Wavelet scattering transform application in classification of retinal abnormalities using OCT images.
Wavelet scattering transform application in classification of retinal abnormalities using OCT images.
复制标题
DOI：
10.1038/s41598-023-46200-1
复制
发表时间：
2023-11-03
影响因子：
4.6
通讯作者：
中科院分区：
文献类型：
--
作者：
计算机辅助诊断在眼科医生诊断视网膜病变中发挥了重要作用。在本文中，一个特殊的卷积神经网络的小波散射变换（WST）的基础上被用来检测一至四个视网膜异常的光学相干断层扫描（OCT）图像。与深度学习方法相比，该网络中的预定义小波滤波器降低了计算复杂度和处理时间。我们使用两层的WST网络，以获得一个直接和有效的模型。WST生成一个稀疏表示的图像，这是不变性和稳定的局部变形。接下来，主成分分析对提取的特征进行分类。我们使用四个公开可用的数据集来评估该模型，与文献进行全面比较。将OCTID数据集的OCT图像分为两类和五类的准确率分别为和。使用基于TOPCON设备的数据集，我们在检测糖尿病黄斑水肿和正常黄斑水肿方面的准确度达到。海德堡和杜克数据集包含DME、黄斑变性相关性黄斑变性和正常类，其中我们分别实现了和的准确度。我们的结果与最先进的模型的比较表明，我们的模型优于这些模型的一些评估或实现几乎最好的结果报告到目前为止，同时具有更小的计算复杂性。
To assist ophthalmologists in diagnosing retinal abnormalities, Computer Aided Diagnosis has played a significant role. In this paper, a particular Convolutional Neural Network based on Wavelet Scattering Transform (WST) is used to detect one to four retinal abnormalities from Optical Coherence Tomography (OCT) images. Predefined wavelet filters in this network decrease the computation complexity and processing time compared to deep learning methods. We use two layers of the WST network to obtain a direct and efficient model. WST generates a sparse representation of the images which is translation-invariant and stable concerning local deformations. Next, a Principal Component Analysis classifies the extracted features. We evaluate the model using four publicly available datasets to have a comprehensive comparison with the literature. The accuracies of classifying the OCT images of the OCTID dataset into two and five classes were and , respectively. We achieved an accuracy of in detecting Diabetic Macular Edema from Normal ones using the TOPCON device-based dataset. Heidelberg and Duke datasets contain DME, Age-related Macular Degeneration, and Normal classes, in which we achieved accuracy of and , respectively. A comparison of our results with the state-of-the-art models shows that our model outperforms these models for some assessments or achieves nearly the best results reported so far while having a much smaller computational complexity.
登录
查看更多内容
影响因子：
7.6
作者：
Rabbani H;Sonka M;Abramoff MD
通讯作者：
Abramoff MD
DOI：
10.3390/e22050567
发表时间：
2020-05-19
期刊：
Entropy (Basel, Switzerland)
影响因子：
--
作者：
Ali A;Qadri S;Khan Mashwani W;Kumam W;Kumam P;Naeem S;Goktas A;Jamal F;Chesneau C;Anam S;Sulaiman M
通讯作者：
Sulaiman M
DOI：
10.3390/s22041667
发表时间：
2022-02-21
期刊：
Sensors (Basel, Switzerland)
影响因子：
--
作者：
Khan AM;Hassan T;Akram MU;Alghamdi NS;Werghi N
通讯作者：
Werghi N
影响因子：
3.9
作者：
Abdelmaksoud, Eman;El-Sappagh, Shaker;Elmogy, Mohammed
通讯作者：
Elmogy, Mohammed
影响因子：
3.2
作者：
Hsieh, Yi-Ting;Chuang, Lee-Ming;Chen, Mingke
通讯作者：
Chen, Mingke