Intelligent prediction method and system for NSCLC lymph node metastasis risk

A technology for lymph node metastasis and intelligent prediction, applied in neural learning methods, diagnostic recording/measurement, instruments, etc., can solve problems such as lack of technology, cost of time, reagents, consumables, etc., and achieve the effect of shortening operation time and reducing the probability of tumor recurrence.

Pending Publication Date: 2022-07-12
南昌市高新区人民医院
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Because it is performed manually, it is limited by human factors, and has time and reagent consumable costs
However, there is a lack of relevant technologies for the application of artificial intelligence in NSCLC lymph node metastasis, which requires effective artificial intelligence algorithms, a large amount of image data support, and powerful graphics central processing unit (GPU), and the entire training process requires the guidance of professional clinical oncologists

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
  • Intelligent prediction method and system for NSCLC lymph node metastasis risk
  • Intelligent prediction method and system for NSCLC lymph node metastasis risk
  • Intelligent prediction method and system for NSCLC lymph node metastasis risk

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] participate Figure 1 to Figure 3 As shown, a CNN-based intelligent prediction method for the risk of lymph node metastasis in NSCLC includes the following steps:

[0054] S1: Obtain the full-field digital image (WSI) of the conventional pathological H&E-stained slides of the surgically resected T1-2 lung cancer mass, and additionally obtain the full-field digital image (WSI) of the pathological H&E-stained slides of normal lung tissue; all images Screened and confirmed by a pathologist;

[0055] S2: The pathologist marks the normal tissue area in the H&E full-field digital slice of lung cancer; the pathologist selects the full-field digital slice of normal lung tissue, and the image does not contain tumor tissue;

[0056] S3: Cut the color block on the acquired image;

[0057] S4: Divide the color patch images into a training set and a validation set according to the cases they belong to. Each set contains 1:1 tumor and non-tumor color patches. The constructed convol...

Embodiment 2

[0078] Another object of the present invention is to provide an intelligent prediction system for NSCLC lymph node metastasis risk based on CNN, including: an input terminal module, a single case processing program module, a learning optimization program module, and a report output terminal module;

[0079] The input terminal module includes performing quality control on the input full-field digital image of lung cancer, and is used to remove the normal tissue color patches next to the cancer, remove the high background color patches, and retain and reorganize the tumor tissue color patches to achieve the preprocessing effect;

[0080] The single case processing program module includes using the previously generated convolutional neural network model to analyze the image to obtain the characteristics of each characteristic color block, and finally to determine the risk;

[0081] The learning optimization program module is used to batch receive full-field digital images of lung ...

Embodiment 3

[0084] data collection:

[0085] The pathological diagnosis of lung non-small cell carcinoma and T staging ≤ T2 in the Department of Pathology, The First Affiliated Hospital of Nanchang University from January 2019 to November 2021 were collected, and all H&E stained sections were screened; and relevant clinical data of the patients were obtained. Pathological special data.

[0086] Quality control:

[0087] Severely decolorized or stained glass slides were excluded from H&E stained slides, and infection, fibrosis, and H&E stained slides affecting normal lung tissue were excluded from H&E stained slides of normal lung tissue; the visual quality of tumor tissue was excluded due to technical artifacts. Such as air bubbles, pen marks, water droplets, dust, air inclusions, hair, etc. on glass slides; all images of full-field digital slices were evaluated by three pathology experts, and images were selected 1:1 into tumor tissue and Non-tumor tissue; images with inconsistent diag...

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 an intelligent prediction method and system for an NSCLC lymph node metastasis risk. The method comprises the following steps: acquiring Hamp of a lung cancer patient; e, dyeing the full-view digital section and clinical data of the patient; performing quality control, color block processing and screening on the image data; and applying the preprocessed color blocks to a two-stage analysis process, wherein the two-stage analysis process comprises establishing a CNN model for identifying non-cancer tissues and a CNN model for predicting early-stage non-small cell lung cancer lymph node metastasis. The analysis of the two parts comprises the steps of dividing color block images into a training set and a verification set, constructing a model through a CNN algorithm, optimizing parameters of the trained CNN model, and verifying the optimized CNN model in the verification set. And finally predicting the risk of early non-small cell lung cancer lymph node metastasis by using the detected CNN model. The deep learning algorithm is adopted to judge the pathological section, so that the accuracy of tumor metastasis risk assessment is improved, the medical cost is greatly reduced, and the method is easy to widely popularize.

Description

technical field [0001] The invention belongs to the technical field of artificial intelligence detection, and in particular relates to a method and system for predicting the risk of lymph node metastasis of early non-small cell lung cancer based on a convolutional neural network. Background technique [0002] Primary lung cancer is currently the malignant tumor with the highest morbidity and mortality in the world, and primary non-small cell lung cancer (NSCLC) is the main pathological type of primary lung cancer. The detection rate of NSCLC is increasing. Precise treatment for early stage NSCLC has important clinical significance. [0003] To achieve precise treatment of early-stage lung cancer, it is necessary to achieve the goal of radical cure on the basis of the least operation time and minimal surgical trauma. However, although the operation time of simple lumpectomy is short, some patients develop regional lymph node metastasis 1 year after surgery. Mass excision a...

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): G16H50/30G16H50/20G06T7/90G06N3/08G06N3/04A61B5/00
CPCG16H50/30G16H50/20A61B5/7275G06T7/90G06N3/04G06N3/08
Inventor 石超邓立彬邱峰娄伟明高建莹张黎
Owner 南昌市高新区人民医院
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap