arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2606.04888cs.CV
收藏

HD-DinoMoE: 一种用于复杂采集场景下巩膜异常分割的类别感知层次化双混合专家网络

HD-DinoMoE: A Class-Aware Hierarchical Dual Mixture-of-Experts Network for Scleral Anomaly Segmentation in Complex Acquisition Scenarios

School of Electronic and Information Engineering, University of Science and Technology Liaoning(辽宁科技大学电子与信息工程学院)

Yinxiang Yu, Maoxiang Chu, Qi Niu, Guanghu Liu, Wei Xu, Haotian Wang, Zhi Chen, Yutian Zhu, Yuelong Fan, Guanghao Liao

AI总结:

针对多源分布差异、异常形态多样和巩膜镜面反射问题,提出类别感知层次化双混合专家网络HD-DinoMoE,结合双流DINOv3特征融合与类别特定多专家解码,实现血管、黄斑和黑斑、血斑的像素级分割,在ML-SASD数据集上达到72.11%的平均Dice和58.44%的平均IoU。

摘要:

中医目诊通过观察巩膜表面异常提供经验性线索,但其临床应用仍具有主观性且难以量化。为支持智能化和可量化的目诊,本研究提出了中医启发的人工智能眼部辅助诊断系统(TAO),并聚焦于像素级巩膜表面异常分割。针对受多源分布差异、异常形态多样和巩膜镜面反射(SSR)影响的临床和用户采集图像,我们提出了HD-DinoMoE,一种类别感知层次化双混合专家网络。HD-DinoMoE结合类别感知双流DINOv3特征融合与类别特定多专家解码,以分割血管、黄斑和黑斑以及血斑。一种三阶段骨干冻结路由策略稳定了双骨干适应;渐进置信惩罚(PCP)损失减少了SSR区域的高置信度假阳性和分割泄漏;类别感知自适应样本加权(CA-ASW)平衡了样本和类别级别的训练贡献。我们进一步构建了多标签巩膜异常分割数据集(ML-SASD),这是一个包含临床、野生和混合设置以及三种异常类别像素级标注的新基准。在ML-SASD-Mix上,HD-DinoMoE实现了72.11%的平均Dice和58.44%的平均交并比,同时保持了良好的边界定位和镜面区域假阳性控制。它在公共SBVPI数据集的血管子集上也显示出有竞争力的泛化能力。这些结果表明,HD-DinoMoE为复杂采集场景下的TAO提供了一种可行的分割解决方案。代码和数据访问信息可在https://github.com/FX-CMX/HD-DinoMoE获取。

英文摘要:

Traditional Chinese Medicine (TCM) ocular inspection provides empirical cues for assessing scleral surface anomalies, but its clinical use remains subjective and difficult to quantify. To support intelligent and quantifiable ocular inspection, this study presents the TCM-inspired Artificial Intelligence Ocular Auxiliary Diagnosis System (TAO) and focuses on pixel-level scleral surface anomaly segmentation. For clinical and user-acquired images affected by multi-source distributional discrepancies, diverse anomaly morphologies, and scleral specular reflection (SSR), we propose HD-DinoMoE, a class-aware hierarchical dual mixture-of-experts network. HD-DinoMoE combines class-aware dual-stream DINOv3 feature fusion with class-specific multi-expert decoding to segment Vessels, Yellow and Black Spots, and Blood Spots. A three-stage backbone-frozen routing strategy stabilizes dual-backbone adaptation; Progressive Confidence Penalty (PCP) Loss reduces high-confidence false positives and segmentation leakage in SSR regions; and Class-Aware Adaptive Sample Weighting (CA-ASW) balances sample- and class-level training contributions. We further construct the Multi-label Scleral Anomaly Segmentation Dataset (ML-SASD), a new benchmark with Clinical, Wild, and Mix settings and pixel-wise annotations for three anomaly categories. On ML-SASD-Mix, HD-DinoMoE achieves a mean Dice of 72.11% and a mean Intersection-over-Union of 58.44%, while maintaining favorable boundary localization and specular-region false-positive control. It also shows competitive generalization on the Vessels subset of the public SBVPI dataset. These results indicate that HD-DinoMoE provides a feasible segmentation solution for TAO under complex acquisition scenarios. The code and data access information are available at https://github.com/FX-CMX/HD-DinoMoE.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑