Check patentability & draft patents in minutes with Patsnap Eureka AI!

Error correction method and system in automatic marking of individual cells of bone marrow image

An automatic labeling and error correction method technology, applied in the field of medical image processing, can solve the problems of time-consuming and energy-consuming, difficult to improve the accuracy of the bone marrow cell automatic labeling model, and low quality of the bone marrow cell public data set, so as to improve the accuracy. Effect

Active Publication Date: 2021-09-17
UNIV OF JINAN
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Existing public datasets of bone marrow cells are of poor quality
For example, the ALL-IDB1 database consists of 109 2592*1944 and 260 257*257 bone marrow pathological images, but it can only be used to identify blast cells in bone marrow cells, and cannot be directly applied to the classification and quantification of bone marrow cells; The bone marrow pathology image dataset provided by the image bank online website of the American Society of Hematology can only be used to identify normal and abnormal pathological images
The inventors found that the reason for this phenomenon is that the construction of the bone marrow cell data set is highly dependent on the manual labeling of pathologists, and consumes a lot of time and energy for doctors, and its labeling quality is also easily affected by doctors’ subjective
In this case, the accuracy of the automatic labeling model of bone marrow cells is difficult to improve

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Error correction method and system in automatic marking of individual cells of bone marrow image
  • Error correction method and system in automatic marking of individual cells of bone marrow image
  • Error correction method and system in automatic marking of individual cells of bone marrow image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] Such as figure 1 As shown, this embodiment provides an error correction method in the automatic labeling of individual cells in bone marrow images, which specifically includes the following steps:

[0039] Step S101: Using a cell labeling model to mark bone marrow pathological images to obtain initial labeling results; the initial labeling results include initial cell types and initial labeling frame position information.

[0040] In this embodiment, the initial labeling result is obtained by using a trained cell labeling model to label bone marrow pathological images.

[0041] It can be understood that the cell labeling model is an existing network model, and its structure can be specifically selected according to the actual situation. By manually marking the box position and the type of the box, that is, the cell type, automatic labeling training is performed to obtain the cell Label the parameters of the model.

[0042] For the process of initial marking, please re...

Embodiment 2

[0069] Such as Image 6 As shown, this embodiment provides an error correction system in the automatic labeling of individual cells in bone marrow images, which specifically includes the following modules:

[0070] An image automatic labeling module 11, which is used to use a cell labeling model to label bone marrow pathological images to obtain initial labeling results; the initial labeling results include initial cell types and initial labeling frame position information;

[0071] Classification error correction confirmation module 12, which is used to use the category error corrector to carry out secondary confirmation on the initial labeling result, obtain the classification score of the cell and the cell type after error correction, and correct the label of the classification error, and obtain the first time Marking results after error correction;

[0072] Positioning error correction confirmation module 13, which is used to use the positioning error corrector to compare...

Embodiment 3

[0077] This embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in the error correction method in the automatic labeling of individual cells in bone marrow images as described in the first embodiment above are implemented .

[0078] This embodiment adopts classification error correction confirmation, positioning error correction confirmation and fuzzy error correction confirmation to correct classification errors, positioning errors and fuzzy marking problems in turn, and automatically corrects errors on the basis of marking to obtain more accurate results, so that bone marrow cells The labeling process is more automated and precise, which greatly improves the efficiency of bone marrow cell labeling and the accuracy of the automatic labeling model.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an error correction method and system in automatic marking of individual cells of a bone marrow image. The method comprises the following steps: marking a bone marrow pathological image by using a cell marking model to obtain an initial marking result; carrying out secondary confirmation on the initial marking result by utilizing a category error corrector to obtain a classification score of the cells and a cell type after error correction, and correcting marks with wrong classification to obtain a marking result after first error correction; comparing the constraint relationship between each cell marking frame in the marking result after the first correction and the corresponding cell size by using a positioning error corrector, and correcting the marking with the positioning error to obtain a marking result after the second correction; and respectively calculating the overlapping rate of the marking frame areas of any two cells in the marking result after the second error correction by using a fuzzy confirmation device, judging whether a fuzzy marking problem exists or not according to the ratio of the overlapping rate to an overlapping threshold value and the category of the marking frame, and correcting the fuzzy marking problem to obtain a final marking result after error correction.

Description

technical field [0001] The invention belongs to the field of medical image processing, in particular to an error correction method and system for automatic labeling of individual cells in bone marrow images. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] The analysis and classification statistics of various cell morphology in bone marrow pathological images are helpful to assist pathologists in diagnosing acute leukemia (ALL or AML), myelodysplastic syndrome (MDS) and multiple myeloma and other blood system diseases. By establishing an automatic labeling model of bone marrow cells based on deep learning, preliminary labeling of bone marrow cells can be achieved, but the error is still relatively large. It is of great significance to study the error correction method in the automatic labeling of bone marrow cells to optimize the pathologi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G16H30/40
CPCG16H30/40
Inventor 苏洁韩金隽
Owner UNIV OF JINAN
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com