Skip to main content
Fig. 1 | Insights into Imaging

Fig. 1

From: Efficient pulmonary nodules classification using radiomics and different artificial intelligence strategies

Fig. 1

Full preprocessing of lung nodule images. a The raw CT imaging data obtained from TCIA and it refers to all CT slices for each patient in 512 × 512 pixels. b Refers to the slice of interest, which was extracted by auto cropping algorithm based on reported annotations of radiologists. c Illustrates the cropping process as carried out for the nodules in 64 × 64 image size. d Otsu threshold method was used to partially remove non-nodular lung tissues using thresholding. e The role of Photo Scissors editor comes to remove manually any remaining non-nodule structure from the images

Back to article page