Fig. 6From: Segmentation of lung parenchyma in CT images using CNN trained with the clustering algorithm generated datasetExamples of segmentation at axial slices (In each sub-figure, there are three columns: the left the left column shows the results of manual segmentation of lung field including both lung parenchyma and lesions; the middle column presents the results of segmentation of lung parenchyma by our proposed method; the right column shows the results after a “hole-filling” operation from those by our method.). a Three example slices (three rows) for subjects with lung cancer (PET-CT scanner). b Three example slices (three rows) for special cases where the pleural effusion, emphysema, inflammation are near the boundary the lung fieldBack to article page