Abstract：
Malaria was regarded as the most devastating infectious disease of the 21st century until the COVID-19 pandemic. Asexual blood staged parasites (ABS) play a unique role in ensuring the parasite\'s survival and pathogenesis. Hitherto, there have been no spectroscopic reports discriminating the life cycle stages of the ABS parasite under physiological conditions. The identification and quantification of the stages in the erythrocytic life cycle is important in monitoring the progression and recovery from the disease. In this study, we explored visible microspectrophotometry coupled to machine learning to discriminate functional ABS parasites at the single cell level. Principal Component Analysis (PCA) showed an excellent discrimination between the different stages of the ABS parasites. Support Vector Machine Analysis provided a 100% prediction for both schizonts and trophozoites, while a 92% and 98% accuracy was achieved for predicting control and ring staged infected RBCs, respectively. This work shows proof of principle for discriminating the life cycle stages of parasites in functional erythrocytes using visible microscopy and thus eliminating the drying and fixative steps that are associated with other optical-based spectroscopic techniques.
摘要：
在COVID-19大流行之前，疟疾被认为是21世纪最具破坏性的传染病。无性系血液分期寄生虫(ABS)在确保寄生虫的生存和发病机制方面发挥着独特的作用。到目前为止，目前还没有光谱报告区分ABS寄生虫在生理条件下的生命周期阶段。红细胞生命周期阶段的识别和量化对于监测疾病的进展和恢复非常重要。在这项研究中,我们探索了可见显微分光光度法与机器学习相结合，以在单细胞水平上区分功能性ABS寄生虫。主成分分析（PCA）显示了ABS寄生虫不同阶段之间的出色区分。支持向量机分析为裂殖体和滋养体提供了100%的预测，虽然预测对照和环分期感染的红细胞达到了92%和98%的准确性，分别。这项工作显示了使用可见显微镜区分功能性红细胞中寄生虫的生命周期阶段的原理证明，从而消除了与其他基于光学的光谱技术相关的干燥和固定步骤。
