FlowLOT：用于流式细胞术分析的线性化最优传输

中文摘要 AI 辅助

多参数流式细胞术生成用于疾病诊断和监测的高维、无序单细胞测量数据，然而分析往往仍依赖于手动设门，限制了可扩展性和可重复性。现有的机器学习方法可以减少注释负担，但通常需要大规模训练队列，且可解释性有限。为解决这些挑战，我们引入了FlowLOT，一个基于最优传输的框架，将患者样本的单细胞测量数据建模为经验细胞分布，并直接将其映射为固定长度的特征向量。在单一透明架构内，FlowLOT统一了高维分类、可解释可视化和连续定量推断。在少样本场景下，在FlowCAP-II上每类仅使用16例患者，在BLAST110上每类仅使用8例患者，它就能准确区分健康样本和急性髓系白血病（AML）样本，分别达到94.3%和98.0%的平衡准确率。底层嵌入揭示了驱动疾病相关群体转移的标志物水平变异，并能够进行定量的可测量残留病（MRD）估计，在保留样本上达到0.82的Pearson相关系数，在跨数据集迁移下达到0.79。此外，在临床相关的白血病相关免疫表型（LAIP）残留病0.1%阈值下，FlowLOT以100%的特异性检测阳性，灵敏度为72%。通过用分布感知框架取代主观手动设门和黑箱深度学习，FlowLOT为在现实临床和实验约束下的高维细胞术提供了一种样本高效、可扩展且可解释的解决方案。

英文摘要

Multiparameter flow cytometry generates high-dimensional, unordered single-cell mea- surement data for disease diagnosis and monitoring, yet analysis often remains dependent on manual gating, limiting scalability and reproducibility. Existing machine-learning ap- proaches can reduce annotation burden but frequently require large training cohorts and offer limited interpretability. To address these challenges, we introduce FlowLOT , an optimal-transport-based framework that models the single-cell measurement data of a pa- tient sample as an empirical cellular distribution and maps it directly into a fixed-length feature vector. Within a single transparent architecture, FlowLOT unifies high-dimensional classification, interpretable visualization, and continuous quantitative inference. In few-shot regimes, using as few as 16 patients per class on FlowCAP-II and 8 patients per class on BLAST110, it accurately distinguishes healthy from acute myeloid leukemia (AML) sam- ples, reaching 94.3% and 98.0% balanced accuracy, respectively. The underlying embedding exposes marker-level variation driving disease-associated population shifts and enables quantitative measurable residual disease (MRD) estimation, achieving a Pearson correlation of 0.82 on held-out samples and 0.79 under cross-dataset transfer. Furthermore, at the clinically relevant 0.1% threshold for leukemia-associated immunophenotype (LAIP) residual disease, FlowLOT detects positivity with 72% sensitivity at 100% specificity. By replacing subjective manual gating and black-box deep learning with a distribution-aware framework, FlowLOT offers a sample-efficient, scalable, and interpretable solution for high- dimensional cytometry under realistic clinical and experimental constraints.

发表机构

• University of Virginia（弗吉尼亚大学）
• North South University（北南大学）
• Purdue University（普渡大学）

机构由 AI 辅助整理，请以论文原文为准。