Medical & Biological Engineering & Computing ( IF 2.6 ) Pub Date : 2021-04-06 , DOI: 10.1007/s11517-021-02357-3 Revathy Sivanandan 1 , J Jayakumari 1
|
Effective ultrasound (US) analysis for preliminary breast tumor diagnosis is constrained due to the presence of complex echogenic patterns. Implementing pretrained models of convolutional neural networks (CNNs) which mostly focuses on natural images and using transfer learning seldom gives good results in medical domain. In this work, a CNN architecture, StepNet, with step-wise incremental convolution layers for each downsampled block was developed for classification of breast tumors as benign/malignant. To increase noise robustness and as an improvement over existing methodologies, neutrosophic preprocessing was performed, and the enhanced images were appended to the original image during training and data augmentation. The final layers’ activation maps are clustered using fuzzy c-means clustering which qualify as a validation method for the prediction of StepNet. Using neutrosophic preprocessing alone had increased the validation accuracy from 0.84 to 0.93, while using neutrosophic preprocessing and augmentation had increased the accuracy to 0.98. StepNet has comparably less training and validation time than other state of the art architectures and methods and shows an increase in prediction accuracy even for challenging isoechoic and hypoechoic tumors.
Graphical abstract
中文翻译：
一种新的 CNN 架构，用于通过基于激活图聚类的预测验证对超声乳腺肿瘤图像进行有效分类
由于存在复杂的回声模式，用于初步乳腺肿瘤诊断的有效超声 (US) 分析受到限制。实现卷积神经网络 (CNN) 的预训练模型，主要关注自然图像，使用迁移学习在医学领域很少有好的结果。在这项工作中，开发了一种 CNN 架构 StepNet，每个下采样块都有逐步增量卷积层，用于将乳腺肿瘤分类为良性/恶性。为了提高噪声鲁棒性并作为对现有方法的改进，进行了中智预处理，并在训练和数据增强期间将增强图像附加到原始图像。最后层的激活图使用模糊 c 均值聚类进行聚类，这有资格作为 StepNet 预测的验证方法。单独使用中智预处理将验证准确度从 0.84 提高到 0.93，而使用中智预处理和增强将准确度提高到 0.98。StepNet 的训练和验证时间比其他最先进的架构和方法要短，并且即使对于具有挑战性的等回声和低回声肿瘤，预测准确性也有所提高。
京公网安备 11010802027423号