Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Network construction method and system for thyroid tumor cytology smear image classification

A technology for thyroid tumors and construction methods, applied in the direction of neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as lack, lack of diagnostic experience, inability to analyze benign and malignant thyroid cytology smears, and reduce workload , avoid disk space, increase speed effect

Active Publication Date: 2018-09-21
FUDAN UNIV SHANGHAI CANCER CENT +1
View PDF3 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the lack of relevant cytopathology professionals and lack of diagnostic experience, many hospitals cannot make accurate benign and malignant analysis of thyroid cytology smears

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
  • Network construction method and system for thyroid tumor cytology smear image classification
  • Network construction method and system for thyroid tumor cytology smear image classification
  • Network construction method and system for thyroid tumor cytology smear image classification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0089] Embodiment 1 classifies with the neural network constructed by VGG-16

[0090] Obtain a certain size of thyroid tumor cytology smear images marked with benign and malignant

[0091] 1. Obtain photomicrographs of thyroid cytology smears

[0092] The data set in this example was collected from patients with thyroid nodules by the Cancer Hospital Affiliated to Fudan University. The hospital conducts a puncture examination of thyroid tumors on patients suspected of having malignant thyroid nodules, obtains thyroid tumor cell samples, conducts smear tests on them, obtains micrographs, and marks them as benign and malignant.

[0093] The micrographs of these cell smears were all at the same magnification of 400×; the dataset contained 159 malignant micrographs and 120 benign micrographs, each from a different patient.

[0094] 2. Extracting discriminative regions from photomicrographs

[0095] Multiple 224×224 pictures were cut from each photomicrograph of thyroid cytology...

Embodiment 2

[0116] Embodiment 2 classifies with the neural network constructed by Inception V3

[0117] The construction and classification method of this embodiment is the same as that of Embodiment 1, the only difference is that when using Inception V3, the image is enlarged to 299×299 before input.

Embodiment 3V

[0118] Comparison of the neural network constructed by embodiment 3VGG-16 and Inception V3

[0119] Test the accuracy of the two network models with the test set in the above embodiment. The test accuracy can accurately reflect the effect of the two convolutional neural networks on the classification task of thyroid tumor fine-needle aspiration cytology smear images. In addition, this embodiment also counts the sensitivity, specificity, positive predictive value, and negative predictive value of the two methods, and the results are shown in Table 1.

[0120] Table 1 The effect of VGG-16 and Inception V3 on the test set

[0121]

[0122] As can be seen from Table 1, the accuracy of VGG-16 on the test set is very high, reaching 97.66%. The effect of Inceptionv3 is relatively poor, but it also reached 92.75%. This shows that the two kinds of neural networks in the present invention have achieved good results in image analysis of thyroid tumor cytology smears.

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 discloses a network construction method and system for thyroid tumor cytology smear image classification. According to the system, an intensive learning method is used to find an existing convolutional neural network which is most suitable for thyroid tumor cytology smear image classification. The specific process of the intensive learning method comprises a step of generating a convolutional neural network by using a circulating neural network, a step of training the convolutional neural network by using a thyroid tumor cytology smear image training set, a step of verifying theaccuracy of the trained convolutional neural network by using a thyroid tumor cytology smear image verification set, setting an accuracy threshold, and determining whether the accuracy is higher thanthe threshold, and a step of taking a convolutional neural network with highest accuracy as a preliminary convolutional neural network and training the network again. The purposes of constructing theconvolutional neural network with high accuracy to assist a doctor in the diagnosis of thyroid tumors and improving the accuracy of diagnosis are achieved.

Description

technical field [0001] The present invention relates to the field of image recognition, in particular to a deep learning and reinforcement learning-based convolutional neural network suitable for classification of fine-needle aspiration cytology smear images of thyroid tumors and its construction method and system. Background technique [0002] Thyroid cancer is the most common malignant tumor of the endocrine system. Thyroid nodules refer to tumors in the thyroid gland. In view of the high incidence of thyroid nodules, thyroid malignancies only account for a very small part of them. If as many thyroid malignancies as possible can be identified through non-surgical methods, then Significantly reducing the number of unnecessary diagnostic operations can not only reduce the damage to patients caused by the operation, but also make more reasonable use of limited medical resources. Thyroid tumor fine-needle aspiration cytology smear is currently the most accurate and cost-effec...

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 Applications(China)
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/082G06V20/693G06V20/695G06V20/698G06N3/044
Inventor 向俊卢宏涛官青王蕴珺平波万晓春李端树杜佳俊秦宇
Owner FUDAN UNIV SHANGHAI CANCER CENT
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com