基于Ano-GAN模型的医学图像缺陷检测研究——以MRI影像为例
Research on Medical Image Defect Detection Based on Ano GAN Model—Taking MRI Imaging Data as an Example
DOI: 10.12677/AAM.2023.1210433, PDF,   
作者: 史丽琴*, 王仲平#, 程正祥：兰州交通大学数理学院，甘肃 兰州
关键词: MRI影像；DCGAN；缺陷检测；图像拟合；MRI Imaging； DCGAN； Defect Detection； Image Fitting
摘要: 图像异常检测是计算机视觉中重要的研究课题，本文主要就MRI医学影像展开研究，致力于提高机器诊断疾病的效率，主要选取人体膝关节的MRI数据进行实验，研究Ano-GAN模型在MRI数据上的缺陷检测效果。通过多次对比实验，就软件最优模型输出的切片影像及生成器和判别器网络的损失，不断调整模型网络的层数及相关参数，从而得到最优模型，并通过最优模型输出的结果，分析模型对数据的检测效果，分析模型在该类数据检测当中的效果及普遍MRI医学影像检测的适用程度，对于提高深度学习模型在异常检测领域的发展有重要意义。
Abstract: Image anomaly detection is an important research topic in computer vision. This article mainly fo-cuses on MRI medical images, aiming to improve the efficiency of machine diagnosis of diseases. It mainly selects MRI data of human knee joints for experiments to study the defect detection effect of the Ano-GAN model on MRI data. Through multiple comparative experiments, we continuously ad-just the number of layers and related parameters of the model network to obtain the optimal model based on the slice images output by the software’s optimal model, as well as the losses of the gener-ator and discriminator network. Through the results of the optimal model output, we analyze the model’s detection effect on data, its effectiveness in detecting such data, and its applicability in general MRI medical image detection. It is of great significance for improving the development of deep learning models in the field of anomaly detection.
文章引用：史丽琴, 王仲平, 程正祥. 基于Ano-GAN模型的医学图像缺陷检测研究——以MRI影像为例[J]. 应用数学进展, 2023, 12(10): 4403-4414. https://doi.org/10.12677/AAM.2023.1210433

参考文献

[1] 吕承侃, 沈飞, 张正涛, 等. 图像异常检测研究现状综述[J]. 自动化学报, 2022, 48(6): 1402-1428.
[2] 李少波, 杨静, 王铮, 等. 缺陷检测技术的发展与应用研究综述[J]. 自动化学报, 2020, 46(11): 2319-2336.
[3] 汤勃, 孔建益, 伍世虔. 机器视觉表面缺陷检测综述[J]. 中国图象图形学报, 2017, 22(12): 1640-1663.
[4] 阿斯科纳D, 麦吉尼斯K, 斯米顿A F. 利用MRNet数据集检测膝关节损伤的现有和新的深度学习方法的比较研究[C]//2020智能数据科学技术与应用国际会议(IDSTA). 2020: 149-155.
[5] Hashemi, R.H., Bradley, W.G. and Lisanti, C.J. (2012) MRI: The Basics. Lippincott Williams & Wilkins, Philadelphia.
[6] Goodfellow, J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. and Bengio, Y. (2014) Generative Adversarial Nets. Advances in Neural Information Processing Systems, 27.
[7] Schlegl, T., Seeböck, P., Waldstein, S.M., et al. (2017) Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery. In: Niethammer, M., et al., Eds., International Conference on Information Processing in Medical Imaging, Springer International Publishing, Cham, 146-157. [Google Scholar] [CrossRef] 
[8] Wang, Y. (2020) A Mathematical Introduction to Generative Adversarial Nets (GAN). arXiv:2009.00169, 2020.
[9] Radford, A., Metz, L. and Chintala, S. (2015) Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. arXiv:1511.06434.
[10] Fang, W., Zhang, F., Sheng, V.S., et al. (2018) A Method for Improving CNN-Based Image Recognition Using DCGAN. Com-puters, Materials & Continua, 57, 167-178. [Google Scholar] [CrossRef] 
[11] Wu, Q., Chen, Y. and Meng, J. (2020) DCGAN-Based Data Augmentation for Tomato Leaf Disease Identification. IEEE Access, 8, 98716-98728. [Google Scholar] [CrossRef] 

为你推荐