Busayakanon Supasuta, Kaewthamasorn Morakot, Pinetsuksai Natchapon, Tongloy Teerawat, Chuwongin Santhad, Boonsang Siridech, Kittichai Veerayuth
Faculty of Medicine, King Mongkut's Institute of Technology Ladkrabang, Bangkok 10520, Thailand.
Department of Pathology, Center of Excellence in Veterinary Parasitology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand.
Vet World. 2024 Nov;17(11):2619-2634. doi: 10.14202/vetworld.2024.2619-2634. Epub 2024 Nov 25.
Zoonotic diseases caused by various blood parasites are important public health concerns that impact animals and humans worldwide. The traditional method of microscopic examination for parasite diagnosis is labor-intensive, time-consuming, and prone to variability among observers, necessitating highly skilled and experienced personnel. Therefore, an innovative approach is required to enhance the conventional method. This study aimed to develop a self-supervised learning (SSL) approach to identify zoonotic blood parasites from microscopic images, with an initial focus on parasite species classification.
We acquired a public dataset featuring microscopic images of Giemsa-stained thin blood films of trypanosomes and other blood parasites, including , and , as well as images of both white and red blood cells. The input data were subjected to SSL model training using the Bootstrap Your Own Latent (BYOL) algorithm with Residual Network 50 (ResNet50), ResNet101, and ResNet152 as the backbones. The performance of the proposed SSL model was then compared to that of baseline models.
The proposed BYOL SSL model outperformed supervised learning models across all classes. Among the SSL models, ResNet50 consistently achieved high accuracy, reaching 0.992 in most classes, which aligns well with the patterns observed in the pre-trained uniform manifold approximation and projection representations. Fine-tuned SSL models exhibit high performance, achieving 95% accuracy and a 0.960 area under the curve of the receiver operating characteristics (ROC) curve even when fine-tuned with 1% of the data in the downstream process. Furthermore, 20% of the data for training with SSL models yielded ≥95% in all other statistical metrics, including accuracy, recall, precision, specification, F1 score, and ROC curve. As a result, multi-class classification prediction demonstrated that model performance exceeded 91% for the F1 score, except for the early stage of , which showed an F1 score of 87%. This may be due to the model being exposed to high levels of variation during the developmental stage.
This approach can significantly enhance active surveillance efforts to improve disease control and prevent outbreaks, particularly in resource-limited settings. In addition, SSL addresses significant challenges, such as data variability and the requirement for extensive class labeling, which are common in biology and medical fields.
由各种血液寄生虫引起的人畜共患疾病是重要的公共卫生问题，影响着全球的动物和人类。传统的寄生虫诊断显微镜检查方法劳动强度大、耗时，且观察者之间容易出现差异，需要高技能和经验丰富的人员。因此，需要一种创新方法来改进传统方法。本研究旨在开发一种自监督学习（SSL）方法，从显微镜图像中识别出人畜共患血液寄生虫，最初重点是寄生虫物种分类。
我们获取了一个公共数据集，其中包含经吉姆萨染色的锥虫和其他血液寄生虫薄血膜的显微镜图像，包括[具体寄生虫名称未给出]，以及白细胞和红细胞的图像。使用以残差网络50（ResNet50）、ResNet101和ResNet152为骨干的自引导潜在（BYOL）算法对输入数据进行SSL模型训练。然后将所提出的SSL模型的性能与基线模型的性能进行比较。
所提出的BYOL SSL模型在所有类别上均优于监督学习模型。在SSL模型中，ResNet50始终保持较高的准确率，在大多数类别中达到0.992，这与预训练的均匀流形逼近和投影表示中观察到的模式非常吻合。微调后的SSL模型表现出高性能，即使在下游过程中用1%的数据进行微调，也能达到95%的准确率和接收器操作特征（ROC）曲线下0.960的面积。此外，使用SSL模型进行训练的20%的数据在所有其他统计指标上，包括准确率、召回率、精确率、特异性、F1分数和ROC曲线，都产生了≥95%的结果。因此，多类分类预测表明，除了[具体寄生虫名称未给出]的早期阶段F1分数为87%外，模型性能在F1分数方面超过了91%。这可能是由于模型在发育阶段面临高水平的变异。
这种方法可以显著加强主动监测工作，以改善疾病控制和预防疫情爆发，特别是在资源有限的环境中。此外，SSL解决了重大挑战，如数据变异性和对广泛类别标签的需求，这些在生物学和医学领域很常见。
