Unlock instant, AI-driven research and patent intelligence for your innovation.

A cancer cell tracking method based on multi-feature fusion

A multi-feature fusion and cancer cell technology, which is applied in the field of cell tracking, can solve problems such as low tracking efficiency, low tracking accuracy, and difficult tracking of cancer cells, and achieve the effects of improving efficiency, improving representation, and improving accuracy

Active Publication Date: 2021-10-12
ZHEJIANG UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the difficulty of tracking cancer cells due to missed and false detections of cancer cells under a phase microscope due to problems such as high density, changeable shape, and occlusion, the tracking efficiency of existing cancer cell tracking methods is low , low tracking accuracy, the present invention provides a cancer cell tracking method based on multi-feature fusion that can effectively improve tracking efficiency and tracking accuracy

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
  • A cancer cell tracking method based on multi-feature fusion
  • A cancer cell tracking method based on multi-feature fusion
  • A cancer cell tracking method based on multi-feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention will be further described below in conjunction with the accompanying drawings.

[0031] refer to Figure 1 to Figure 6 , a cancer cell tracking method based on multi-feature fusion, including the following steps:

[0032] Step 1, detection of cancer cells, comprises the following steps:

[0033] 1.1. Make the dataset: Use the voc2007 dataset format to manually make Ground Truth (GT) on the dataset as the training set of the network;

[0034] 1.2. Generation of candidate regions: The Faster R-CNN algorithm used is to generate candidate regions through the RPN network;

[0035] 1.3. Classification of candidate regions and refinement of candidate frames: The Faster R-CNN algorithm used is to classify candidate regions and refine the position of candidate frames through the Fast R-CNN network;

[0036] Step 2, feature extraction of cancer cells, includes the following steps:

[0037] 2.1. Extraction of centroid features: After the Faster R-CNN algo...

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

A cancer cell tracking method based on multi-feature fusion, comprising the following steps: Step 1, detection of cancer cells, including the following steps: 1.1, making a data set; 1.2, generation of candidate regions; 1.3, classification of suggested regions and candidate frames refinement; step 2, feature extraction of cancer cells, including the following steps: 2.1, centroid feature extraction; 2.2, convolution feature extraction; step 3, primary tracking of cancer cells, including the following steps: 3.1, cancer cell category Judgment; 3.2. Preliminary tracking by category; Step 4, re-tracking of cancer cells, including the following steps: 4.1. Correlation matching of missed detection areas; 4.2. Correlation matching of repeated detection areas. The invention provides a cancer cell tracking method based on multi-feature fusion that effectively improves tracking efficiency and tracking accuracy.

Description

technical field [0001] The invention belongs to the field of cell tracking and designs a cancer cell tracking method based on multi-feature fusion. Specifically, the detection of cancer cells is realized through the deep learning Faster R-CNN algorithm, that is, the RPN network is used to extract the candidate area, and then the Fast R-CNN network is used to refine the position of the candidate frame and determine the target category, and then extract the detection results Finally, the multi-feature fusion tracking algorithm is used to complete the correct association and matching of cancer cells, and realize the continuous tracking of cancer cells between frames. Background technique [0002] According to the latest 15-year cancer incidence and death data in China released by the National Cancer Center, new cancer cases in China are increasing every day. The early diagnosis of cancer cells and the testing of anticancer drugs have always been a topic of great concern to the ...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/246G06T7/60G06K9/62G06N3/04
CPCG06T7/246G06T7/60G06T2207/30096G06N3/045G06F18/22
Inventor 胡海根周莉莉肖杰管秋周乾伟陈胜勇
Owner ZHEJIANG UNIV OF TECH