An effective approach for CT lung segmentation using mask region-based convolutional neural networks

作者：

Highlights：

• This study presents a new approach for automatic segmentation of lung CT scans.

• An original method for image processing using CNNs as kernels.

• It comprehends a new model based on deep learning using Mask R-CNN combined with supervised and unsupervised machine learning methods.

• Our model achieved high quality segmentation results that are better than the results generated by using the standard Mask R-CNN itself.

摘要

•This study presents a new approach for automatic segmentation of lung CT scans.•An original method for image processing using CNNs as kernels.•It comprehends a new model based on deep learning using Mask R-CNN combined with supervised and unsupervised machine learning methods.•Our model achieved high quality segmentation results that are better than the results generated by using the standard Mask R-CNN itself.

论文关键词：Image segmentation lung,Mask R-CNN,Machine learning

论文评审过程：Received 14 July 2019, Revised 6 December 2019, Accepted 2 January 2020, Available online 8 January 2020, Version of Record 16 January 2020.

论文官网地址：https://doi.org/10.1016/j.artmed.2020.101792