Multi-region scanning method based on digital pathology slice scanner

Abstract

The invention discloses a multi-region scanning method based on a digital pathology slice scanner. According to training data prepared in advance, training is conducted in advance to form an identification model used for identifying scanned regions required to be scanned on a digital pathology slice image. The method specifically includes the following steps that a scanning camera photographs a pathology slice and obtains the corresponding digital pathology slice image; the identification model is adopted for identifying the digital pathology slice image to identify the scanned regions; the pathology slice scanner is adopted for scanning the scanned regions, and finally pathology slice scanning files are obtained and automatically saved. Compared with the prior art, the method has the advantages that multiple scanning regions of the digital pathology slice image can be simultaneously selected for scanning, compression and integration are conducted on the output scanning files, the workefficiency is improved, the output image width is reduced, and the practicability of the digital pathology slice scanner is improved.

Description

technical field [0001] The invention relates to an image scanning method, in particular to a multi-region scanning method based on a digital pathological slide scanner. Background technique [0002] Existing digital pathological slice scanners first take a group of panoramic images of pathological slices when scanning sample tissue, and then technicians select the sample area that needs to be scanned in depth through the platform software according to the inspection needs, and the digital pathological slice scanner then scans the selected area. The area is focused and scanned to obtain the final scanned file. [0003] Usually, as figure 1 As shown, the sample to be detected on the pathological slide is a tissue, which is concentrated in one area, so most digital pathological slide scanners currently on the market only support single-area scanning, that is to say, technicians can only select the area of ​​the panoramic image. A single area is scanned in depth. However, in ...

AI Technical Summary

Problems solved by technology

[0003] Usually, as figure 1 As shown, the sample to be detected on the pathological slide is a tissue, which is concentrated in one area, so most digital pathological slide scanners currently on the market only support single-area scanning, that is to say, technicians can only select the area of ​​the panoramic image. Deep scan of a single area

However, in actual operation, some morphologically dispersed tissues will appear on pathological sections, such as figure 2 As shown, if the usual scanning method is still followed, the digital pathology slide scanner will scan multiple scattered tissues into the same area as a whole, such as image 3 As shown, it can be seen that many blank areas are left between multiple tissues scanned into the same area, and these blank scans are meaningless for the detection of tissue samples, which not only increases the scanning time, but also increases the output image size , wasting computer storage space

Images