A Novel Unsupervised COVID-19 Lesion Segmentation from CT Images Based-on the Lung Tissue Detection

Abstract

Image segmentation plays a significant role in quantitative image analysis. Lung segmentation of CT images has received more importance in fighting against COVID-19. In this work, a novel unsupervised framework was developed for COVID-19 infectious lesion segmentation from CT images without using annotation data. For lung segmentation, 450 normal subjects and 450 COVID cases were separately used for supervised training of a residual network (DL-Covid and DL-Norm). The lung masks predicted by both models were in the form of voxel-vise probability maps. For COVID lesion segmentation, DL-Covid and DL-Norm models were applied to COVID CT images. The DL-Norm model (trained with normal CT images) is only familiar with the healthy lung tissues and would assign low probabilities to the COVID infections. However, the DL-Covid model is familiar with COVID infections as well as healthy lung tissues. Accordingly, the lung lesion probability maps were generated by subtraction of the two lung probability maps predicted by the DL-Covid and DL-Norm. The performance of the infection segmentation framework was assessed on 50 COVID CT images considering the manual lesion segmentation as reference. Different parameters such as Dice coefficients, Jaccard index (JC), false-positive, and false-negative ratios were calculated. Lung segmentation in normal and COVID subjects resulted in Dice coefficients of 0.985 ± 0.003 and 0.978 ± 0.010, respectively. Quantitative analysis of COVID lesion segmentation revealed the Dice coefficient and JC of 0.67 ± 0.033 and 0.60 ± 0.06, respectively. Furthermore, a false-positive ratio of 0.072 ± 0.049 and a false-negative ratio of 0.062 ± 0.042 were obtained for the COVID lesion segmentation. The proposed unsupervised approach for COVID-19 infection segmentation showed satisfying performance. The outcome of this approach could be employed in supervised deep learning algorithms with noisy labels or weakly annotated data to achieve higher accuracy of the lung lesion segmentation.

Ähnliche Arbeiten