Head-and-Neck PET/CT Lesion Segmentation via SSIMH and SegResNet

Shuai Huang, Yuqi Cheng, Shilin Yu, Hongtu Zhu

Keywords: HECKTOR25 · Head and neck cancer · PET/CT segmentation · Domain adaptation · SegResNet

Abstract: We present a simple and effective pipeline for automatic de-tection and segmentation of primary tumors and lymph nodes in FDG-PET/CT for HECKTOR 2025 Task 1. The method starts with an anatomy-aware pre-crop of the head-and-neck region to suppress irrelevant con-text, followed by modality-specific intensity normalization with soft clamp-ing. To mitigate cross-center domain shift, we apply single-subject, SSIM-guided spectrum swapping (SSIMH) on CT in the frequency domain without external references. For segmentation, we use a residual U-Net–style SegResNet with deep supervision and a combined Dice + Cross-Entropy loss. Training employs stratified five-fold cross-validation with foreground-centered sampling to emphasize small lesions. At inference, we use sliding-window tiling on the cropped volumes, lightweight post-processing to remove small isolated components, and a five-model ensem-ble by averaging per-voxel logits before softmax. On the official HECK-TOR 2025 Task 1 test set, our approach achieves a GTVp Dice of 0.5779, a GTVn aggregated Dice (DSCagg) of 0.5280, and a GTVn aggregated lesion-wise F1 of 0.3207. The overall recipe is concise and reproducible, providing a strong and transparent baseline for multi-center head-and-neck PET/CT segmentation under domain shift. (Team name: BIGS2)

Submission Number: 14