Automatic blood analysis method and system, hematology analyzer and storage medium

Abstract

The invention discloses an automatic blood analysis method and system, a blood cell analyzer and a storage medium, and the method comprises the steps: respectively obtaining microscopic images of a to-be-detected sample when the distances between an objective table and an objective lens are h0-d, h0 and h0 + d, wherein the distance h0 between the objective table and the objective lens is the focusing position of the to-be-detected sample; inputting the microscopic images of the to-be-detected samples corresponding to the distances h0-d, h0 and h0 + d into a trained feature fusion network for feature fusion, and outputting a fusion feature map; inputting the fusion feature map into a trained RPN network and a trained FCN network for frame regression, classification and masking, and outputting an instance segmentation map including a classification result and a mask; and counting the number of to-be-detected targets in the to-be-detected sample according to the classification result andthe mask. According to the invention, the problem of low accuracy of the number of blood cells counted by the cell identification and statistics method based on the optimal focusing microscopic imageis solved.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to an automatic blood analysis method, system, blood cell analyzer and computer-readable storage medium. Background technique [0002] At present, to count the blood cells in the blood sample, the inspectors count the blood cells in the blood sample based on the blood smear, mainly by placing the blood smear under the microscope, and then using the traditional machine vision method to analyze the blood smear under the microscope Cell identification and statistics were performed on the best focused microscope images observed below to obtain the number of blood cells in the final blood smear. However, since all the blood cells in the blood smear are not at the same level, and the position of the blood cells is high or low, some blood cells are not in focus in the best focus microscopic image, and the definition in the image is low. As a result, in the use of machine vision, it is eas...

AI Technical Summary

Problems solved by technology

[0003] The main purpose of the present invention is to propose an automatic blood analysis method, system, blood cell analyzer and computer-readable storage medium, aiming to solve the problem of blood cells counted by cell identification and statistical methods based on best-focus microscopic images in the prior art. Quantitative accuracy problem

Images