Pulmonary nodule classification method based on multiple views, multiple scales and multiple tasks
A classification method and pulmonary nodule technology, applied in the field of medical image processing, can solve problems such as unbalanced data in joint multi-task classification
Active Publication Date: 2020-05-12
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 7 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The present invention provides a multi-view, multi-scale, and multi-task pulmonary nodule classification method for classifying multiple semantic features of pulmonary nodules and solving the problem of unbalanced joint multi-task classification data
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0049] Embodiment 1: as Figure 1-5 As shown, based on the multi-view, multi-scale, multi-task pulmonary nodule classification method, the specific steps of the multi-view, multi-scale, multi-task pulmonary nodule classification method are as follows:
[0050] Step1, extract 2D nodule slices of 9 views from the 3-D view;
[0051] Further, the specific steps of the step Step1 are:
[0052] Step1.1, extract the 40X40X40mm cube centered on the nodule for each candidate nodule, the size of the cube is selected to include all nodule information and include sufficient background information;
[0053] Among them, the proposed CAD system is trained and verified using a large public dataset, the Lung Image Database Consortium (LIDC-IDRI). LIDC-IDRI contained 1018 heterogeneous cases from 7 institutions. The slice thickness of the CT images ranged from 0.6 mm to 5.0 mm, with a median of 2.0 mm. For each example image, a two-stage diagnostic annotation was performed by 4 experienced ...
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
Login to View More Abstract
The invention relates to a pulmonary nodule classification method based on multiple views, multiple scales and multiple tasks, and belongs to the technical field of medical image processing. The method comprises the following steps of firstly, extracting the 2D nodule slices of 9 views from a 3-D view; extracting the patches of 10 mm, 20 mm and 40 mm on the 2D nodule slices; constructing three convolutional neural network models, using the patches extracted from each planar view for training according to scales, and then fusing the full connection layers of the three models for the feature fusion of all images of pulmonary nodules; and performing joint training on the semantic features of the pulmonary nodules in the full connection layer to obtain a classification result of the semantic features. According to the method, a plurality of semantic features of the pulmonary nodule are classified, and finally, the semantic features of the pulmonary nodule are accurately marked by combiningmultiple classifications.
Description
technical field [0001] The invention relates to a method for classifying pulmonary nodules based on multi-view, multi-scale and multi-task, and belongs to the technical field of medical image processing. Background technique [0002] Lung cancer is one of the leading causes of cancer death in the world. In 2019, the estimated new cases of lung cancer will account for 13% of all new cases of cancer (a total of 45 cancers) in the United States, while the mortality rate is as high as 63%. The US National Lung Screening Trial (NLST) showed that the use of low-dose CT screening can reduce lung cancer mortality. Therefore, the detection and diagnosis of early lung cancer are very important for later treatment. With the advancement of screening technology, the detection rate of nodules continues to increase, but it is highly subjective for radiologists to make a diagnosis of pulmonary nodules based on their own level and experience. [0003] Providing meaningful diagnostic feature...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V2201/03G06N3/045G06F18/241
Inventor 黄青松张帅威刘利军冯旭鹏
Owner KUNMING UNIV OF SCI & TECH