Δ xyz . The polarimetric parameters can be used for ultrastructural characterization of the histology tissue and for cancer diagnostics [42].…”

“…The polarimetric measurements are performed by employing liquid crystal variable retarders (LCVR), one for PSG and one for PSA. Two perpendicular linear or circular polarization states can be produced with the LCVR oriented with fast axis at 45 degrees from the incident linear polarization orientation [15, 42]. The SHG signal intensity, I M , measured after the PSA containing a single LCVR oriented at 45 degrees with respect to vertical polarization can be characterized with: where s 0 , s 1 , s 2 , s 3 are the measured SHG Stokes vector components [44, 45] and θ is the retardance angle of the LCVR.…”

“…By measuring the intensities at retardances for the PSA-LCVR, we obtain a system of 4 equations with 3 Stokes vector components, which are solved for s 0 , s 1 , and s 3 . The obtained values are averaged to find a unique value for each of the components [42]. Note, that s 2 of SHG signal component can be measured with the addition of a second LCVR on the PSA.…”

“…The wide-field microscope is based on the previously described setup [33] with addition of polarization state generator and polarization state analyzer [42]. The setup outline is presented in Fig.…”

“…An example of polarimetric wide-field SHG microscopy for a large area (700`< x 700`<) imaging of histology tissue microarray cores is presented. The newly developed wide-field polarimetric imaging method can be readily applied for nonlinear digital pathology investigations [42].…”