Multi-focal-length video acquisition method and system for urine red blood cells

Abstract

The invention belongs to the technical field of artificial intelligence assisted medical treatment, and particularly relates to a urine red blood cell multi-focal-length video acquisition method and system. The method comprises the following steps: S1, continuously adjusting the focal length of a microscope, and collecting a microscope zoom video of a sample; S2, performing definition calculation on each frame of image in the zoom video, wherein the definition calculation adopts an improved DCT image definition evaluation function to perform calculation; S3, drawing a definition change curve of the zoom video according to the definition value obtained through calculation; and in the definition value change curve, taking the corresponding image focal length when the curve slope is maximum as the starting point focal length of the collected video, taking the corresponding image focal length when the curve slope is minimum as the ending point focal length of the collected video, and collecting and storing the zoom video of the microscope. According to the invention, multi-focal-length video acquisition is carried out by using the evaluation function of the image, so that the acquisition precision and efficiency are improved.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence-assisted medical treatment, and in particular relates to a method and system for collecting urine red blood cell multi-focus video. Background technique [0002] The clinical manifestations of patients with kidney disease usually include changes in urine. Generally, it can be observed through a microscope that in addition to normal urine red blood cells, there are also a large number of blood cells formed by the extrusion of glomeruli and other tissues. Abnormal urine red blood cells. Doctors usually diagnose kidney disease by collecting and looking at a sample of a person's urine. Among them, some abnormal urinary red blood cells will show different morphological characteristics under different focal lengths of the microscope, and may even have similar morphological characteristics to other types of abnormal urinary red blood cells at a certain focal length. For the identificati...

AI Technical Summary

Problems solved by technology

For the identification of such abnormal urine red blood cells, if the microscope focal length is not suitable for observation, it is very easy to misidentify such abnormal urine red blood cells

In addition, the morphological characteristics of some abnormal urinary red blood cells are also very similar to each other, which is very easy to be confused when identifying

[0003] At present, there are published invention patents that propose a classification and statistics method and system for abnormal urine red blood cells, and propose the concept of multi-focus video data of urine red blood cells under a microscope, but the video data used are manually collected

Manual data collection usually has many uncontrollable factors, such as changes in the external contrast environment when collecting data, camera exposure adjustment, etc.

In the process of collecting multi-focal length video data, the more important problem is that it is difficult to judge the focal length corresponding to the correct abnormal urine red blood cell characteristics when collecting data manually, and the judgment standard is subjective, and it is impossible to obtain a unified standard. High-resolution multi-focal video data for accurate characterization of abnormal urine red blood cells

[0004] In the prior art, there is no video data acquisition equipment and system that can automatically analyze and collect multi-focus videos of urine red blood cells under a microscope

Images