论文概要
研究领域: ML 作者: Serena Su, Yifan Wang, Senwei Liang 发布时间: 2026-08-17 arXiv: 2608.16870
中文摘要
循环肿瘤细胞（CTC）表型的准确分类可为评估转移潜力提供有价值的信息。无标记微流控设备提供水动力障碍赛道，将CTC的微妙生物物理特征（包括大小和可变形性）转化为独特的运动轨迹。然而，控制这些轨迹的高度非线性流固耦合使得从轨迹数据推断细胞表型的逆问题在解析上难以处理。虽然深度神经网络（DNN）已成为解决这一逆问题的强大方法，但其有效性受限于轨迹数据的有限可用性和缺乏物理可解释性。为此，我们提出了一个可解释且数据高效的DNN框架用于基于轨迹的CTC分类。为缓解数据稀缺，我们开发了SubSeq（子序列）策略，在训练期间随机提取信息丰富的局部轨迹段以促进从局部模式学习。我们进一步应用梯度加权类激活映射来识别驱动模型预测的轨迹特征和微流控设备的物理区域。可解释性分析表明，局部轨迹段包含与准确分类相关的大量生物物理信息，这凸显了全长度轨迹的冗余性。
原文摘要
Accurate classification of circulating tumor cell (CTC) phenotypes can provide valuable information for assessing metastatic potential. Label free microfluidic devices provide a hydrodynamic obstacle course that transforms subtle biophysical characteristics of CTCs, including size and deformability, into distinct kinematic trajectories. However, the highly nonlinear fluid structure interactions governing these trajectories make the inverse problem of inferring cellular phenotype from trajectory data analytically intractable. While deep neural networks (DNNs) have emerged as a powerful approach for addressing this inverse problem, their effectiveness is constrained by the limited availability of trajectory data and the lack of physical interpretability. To address these challenges, we propo…
— 自动采集于 2026-08-19
#论文 #arXiv #ML #小凯