Robust lumen segmentation based on temporal residual U-Net using spatiotemporal features in intravascular optical...

Lumen segmentation in intravascular optical coherence tomography (IVOCT) images is essential for quantifying vascular stenosis severity, location, and length. Current methods relying on manual parameter tuning or single-frame spatial features struggle with image artifacts, limiting clinical utility.

We aim to develop a temporal residual U-Net (TR-Unet) leveraging spatiotemporal feature fusion for robust IVOCT lumen segmentation, particularly in artifact-corrupted images.

We integrate convolutional long short-term memory networks to capture vascular morphology evolution across pullback sequences, enhanced ResUnet for spatial feature extraction, and coordinate attention mechanisms for adaptive spatiotemporal fusion.

By processing 2451 clinical images, the proposed TR-Unet model shows a well performance as Dice coefficient = 98.54%, Jaccard similarity (JS) = 97.17%, and recall = 98.26%. Evaluations on severely blood artifact-corrupted images reveal improvements of 3.01% (Dice), 1.3% (ACC), 5.24% (JS), 2.15% (recall), and 2.06% (precision) over competing methods.

TR-Unet establishes a robust and effective spatiotemporal fusion paradigm for IVOCT segmentation, demonstrating significant robustness to artifacts and providing architectural insights for temporal modeling optimization.