Pineda-Antunez Carlos, Seguin Claudia, van Duuren Luuk A, Knudsen Amy B, Davidi Barak, de Lima Pedro Nascimento, Rutter Carolyn, Kuntz Karen M, Lansdorp-Vogelaar Iris, Collier Nicholson, Ozik Jonathan, Alarid-Escudero Fernando
The Comparative Health Outcomes, Policy, and Economics (CHOICE) Institute, University of Washington, Seattle, WA, United States.
Institute for Technology Assessment, Massachusetts General Hospital, Boston, MA, United States.
medRxiv. 2024 Feb 5:2023.02.27.23286525. doi: 10.1101/2023.02.27.23286525.
To calibrate Cancer Intervention and Surveillance Modeling Network (CISNET) 's SimCRC, MISCAN-Colon, and CRC-SPIN simulation models of the natural history colorectal cancer (CRC) with an emulator-based Bayesian algorithm and internally validate the model-predicted outcomes to calibration targets.
We used Latin hypercube sampling to sample up to 50,000 parameter sets for each CISNET-CRC model and generated the corresponding outputs. We trained multilayer perceptron artificial neural networks (ANN) as emulators using the input and output samples for each CISNET-CRC model. We selected ANN structures with corresponding hyperparameters (i.e., number of hidden layers, nodes, activation functions, epochs, and optimizer) that minimize the predicted mean square error on the validation sample. We implemented the ANN emulators in a probabilistic programming language and calibrated the input parameters with Hamiltonian Monte Carlo-based algorithms to obtain the joint posterior distributions of the CISNET-CRC models' parameters. We internally validated each calibrated emulator by comparing the model-predicted posterior outputs against the calibration targets.
The optimal ANN for SimCRC had four hidden layers and 360 hidden nodes, MISCAN-Colon had 4 hidden layers and 114 hidden nodes, and CRC-SPIN had one hidden layer and 140 hidden nodes. The total time for training and calibrating the emulators was 7.3, 4.0, and 0.66 hours for SimCRC, MISCAN-Colon, and CRC-SPIN, respectively. The mean of the model-predicted outputs fell within the 95% confidence intervals of the calibration targets in 98 of 110 for SimCRC, 65 of 93 for MISCAN, and 31 of 41 targets for CRC-SPIN.
Using ANN emulators is a practical solution to reduce the computational burden and complexity for Bayesian calibration of individual-level simulation models used for policy analysis, like the CISNET CRC models. In this work, we present a step-by-step guide to constructing emulators for calibrating three realistic CRC individual-level models using a Bayesian approach.
使用基于模拟器的贝叶斯算法校准癌症干预与监测建模网络（CISNET）的自然史结直肠癌（CRC）的SimCRC、MISCAN - 结肠和CRC - SPIN模拟模型，并将模型预测结果内部验证至校准目标。
我们使用拉丁超立方抽样为每个CISNET - CRC模型抽取多达50,000个参数集并生成相应输出。我们使用每个CISNET - CRC模型的输入和输出样本训练多层感知器人工神经网络（ANN）作为模拟器。我们选择具有相应超参数（即隐藏层数、节点数、激活函数、轮次和优化器）的ANN结构，以最小化验证样本上的预测均方误差。我们用概率编程语言实现ANN模拟器，并使用基于哈密顿蒙特卡洛的算法校准输入参数，以获得CISNET - CRC模型参数的联合后验分布。我们通过将模型预测的后验输出与校准目标进行比较来内部验证每个校准后的模拟器。
SimCRC的最优ANN有4个隐藏层和360个隐藏节点，MISCAN - 结肠有4个隐藏层和114个隐藏节点，CRC - SPIN有1个隐藏层和140个隐藏节点。SimCRC、MISCAN - 结肠和CRC - SPIN训练和校准模拟器的总时间分别为7.3小时、4.0小时和0.66小时。SimCRC的110个目标中有98个、MISCAN的93个目标中有65个、CRC - SPIN的41个目标中有31个，模型预测输出的均值落在校准目标的95%置信区间内。
使用ANN模拟器是一种切实可行的解决方案，可减少用于政策分析的个体水平模拟模型（如CISNET CRC模型）的贝叶斯校准的计算负担和复杂性。在这项工作中，我们提供了一个逐步指南，介绍如何使用贝叶斯方法构建模拟器来校准三个实际的CRC个体水平模型。
