Polarization Sensitive Optical Coherence Tomography for Surgical Margin Assessment for Canine Soft Tissue Sarcoma

Soft tissue sarcomas (STSs) and skeletal muscle both appear high scattering on spectral-domain optical coherence tomography (SD-OCT) imaging, which resulted in instances of misclassification by observers in prior soft tissue sarcoma margin assessment studies. Polarization-sensitive optical coherence tomography (PS-OCT) is an extension of OCT that provides additional information about tissue polarization contrast. In this prospective study we wanted to evaluate whether PS-OCT improves tissue differentiation and diagnostic accuracy compared to spectral-domain OCT (SD-OCT) alone. PS-OCT images from surgical margins of 10 resected STS were compared with histopathology to identify characteristic features of normal and neoplastic tissues. Then, six observers of varying OCT image interpretation experience assessed surgical margins from 30 resected STS using SD-OCT alone and SD-OCT combined with PS-OCT. PS-OCT images demonstrated distinct birefringence (optical patterns), corresponding to tissue organization: skeletal muscle exhibited alternating high and low birefringence bands, fascia showed variable birefringence and STS had uniformly low birefringence. Overall diagnostic accuracy was high for both modalities [correct classification rate: SD-OCT alone 93.7% (95% confidence interval (CI): 90.1, 96.0) vs. SD-OCT combined with PS-OCT 92.6% (95% = CI: 89.3, 95.4)]; adding PS-OCT did not significantly improve accuracy across all observers (p = 0.82), though expert observers benefited significantly (p = 0.0018). Interobserver and intraobserver agreement ranged from substantial to near perfect. These findings suggest PS-OCT can be interpreted reliably after brief training and may enhance accuracy for experienced users. Further research should explore structured training and real-time applications for comprehensive surgical margin evaluation.