Hybrid COVID-19 segmentation and recognition framework (HMB-HCF) using deep learning and genetic algorithms

作者：

Highlights：

• Proposing a hybrid COVID-19 framework with a hybrid hierarchy working mechanism.

• Suggesting a lung segmentation algorithm using computerized tomography images.

• Designing an abstract CNN model and working with pre-trained transfer learning models.

• Suggesting a combined DL and GA overall algorithm for learning and optimization.

• Studying the effects of regularization, optimization, dropout, and data augmentation.

• Reporting the state-of-the-art performance metrics compared with other related works.

摘要

•Proposing a hybrid COVID-19 framework with a hybrid hierarchy working mechanism.•Suggesting a lung segmentation algorithm using computerized tomography images.•Designing an abstract CNN model and working with pre-trained transfer learning models.•Suggesting a combined DL and GA overall algorithm for learning and optimization.•Studying the effects of regularization, optimization, dropout, and data augmentation.•Reporting the state-of-the-art performance metrics compared with other related works.

论文关键词：Classification,Convolutional neural network (CNN),COVID-19,Data augmentation (DA),Deep learning (DL),Genetic algorithms (GA),Optimization,Transfer learning (TL)

论文评审过程：Received 15 February 2021, Revised 8 August 2021, Accepted 17 August 2021, Available online 28 August 2021, Version of Record 7 September 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102156