发表机构
Institute of Electronics and Computer Science (EDI); Faculty of Science and Technology, University of Latvia（电子与计算机科学研究所（EDI）; 拉脱维亚大学科学与技术学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究评估重建全息图在兽医细胞学花粉分析中的效果，发现其缩小了与明场显微镜的差距，并利用AHIR协议区分了解释失败与模型脆弱性伪影。
AI 中文摘要
自动化花粉分析支持兽医细胞学诊断，但明场显微镜比无透镜数字同轴全息显微镜成本更高且更复杂。我们评估重建全息图能否缩小这一差距，以及模型解释在模态变化下是否保持可靠。对六种花粉物种分别采用明场和全息显微镜成像。在基于锚点的标注迁移后，使用YOLOv26s检测和MobileNetV4分类对原始、单次反向传播和迭代相位恢复全息图进行评估。采用六种归因方法，通过归因健康检查与修复（AHIR）协议评估空间定位和忠实度，该协议测试模型在弱噪声下的脆弱性，并在需要时校正归因图的粒度。明场检测达到0.6890 mAP50-95（0.8865 mAP50），分类达到0.9687宏F1（0.9705准确率）。重建全息图缩小了差距，但呈现任务依赖性分化：p型在检测上最强，达到0.5324 mAP50-95（0.8229 mAP50），而r型在分类上最强，达到0.7695宏F1（0.7866准确率），两者均远高于原始全息图基线。基于激活的解释在花粉颗粒上定位强烈，基于区域的方法在全息成像下保留了约60%至80%的忠实度。全息检测器对弱扰动高度脆弱，使基于删除的评估饱和，而基于插入的评估仍具信息量。像素级梯度解释接近随机基线，但空间平滑将p型梯度忠实度从0.05恢复至0.51。对于全息分类，基于扰动的解释保持忠实，而基于梯度的方法低于随机基线。重建改善了低成本全息花粉分析，而AHIR区分了真正的归因失败与由模型脆弱性和图粒度引起的伪影。
英文摘要
Automated pollen analysis supports veterinary cytology, but brightfield microscopy is costlier and more complex than lens-less digital in-line holographic microscopy. We evaluate whether reconstructed holograms can narrow this gap and whether model explanations remain reliable under modality change. Six pollen species were imaged by brightfield and holographic microscopy. Raw, single back-propagation and iterative phase retrieval holograms were evaluated with YOLOv26s detection and MobileNetV4 classification after anchor-based annotation transfer. Six attribution methods were assessed for spatial grounding and faithfulness with the Attribution Health Inspection and Repair (AHIR) protocol, which tests model brittleness under weak noise and corrects attribution-map granularity when needed. Brightfield achieved 0.6890 mAP50-95 (0.8865 mAP50) for detection and 0.9687 macro-F1 (0.9705 accuracy) for classification. Reconstructed holograms narrowed the gap with a task-dependent split: p-type was strongest for detection at 0.5324 mAP50-95 (0.8229 mAP50), while r-type was strongest for classification at 0.7695 macro-F1 (0.7866 accuracy), both far above raw-hologram baselines. Activation-based explanations localized strongly on grains, and region-based methods retained ~60 to ~80% of faithfulness under holography. The holographic detector was highly brittle to weak perturbations, saturating deletion-based evaluation while insertion remained informative. Pixel-level gradient explanations approached random floor, yet spatial smoothing restored p-type gradient faithfulness from 0.05 to 0.51. For holographic classification, perturbation-based explanations remained faithful while gradient-based methods fell below random floor. Reconstruction improves low-cost holographic pollen analysis, while AHIR distinguishes genuine attribution failure from artifacts caused by model brittleness and map granularity.
Comments25 pages, 17 figures, 20 tables. Supplementary material included. Submitted to a journal for peer review
