Computers in Biology and Medicine ( IF 7.7 ) Pub Date : 2020-07-15 , DOI: 10.1016/j.compbiomed.2020.103917 D Osaku 1 , C F Cuba 1 , C T N Suzuki 1 , J F Gomes 1 , A X Falcão 1
Intestinal parasites are responsible for several diseases in human beings. In order to eliminate the error-prone visual analysis of optical microscopy slides, we have investigated automated, fast, and low-cost systems for the diagnosis of human intestinal parasites. In this work, we present a hybrid approach that combines the opinion of two decision-making systems with complementary properties: () a simpler system based on very fast handcrafted image feature extraction and support vector machine classification and () a more complex system based on a deep neural network, Vgg-16, for image feature extraction and classification. is much faster than , but it is less accurate than . Fortunately, the errors of are not the same of . During training, we use a validation set to learn the probabilities of misclassification by on each class based on its confidence values. When quickly classifies all images from a microscopy slide, the method selects a number of images with higher chances of misclassification for characterization and reclassification by . Our hybrid system can improve the overall effectiveness without compromising efficiency, being suitable for the clinical routine — a strategy that might be suitable for other real applications. As demonstrated on large datasets, the proposed system can achieve, on average, 94.9%, 87.8%, and 92.5% of Cohen’s Kappa on helminth eggs, helminth larvae, and protozoa cysts, respectively.
中文翻译:
肠道寄生虫的自动诊断:一种新的混合方法及其好处。
肠道寄生虫是人类几种疾病的原因。为了消除光学显微镜载玻片易于出错的视觉分析,我们研究了用于诊断人类肠道寄生虫的自动化,快速且低成本的系统。在这项工作中,我们提出了一种混合方法,该方法结合了两个具有互补属性的决策系统的意见:)基于非常快速的手工图像特征提取和支持向量机分类的更简单系统,以及()基于深度神经网络Vgg-16的更复杂的系统,用于图像特征提取和分类。 比 ,但准确性不如 。幸运的是, 与...不同 。在训练期间,我们使用验证集来了解错误分类的可能性在每个类别上根据其置信度值。什么时候 快速地从显微镜载玻片上对所有图像进行分类,该方法选择了许多错误分类几率较高的图像,以进行表征和重新分类 。我们的混合系统可以在不影响效率的情况下提高整体效率,适合临床常规操作-一种可能适合其他实际应用的策略。如大型数据集所示,该系统在蠕虫卵,蠕虫幼虫和原生动物囊肿上平均可分别达到Cohen's Kappa的94.9%,87.8%和92.5%。