Computer-aided diagnosis system for medical images using deep convolutional neural networks

A convolutional neural network and medical image technology, which is applied in the field of computer-aided diagnosis system for medical images using deep convolutional neural network, can solve the problem of analyzing medical image length

Inactive Publication Date: 2017-06-27
图兮深维医疗科技(苏州)有限公司
View PDF4 Cites 71 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, medical imaging generates large amounts of data, and analyzing medical images requires a long process

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
  • Computer-aided diagnosis system for medical images using deep convolutional neural networks
  • Computer-aided diagnosis system for medical images using deep convolutional neural networks
  • Computer-aided diagnosis system for medical images using deep convolutional neural networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] Example 1 - Report Generation

[0078] Figure 4 Examples of analysis inclusions are shown. The location of the nodule identified by the system is in the right upper lung. Nodules were determined to be non-solid and ground-glass-like. The x-y-z coordinates of the centroid of the nodule are 195-280-43. The diameter of the nodule is 11 mm along the long axis and 7 mm along the short axis. The borders of nodules are irregular in shape. The probability of malignancy is 84%.

[0079] Figure 5 An example of a diagnostic report is shown. In this embodiment, the report includes a user interface. The user is presented with three projected views 501 , 502 and 503 of a lung CT scan. Views 501, 502 and 503 include cursors to allow the user to pinpoint a location. Window 504 shows the 3D model of the segmented lung and detected nodule candidates (square points). The 3D model is superimposed in the projected views 501 , 502 and 503 .

Embodiment 2

[0080] Example 2 - Validation of pulmonary nodule detection

[0081] The techniques disclosed herein were applied to the publicly available lung nodule database (LIDC-IDRI) provided by the National Institutes of Health (NIH). The database included 1018 CT lung scans from 1012 patients. Scans were captured using various sets of CT machines and various parameter settings. Each voxel within the scan has been carefully annotated as normal (non-nodular) or abnormal (nodular) by four radiologists. In the training step, five-fold cross-validation is used for evaluation. All scans are first divided into five sets, each set containing approximately the same number of scans. Each scan was randomly assigned to one of five sets. Each evaluation iteration uses four of the five sets to train the lung nodule detector, and the remaining set is used to evaluate the trained detector. During training iterations, split and cascade detections are applied to the database. The evaluation was r...

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

Described are systems, media, and methods for applying deep convolutional neural networks to medical images to generate a real-time or near real-time diagnosis.

Description

Background technique [0001] Disease diagnosis is an important step in health checkup. Medical imaging is a useful tool for diagnosing many diseases and providing a non-invasive diagnosis, which has advantages over other tools. However, medical imaging generates a large amount of data, and analyzing medical images requires a long process. In the early stages of disease diagnosis, abnormal tissue may not be prominent even in high-resolution imaging modalities. Therefore, new technology to solve the problem is necessary. Contents of the invention [0002] A computer-aided diagnosis (CAD) system for medical images aims to help doctors diagnose diseases more efficiently by reducing examination time, increasing diagnostic accuracy, and reducing diagnostic variation due to experience and individual conditions. Using advanced computer technology, CAD systems highlight areas of potential medical condition for physicians to examine in detail and make a final diagnostic decision. T...

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): G06F19/00G06T7/00
CPCG06T7/0012G06T2207/20081G06T2207/20084G06T2207/30004G06T2207/30064G16H15/00G16H50/20G06T11/008G06T2207/10072G06T2207/10116G06T2207/20016A61B5/0022A61B5/055A61B5/08A61B5/4887A61B5/7267A61B5/743A61B5/7485A61B6/032A61B6/037A61B6/50A61B6/5211A61B6/5217A61B6/5235A61B6/563A61B8/5223A61B8/565A61B2576/00
Inventor 高大山钟新
Owner 图兮深维医疗科技(苏州)有限公司
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