Deep learning-based bone marrow cell classification and identification method, device and system

Abstract

The invention provides a bone marrow cell classification and identification method, device and system based on deep learning, and belongs to the field of medical image processing. The method comprises the following steps: (1) acquiring a low-power lens full-slice image of a bone marrow smear by using a low-power lens; (2) cutting the low-power lens full slice image of the bone marrow smear into a plurality of low-power lens small images with the same size as the high-power lens imaging view; (3) classifying the low-power mirror small images obtained in the step (2) by using a trained image classification model to obtain low-power mirror small images with good visual fields; (4) scanning and imaging the low-power lens small image with a good visual field by using a high-power lens to obtain a corresponding high-power lens image; (5) classifying and counting bone marrow cells in the high-power lens image obtained in the step (4) by using the trained target detection model; (6) predicting blood diseases according to classification and counting results of the bone marrow cells. Therefore, identification and classification of bone marrow cells can be reliably and automatically realized, and the accuracy of blood disease prediction can be improved.

Description

technical field [0001] The invention relates to a method, device and system for classification and recognition of bone marrow cells based on deep learning, and belongs to the technical field of medical image processing. Background technique [0002] Microscopic examination and classification of blood cells has always been an important basis for hematological diagnosis. Morphological examination of white blood cells in peripheral blood and bone marrow samples is the initial step in the diagnosis of blood diseases (such as acute myeloid leukemia, acute lymphoblastic leukemia, etc.), among which, the most commonly used acute leukemia classification method FAB method is strongly dependent on cell morphology. [0003] Morphological analysis of white blood cells in bone marrow is crucial to the diagnosis of blood diseases, but so far, in practical applications, the morphological examination of cells in bone marrow smears still relies on manual microscopy. Manual microscopic exam...

AI Technical Summary

Problems solved by technology

Manual microscopic examination requires the observer to have certain professional knowledge and experience, and requires a lot of manpower and time, which is cumbersome and time-consuming, and long-term observation can easily lead to eye fatigue and human errors; These often vary with different observers, resulting in strong subjectivity of manual microscopic examination results and difficult to achieve standardization. These factors will cause the accuracy of classification and counting to be disturbed, thereby reducing the reliability and stability of microscopic examination results.

[0004] To sum up, the current morphological examination of bone marrow cells relying on manual microscopic examination is not only tedious and time-consuming, but also has strong subjectivity and low reliability in the results of microscopic examination. Accuracy of blood disease predictions was also lower

Images