Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance
Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance
复制标题
DOI：
10.1016/j.neunet.2007.12.031
复制
发表时间：
2008-03-01
期刊：
影响因子：
7.8
通讯作者：
Tourassi, Georgia D.
中科院分区：
文献类型：
--
作者：
Mazurowski, Maciej A.;Habas, Piotr A.;Tourassi, Georgia D.
本研究探讨了在开发用于计算机辅助医疗诊断的神经网络分类器时，训练数据中类不平衡的影响。调查是在存在其他特征的情况下进行的，这些特征在医学数据中是典型的，即小训练样本大小、大量特征以及特征之间的相关性。神经网络训练的两种方法进行了探讨：经典的反向传播（BP）和粒子群优化（PSO）与临床相关的训练标准。利用模拟数据进行了实验研究，并在乳腺癌诊断的真实的临床数据上进一步验证了结论。结果表明，即使在训练数据中存在适度的类不平衡，分类器的性能也会恶化。此外，它表明，BP通常是优于PSO的不平衡的训练数据，特别是与小数据样本和大量的功能。最后，它表明，过采样和无补偿方法之间没有明确的偏好，并提供了一些指导，就适当的选择。(C)2007爱思唯尔有限公司保留所有权利。
This Study investigates the effect of class imbalance in training data when developing neural network classifiers for computer-aided medical diagnosis. The investigation is performed in the presence of other characteristics that are typical among medical data, namely small training sample size, large number of features, and correlations between features. Two methods of neural network training are explored: classical backpropagation (BP) and particle swarm optimization (PSO) with clinically relevant training criteria. An experimental Study is performed using simulated data and the conclusions are further validated on real clinical data for breast cancer diagnosis. The results show that classifier performance deteriorates with even modest class imbalance in the training data. Further, it is shown that BP is generally preferable over PSO for imbalanced training data especially with small data sample and large number of features. Finally, it is shown that there is no clear preference between oversampling and no compensation approach and some guidance is provided regarding a proper selection. (C) 2007 Elsevier Ltd. All rights reserved.