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

Diagnosis model constructed based on artificial intelligence fusion multi-modal information and used for various pathology types of benign and malignant pulmonary nodules

A technology of artificial intelligence and diagnostic models, applied in the field of medical image processing, can solve problems such as low work efficiency, high dependence on personal clinical experience, and inability to classify and diagnose pulmonary nodules pathological types, and achieve the effect of improving diagnostic efficiency

Active Publication Date: 2020-08-25
SHANDONG UNIV +1
View PDF9 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide an artificial intelligence-based diagnostic method for the above-mentioned existing benign and malignant pulmonary nodules with low practical work efficiency, high dependence on personal clinical experience, and inability to classify and diagnose specific pathological types of pulmonary nodules at the same time. Constructing a diagnostic model of benign and malignant pulmonary nodules with multiple pathological types by fusing multimodal information

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
  • Diagnosis model constructed based on artificial intelligence fusion multi-modal information and used for various pathology types of benign and malignant pulmonary nodules
  • Diagnosis model constructed based on artificial intelligence fusion multi-modal information and used for various pathology types of benign and malignant pulmonary nodules
  • Diagnosis model constructed based on artificial intelligence fusion multi-modal information and used for various pathology types of benign and malignant pulmonary nodules

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be described in further detail below in conjunction with the accompanying drawings and specific implementation. The present invention realizes the construction of multiple pathological types of benign and malignant pulmonary nodules diagnostic models based on artificial intelligence fusion of multiple modal information mainly including the following steps:

[0029] S1 builds a multi-resolution 3D multi-category deep learning network model.

[0030] S1.1 Collect CT image data with clear pathological types, and the slice thickness is 1mm.

[0031] Collect CT image data of four pathological types, including inflammation, squamous cell carcinoma, adenocarcinoma, and benign others. It is required to have a clear pathological type, and the layer thickness is 1mm, and the diameter D of the nodule is 3mm≤D≤30mm. It is divided into training set, verification set and test set according to the proportion of categories.

[0032] S1.2 Extract and store t...

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 relates to a diagnosis model constructed based on artificial intelligence fusion multi-modal information and used for various pathology types of benign and malignant pulmonary nodules. The method comprises the steps: constructing a multi-resolution 3D multi-classification deep learning network model; constructing a machine learning multi-classification model; training the constructedmulti-resolution 3D deep learning model by using CT image, and obtaining a weight; training the constructed machine learning multi-classification model by using lung tumor marker information, and obtaining a weight; and fusing the lung CT imaging information and the lung tumor marker information at the tail end of the model by migrating the weights of the deep learning network and the machine learning network model through migration learning. According to the invention, a deep learning network model and a machine learning model are adopted to respectively mine deep features related to pathological type classification in a lung CT image and a lung tumor marker, and the fusion of the CT image and the lung tumor marker multi-modal information is realized by fusing the two network models to rapidly diagnose the specific pathology type of the pulmonary nodule.

Description

technical field [0001] The invention belongs to the field of medical image processing, and in particular relates to constructing diagnostic models of various pathological types of benign and malignant pulmonary nodules based on fusion of lung CT images and tumor marker multimodal information. Background technique [0002] Early detection and treatment of pulmonary nodules in patients with lung diseases is a key measure to reduce lung cancer mortality. The traditional diagnosis of benign and malignant pulmonary nodules is mostly based on the combination of lung CT imaging examination and lung tumor marker examination by clinical medical workers to jointly diagnose benign and malignant pulmonary nodules. This needs to rely on the personal clinical experience of medical workers, increasing their work pressure, and because the doctor's personal experience level is uneven, the traditional diagnosis method is easily affected by the doctor's personal experience level, and the diagn...

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): G16H50/20G06T7/00
CPCG16H50/20G06T7/0012G06T2207/10081G06T2207/20081G06T2207/20084G06T2207/30064G06T2207/30096
Inventor 董恩清金叶纪惠中倪天骄薛鹏傅宇曹海崔文韬
Owner SHANDONG UNIV
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