Fibrous cap thickness is a key clinical marker for assessing carotid plaque vulnerability. While intravascular optical coherence tomography (OCT) enables in vivo visualization of fibrous caps, its design for coronary arteries poses challenges in carotid imaging, such as larger vessel size, faster blood flow, limited penetration, and restricted imaging range, leading to incomplete visualization and poor image quality. To address these limitations, we propose a dual-coordinate segmentation framework for carotid OCT fibrous cap segmentation. This framework integrates Cartesian images, which preserve global spatial context, with linear-polar transformed images, effectively representing the annular geometry of fibrous caps. The fusion of dual-coordinate features mitigates incomplete vascular walls and blood artifacts, enhancing segmentation accuracy and robustness. We introduce a Cross-Coordinate Feature Fusion Module (CCFFM) to efficiently integrate features from both coordinate systems and reduce interference from redundant information. Additionally, the Kolmogorov-Arnold Network (KAN) block is incorporated to extract complex nonlinear features while improving model interpretability. Our method achieves state-of-the-art performance on an external carotid OCT dataset, demonstrating the potential of OCT for advancing carotid imaging and improving plaque vulnerability assessment.