Digital pathological image quality evaluation system

Abstract

The invention provides a digital pathological image quality evaluation system, and the system is characterized by comprising a digital pathological image preprocessing module; a stain detection module; a scanning quality evaluation module; and a scoring module which is used for calculating a final effective area, and all subitem detection scores given by the digital pathological image preprocessing module, the stain detection module and the scanning quality evaluation module are simultaneously collected to give an overall evaluation of the pathological image quality. According to the digital pathological image quality evaluation system provided by the invention, unified quality evaluation is performed on conventionally acquired medical pathological images, and useless information in the digital pathological images is effectively removed, so that doctors are more focused on effective pathological tissue structures, and reliable data guarantee is provided for academic research based on the medical pathological images.

Description

technical field [0001] The invention relates to an image processing method, in particular to a digital pathological image quality evaluation system. Background technique [0002] In the process of conventional slide preparation and digitalization of pathological images, noise will inevitably be introduced. The common ones are: knife marks / cracks on slides, folds / stacking of slices, bubbles / glue / doctor's manual annotation, and focus blur. These noises not only affect the judgment of clinicians, but also are not conducive to the research of researchers. For example, the current academic issues related to the intelligent interpretation of digital pathological images based on deep learning algorithms are greatly limited by the quality of digital pathological images themselves. Therefore, correct evaluation of digital pathological image quality can provide reliable guarantee for pathological diagnosis and medical academic research. [0003] At present, the existing medical patho...

AI Technical Summary

Problems solved by technology

These noises not only affect the judgment of clinicians, but also are not conducive to the research of researchers. For example, the current academic issues related to the intelligent interpretation of digital pathological images based on deep learning algorithms are greatly limited by the quality of digital pathological images.

Images