White blood cell detection and identification method, device, computer storage medium and electronic equipment

Abstract

The invention discloses a white blood cell detection and identification method, device, computer storage medium and electronic equipment, belonging to the field of blood cell detection. The present invention firstly performs full-field photography on a large number of blood slides to establish a slide scan image group. The training and validation datasets were prepared for the image restoration model, image segmentation and multi-task classification model, respectively, from the set of slide scan images. Using deep learning and convolutional neural network technology for model training, and optimizing the model through continuous parameter tuning and error analysis, the final image restoration model, image segmentation and multi-task classification model are obtained. During the application process, the blood glass slide to be tested is also subjected to full-field photography to establish a slide scan image. The restored clear slide scan image is obtained through the image restoration model, and the location and category of the white blood cells are obtained through image segmentation and multi-task classification model. The invention can quickly, comprehensively, accurately and objectively realize the detection and identification of white blood cells.

Description

technical field [0001] The invention relates to the field of blood cell detection, in particular to a white blood cell detection and identification method, device, computer storage medium and electronic equipment. Background technique [0002] Blood cells are roughly divided into three categories: red blood cells, white blood cells, and platelets, and white blood cells include various subtypes. In the field of biological analysis, especially in the field of clinical test analysis, accurate detection and identification of different subtypes of leukocytes plays a very important role in diagnosis and research, and the changes in the content of each subtype of leukocytes are clinically used for diagnosis and treatment. important indicators. [0003] The current blood testing process in the hospital is: blood sample—blood analyzer—dyeing machine—manual microscopic examination. The whole process takes 60 minutes. Blood samples are obtained by manually drawing blood from patients...

AI Technical Summary

Problems solved by technology

[0005] However, the manual microscope inspection operation is not only time-consuming, but not suitable for the screening of large batches of specimens; moreover, due to the influence of the experience of the cytomorphological test workers and the number of jobs, the objectivity of the judgment results will be insufficient. Resulting false-negative results may cause patients to miss treatment or even misdiagnose

[0006] Therefore, the existing white blood cell detection and identification methods have the following problems: (1) manual counting of white blood cells under a microscope is inefficient, and the operation is cumbersome and boring; (2) the accuracy of manual statistical results is low, and it is easily interfered by subjective experience and human factors

Images