Deep learning models for COVID-19 infected area segmentation in CT images

Recent studies indicated that detecting radiographic patterns on CT chest scans can yield high sensitivity and specificity for COVID-19 detection. In this work, we scrutinize the effectiveness of deep learning models for semantic segmentation of pneumonia infected area segmentation in CT images for the detection of COVID-19. We explore the efficacy of U-Nets and Fully Convolutional Neural Networks in this task using real-world CT data from COVID-19 patients. The results indicate that Fully Convolutional Neural Networks are capable of accurate segmentation despite the class imbalance on the dataset and the man-made annotation errors on the boundaries of symptom manifestation areas, and can be a promising method for further analysis of COVID-19 induced pneumonia symptoms in CT images.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Zhiqiang He,et al.  Towards Efficient COVID-19 CT Annotation: A Benchmark for Lung and Infection Segmentation , 2020, ArXiv.

[3]  Yaozong Gao,et al.  Lung Infection Quantification of COVID-19 in CT Images with Deep Learning , 2020, ArXiv.

[4]  L. Xia,et al.  Coronavirus Disease 2019 (COVID-19): Role of Chest CT in Diagnosis and Management. , 2020, AJR. American journal of roentgenology.

[5]  Jennifer Couzin-Frankel Meredith Wadman,et al.  How does coronavirus kill? Clinicians trace a ferocious rampage through the body, from brain to toes , 2020 .

[6]  K. Cao,et al.  Artificial Intelligence Distinguishes COVID-19 from Community Acquired Pneumonia on Chest CT , 2020, Radiology.

[7]  Guang Yang,et al.  Automatic Brain Tumor Detection and Segmentation Using U-Net Based Fully Convolutional Networks , 2017, MIUA.

[8]  Jun Zhou,et al.  Chest CT findings of COVID-19 pneumonia by duration of symptoms , 2020, European Journal of Radiology.

[9]  Li Chen,et al.  COVID-19 CT Lung and Infection Segmentation Dataset , 2020 .

[10]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[11]  Sébastien Ourselin,et al.  Real-Time Segmentation of Non-rigid Surgical Tools Based on Deep Learning and Tracking , 2016, CARE@MICCAI.

[12]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Hao Li,et al.  Real-Time Facial Segmentation and Performance Capture from RGB Input , 2016, ECCV.

[14]  T. Fan,et al.  CT imaging and clinical course of asymptomatic cases with COVID-19 pneumonia at admission in Wuhan, China , 2020, Journal of Infection.

[15]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[16]  Nikolaos Doulamis,et al.  Deep Learning for Computer Vision: A Brief Review , 2018, Comput. Intell. Neurosci..

[17]  Anastasios Doulamis,et al.  Sample selection algorithms for credit risk modelling through data mining techniques , 2019 .

[18]  Nikolaos F. Matsatsinis,et al.  Semi-supervised vision-based maritime surveillance system using fused visual attention maps , 2016, Multimedia Tools and Applications.

[19]  Junwei Su,et al.  Deep learning system to screen coronavirus disease 2019 pneumonia , 2020 .

[20]  Y. Hu,et al.  Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China , 2020, The Lancet.

[21]  Nuno Fernandes,et al.  Economic Effects of Coronavirus Outbreak (COVID-19) on the World Economy , 2020, SSRN Electronic Journal.

[22]  Yuan Gao,et al.  Weakly Supervised Deep Learning for COVID-19 Infection Detection and Classification From CT Images , 2020, IEEE Access.

[23]  Tania Stathaki,et al.  Automatic crack detection for tunnel inspection using deep learning and heuristic image post-processing , 2018, Applied Intelligence.

[24]  Hayit Greenspan,et al.  Rapid AI Development Cycle for the Coronavirus (COVID-19) Pandemic: Initial Results for Automated Detection & Patient Monitoring using Deep Learning CT Image Analysis , 2020, ArXiv.