Automatic abnormal cell recognition method based on image splicing

A method, comprising:

• 1) scanning a slide using a digital pathological scanner and obtaining a cytological slide image;
• 2) obtaining a set of centroid coordinates of all nuclei that is denoted as CentroidOfNucleus by automatically localizing nuclei of all cells in the cytological slide image using a feature fusion based localizing method;
• 3) obtaining a set of cell square region of interest (ROI) images that are denoted as ROI_images, wherein each square cell ROI image corresponds to a cell in the slide, and is determined with a Centroid_i as a center of the each square cell ROI image and a fixed length L as a side length of the each square cell ROI image;
• 4) grouping all cell images in the ROI images into different groups based on sampling without replacement, wherein each group contains ROW×COLUMN cell images with preset ROW and COLUMN parameters;
• obtaining a set of splice images wherein a splice image with length being ROW×L and width being COLUMN×L is generated by randomly splicing the ROW×COLUMN cell images in the same group into ROW rows and COLUMN columns;
• and
• 5) classifying all cell images in the splice image simultaneously by using the splice image as an input of a trained deep neural network;
• and recognizing cells classified as abnormal categories.