Leukocyte nucleus and cytoplasm automatic segmentation method and system based on deep learning

Abstract

The invention discloses a leukocyte nucleus and cytoplasm automatic segmentation method and a leukocyte nucleus and cytoplasm automatic segmentation system based on deep learning. The leukocyte nucleus and cytoplasm automatic segmentation method comprises the steps of: constructing a U-shaped neural network segmentation model, wherein the U-shaped neural network segmentation model comprises an encoder and a decoder, the encoder adopts an improved neural network structure for feature extraction, the decoder recovers details and spatial information of an image through up-sampling, and the encoder and the decoder adopt jump connection to supplement underlying information lost in the pooling process; training the U-shaped neural network segmentation model by adopting a leukocyte training set,setting a learning rate by taking leukocyte verification set loss as a monitoring index, and adjusting the learning rate when the monitoring index is not changed; and segmenting a leukocyte test set by adopting the trained U-shaped neural network segmentation model, and acquiring a segmentation result of cell nucleus and cytoplasm according to the classification probability of each pixel point ofa to-be-segmented image. Through adopting the improved U-shaped neural network segmentation model, morphological information of leukocyte nucleuses and cytoplasm can be rapidly obtained, and automaticsemantic segmentation of leukocyte nucleuses and cytoplasm is realized.

Description

technical field [0001] The present disclosure relates to the technical field of medical image processing, in particular to a deep learning-based automatic segmentation method and system for nucleoplasm of white blood cells. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] At present, there are two main methods for the identification and detection of white blood cells. One is the use of manual microscopic examination. Jobs affect the status of workers and thus the accuracy of results. Another method is to detect with the help of a blood analyzer. This method can automatically classify and count white blood cells, which greatly reduces the workload of the inspectors. However, its biggest limitation is that it can only obtain quantitative information but not cells. The morphological information of white blood cells cannot detect the morphol...

AI Technical Summary

Problems solved by technology

[0003] At present, there are two main methods for the identification and detection of white blood cells. One is the use of manual microscopic examination. Jobs can affect the status of workers and thus the accuracy of results

Another method is to detect with the help of a blood analyzer. This method can automatically classify and count white blood cells, which greatly reduces the workload of the inspectors. However, its biggest limitation is that it can only obtain quantitative information but not cells. The morphological information of white blood cells cannot detect the morphological abnormalities of white blood cells; at the same time, there are also problems such as expensive equipment and classification accuracy to be improved.

Images