Tissue optical properties estimation from cross-polarization OCT data for breast cancer margin assessment

The aim of this study was to investigate the application of optical coefficients obtained from cross-polarization optical coherence tomography (CP OCT) data to differentiate breast cancer tissues from normal breast tissues. For this, surgically obtained breast specimens from 35 patients were investigated using CP OCT in order to construct pseudocoloren faceOCT maps based on a calculation of three optical coefficients: the commonly used, rate of attenuation in the co-channel (Att(co-)) and, additionally, the attenuation in the cross-channel (Att(cross-)); and the interchannel attenuation difference (uD). It was shown that the use of these optical coefficients significantly increased the information available from the OCT data in comparison with unprocessed images, and that this enabled objective quantification for differentiating non-tumorous and tumorous tissue (adipose tissue, normal stroma, tumor stroma and agglomerates of tumor cells). The Att(cross-)and the uD provided greater contrast for the visualization of the different breast cancer structures compared to the Att(co-)coefficient. While the Att(co-)of tumor cells was significantly higher (p < 0.05) than that of adipose tissue, the Att(co-)difference between normal or tumor stroma and tumor cells was not observed. On the other hand, normal or tumor stroma was associated with significantly higher (p < 0.05) Att(cross-)and significantly lower (p < 0.05) uD as compared to tumor cells. Furthermore, Att(cross-)of tumor stroma was significantly higher (p < 0.05) than that of normal (non-tumor) stroma. The findings of this study suggest that CP OCT and an assesment of the optical coefficients of such breast cancer images may, in the future, enable real-time feedback to the surgeon about accurate resection margin locations in patients with breast cancer.