Reticulocyte classification counting model construction method and application

Abstract

The invention relates to a reticulocyte classification and counting model construction method and application. A reticulocyte classification and counting model is constructed and used for classifyingand respectively counting reticulocytes and other common erythrocytes; the existing mature data set is taken as a first database; a plurality of blood smear micrographs are obtained, and a second database is formed after outline and category labeling of common red blood cells and reticulocytes is conducted; and samples are selected from the first database and the second database to complete two-stage training. Through the reticulocyte classification model, erythrocytes are automatically and efficiently classified and counted, and compared with manual statistics, the efficiency is qualitativelyimproved. The method is fair and objective, eliminates man-made and subjective factors, and has no interference of fatigue and other objective factors. The model has a self-learning attribute, the training efficiency of the recognition model is gradually improved along with the increase of high-quality labeled pictures, and the recognition and classification accuracy of the model can be continuously optimized.

Description

Technical field [0001] The invention relates to the technical field of medical images, in particular to a method for constructing a reticulocyte classification and counting model and its application. Background technique [0002] Reticulocytes are red blood cells that are not yet fully mature and are located between late young red blood cells and mature red blood cells. Reticulocyte count is an important indicator for evaluating bone marrow hematopoietic function. Counting reticulocytes is a basic experiment for evaluating bone marrow hematopoietic function in clinical blood tests, and is an important indicator in clinical testing. [0003] The basic principle of the determination of reticulocytes is to use a certain dye to bind to intracellular RNA, and then detect the stained reticulocytes by microscopic examination or using flow cytometry (FCM) and other methods. At present, most domestic hospitals use brilliant tar blue ethanol solution for staining and smearing, and directly ...

AI Technical Summary

Problems solved by technology

However, this method requires manual counting of a large number of cells under a microscope

[0004] The existing measurement methods mainly have technical problems: (1) Manual counting of reticulocytes 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