Chair of Embedded Intelligence for Health Care & Wellbeing, Faculty of Applied Computer Science, University of Augsburg, Augsburg, Germany.
Department of Environmental Medicine, Faculty of Medicine, University Clinic of Augsburg & University of Augsburg, Augsburg, Germany; Institute of Environmental Medicine, Helmholtz Center Munich, German Research Center for Environmental Health, Augsburg, Germany.
Sci Total Environ. 2023 Sep 15;891:164295. doi: 10.1016/j.scitotenv.2023.164295. Epub 2023 May 19.
Airborne pollen monitoring has been conducted for more than a century now, as knowledge of the quantity and periodicity of airborne pollen has diverse use cases, like reconstructing historic climates and tracking current climate change, forensic applications, and up to warning those affected by pollen-induced respiratory allergies. Hence, related work on automation of pollen classification already exists. In contrast, detection of pollen is still conducted manually, and it is the gold standard for accuracy. So, here we used a new-generation, automated, near-real-time pollen monitoring sampler, the BAA500, and we used data consisting of both raw and synthesised microscope images. Apart from the automatically generated, commercially-labelled data of all pollen taxa, we additionally used manual corrections to the pollen taxa, as well as a manually created test set of bounding boxes and pollen taxa, so as to more accurately evaluate the real-life performance. For the pollen detection, we employed two-stage deep neural network object detectors. We explored a semi-supervised training scheme to remedy the partial labelling. Using a teacher-student approach, the model can add pseudo-labels to complete the labelling during training. To evaluate the performance of our deep learning algorithms and to compare them to the commercial algorithm of the BAA500, we created a manual test set, in which an expert aerobiologist corrected automatically annotated labels. For the novel manual test set, both the supervised and semi-supervised approaches clearly outperform the commercial algorithm with an F1 score of up to 76.9 % compared to 61.3 %. On an automatically created and partially labelled test dataset, we obtain a maximum mAP of 92.7 %. Additional experiments on raw microscope images show comparable performance for the best models, which potentially justifies reducing the complexity of the image generation process. Our results bring automatic pollen monitoring a step forward, as they close the gap in pollen detection performance between manual and automated procedure.
空气传播花粉监测已经进行了一个多世纪，因为了解空气传播花粉的数量和周期性有多种用途，例如重建历史气候和跟踪当前气候变化、法医应用，甚至向花粉引起的呼吸道过敏人群发出预警。因此，已经有相关的自动化花粉分类工作。相比之下，花粉的检测仍然是手动进行的，它是准确性的金标准。因此，在这里我们使用了新一代的、自动化的、接近实时的花粉监测采样器 BAA500，并使用了包括原始和合成显微镜图像的数据。除了自动生成的、商业化标注的所有花粉类别的数据外，我们还额外对花粉类别的数据进行了手动修正，以及手动创建的边界框和花粉类别的测试集，以更准确地评估实际应用的性能。对于花粉检测，我们采用了两阶段的深度神经网络目标检测器。我们探索了一种半监督训练方案来弥补部分标注的不足。使用教师-学生方法，模型可以在训练过程中添加伪标签来完成标注。为了评估我们的深度学习算法的性能，并将其与 BAA500 的商业算法进行比较，我们创建了一个手动测试集，其中一位专家空气生物学家长期对自动标注的标签进行修正。对于新的手动测试集，无论是监督学习还是半监督学习方法都明显优于商业算法，F1 分数高达 76.9%，而商业算法仅为 61.3%。在自动创建的部分标注测试数据集上，我们获得了最高的 mAP 为 92.7%。在原始显微镜图像上的额外实验表明，最佳模型的性能相当，这可能证明了减少图像生成过程复杂性的合理性。我们的结果使自动花粉监测向前迈进了一步，因为它们缩小了手动和自动程序在花粉检测性能方面的差距。
