Balaha Hossam Magdy, Balaha Magdy Hassan, Ali Hesham Arafat
Computer Engineering and Control Systems Department, Faculty of Engineering, Mansoura University, Egypt.
Obstetrics and Gynecology, Faculty of Medicine, Tanta University, Egypt.
Artif Intell Med. 2021 Sep;119:102156. doi: 10.1016/j.artmed.2021.102156. Epub 2021 Aug 28.
COVID-19 (Coronavirus) went through a rapid escalation until it became a pandemic disease. The normal and manual medical infection discovery may take few days and therefore computer science engineers can share in the development of the automatic diagnosis for fast detection of that disease. The study suggests a hybrid COVID-19 framework (named HMB-HCF) based on deep learning (DL), genetic algorithm (GA), weighted sum (WS), and majority voting principles in nine phases. Its segmentation phase suggests a lung segmentation algorithm using X-Ray images (named HMB-LSAXI) for extracting lungs. Its classification phase is built from a hybrid convolutional neural network (CNN) architecture using an abstractly-designed CNN (named HMB1-COVID19) and transfer learning (TL) pre-trained models (VGG16, VGG19, ResNet50, ResNet101, Xception, DenseNet121, DenseNet169, MobileNet, and MobileNetV2). The hybrid CNN architecture is used for learning, classification, and parameters optimization while GA is used to optimize the hyperparameters. This hybrid working mechanism is combined in an overall algorithm named HMB-DLGA. The study experiments implemented the WS approach to evaluate the models' performance using the loss, accuracy, F1-score, precision, recall, and area under curve (AUC) metrics with different pre-defined ratios. A collected, combined, and unified X-Ray dataset from 8 different public datasets was used alongside the regularization, dropout, and data augmentation techniques to limit the overall overfitting. The applied experiments reported state-of-the-art metrics. VGG16 reported 100% WS metric (i.e., 0.0097, 99.78%, 0.9984, 99.89%, 99.78%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the highest WS. It also reported a 99.92% WS metric (i.e., 0.0099, 99.84%, 0.9984, 99.84%, 99.84%, and 0.9996 for the loss, accuracy, F1, precision, recall, and AUC respectively) concerning the last reported WS result. HMB-HCF was validated on 13 different public datasets to verify its generalization. The best-achieved metrics were compared with 13 related studies. These extensive experiments' target was the applicability verification and generalization.
新冠病毒病（COVID-19）迅速升级，直至成为一种大流行病。常规的人工医学感染检测可能需要数天时间，因此计算机科学工程师可以参与到自动诊断技术的开发中，以实现对该疾病的快速检测。该研究提出了一种基于深度学习（DL）、遗传算法（GA）、加权和（WS）以及多数投票原则的混合式COVID-19框架（名为HMB-HCF），共分为九个阶段。其分割阶段提出了一种使用X射线图像的肺部分割算法（名为HMB-LSAXI）来提取肺部。其分类阶段由一个混合卷积神经网络（CNN）架构构建而成，该架构使用了一个抽象设计的CNN（名为HMB1-COVID19）以及迁移学习（TL）预训练模型（VGG16、VGG19、ResNet50、ResNet101、Xception、DenseNet121、DenseNet169、MobileNet和MobileNetV2）。混合CNN架构用于学习、分类和参数优化，而GA则用于优化超参数。这种混合工作机制被整合到一个名为HMB-DLGA的整体算法中。该研究实验采用WS方法，使用损失、准确率、F1分数、精确率、召回率和曲线下面积（AUC）指标，以不同的预定义比例来评估模型的性能。使用从8个不同公共数据集中收集、合并和统一的X射线数据集，同时采用正则化、随机失活和数据增强技术来限制整体过拟合。所进行的实验报告了先进的指标。VGG16在最高WS方面报告了100%的WS指标（即损失、准确率、F1、精确率、召回率和AUC分别为0.0097、99.78%、0.9984、99.89%、99.78%和0.9996）。它还在最后报告的WS结果方面报告了99.92%的WS指标（即损失、准确率、F1、精确率、召回率和AUC分别为0.0099、99.84%、0.9984、99.84%、99.84%和0.9996）。HMB-HCF在13个不同的公共数据集上进行了验证，以检验其泛化能力。将取得的最佳指标与13项相关研究进行了比较。这些广泛实验的目标是验证适用性和泛化能力。
