人工神经网络用于脓毒症背景下的疾病轨迹预测
Artificial neural networks for disease trajectory prediction in the context of sepsis
摘要：
临床脓毒症的疾病轨迹，在时间细胞因子和表型动力学方面，可以被解释为一个随机动力学系统。从临床测量中准确预测患者状态的能力一直困扰着生物医学界，主要是由于缺乏相关和高分辨率的数据。我们已经利用两种不同的神经网络架构，长短期记忆和多层感知器，采取的时间序列的五个测量值的11个模拟血清细胞因子浓度作为输入，并返回未来的细胞因子轨迹，以及代表患者的健康状态的总指标。神经网络在50个时期内收敛，用于细胞因子轨迹预测和健康指标回归，具有预期的误差量（由于模拟中的随机性）。从特定细胞因子谱到健康状态的映射并不是唯一的，炎症水平的增加导致预测不太准确。由于机器学习误差的传播与计算模型的随机性相结合，网络应该每天在现实中重新接地，因为随着系统朝着概率吸引盆地发展，预测可能会偏离真实的模型轨迹。这项工作作为使用人工神经网络预测脓毒症疾病进展的概念验证。这项工作并不是为了取代训练有素的临床医生，而是为了用可量化的统计信息增强直觉，帮助他们做出最佳决策。我们注意到，这依赖于所讨论系统的有效计算模型，因为不存在足够的数据来通知机器学习训练的人工智能控制器。
英文摘要：
The disease trajectory for clinical sepsis, in terms of temporal cytokine and phenotypic dynamics, can be interpreted as a random dynamical system. The ability to make accurate predictions about patient state from clinical measurements has eluded the biomedical community, primarily due to the paucity of relevant and high-resolution data. We have utilized two distinct neural network architectures, Long Short-Term Memory and Multi-Layer Perceptron, to take a time sequence of five measurements of eleven simulated serum cytokine concentrations as input and to return both the future cytokine trajectories as well as an aggregate metric representing the patient's state of health. The neural networks converged within 50 epochs for cytokine trajectory predictions and health-metric regressions, with the expected amount of error (due to stochasticity in the simulation). The mapping from a specific cytokine profile to a state-of-health is not unique, and increased levels of inflammation result in less accurate predictions. Due to the propagation of machine learning error combined with computational model stochasticity over time, the network should be re-grounded in reality daily as predictions can diverge from the true model trajectory as the system evolves towards a probabilistic basin of attraction. This work serves as a proof-of-concept for the use of artificial neural networks to predict disease progression in sepsis. This work is not intended to replace a trained clinician, rather the goal is to augment intuition with quantifiable statistical information to help them make the best decisions. We note that this relies on a valid computational model of the system in question as there does not exist sufficient data to inform a machine-learning trained, artificially intelligent, controller.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
