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

Intracranial hemorrhage automatic sketching method and device based on head CT (Computed Tomography) plain-scan image

A technology for intracranial hemorrhage and head, which is applied in the field of AI medical image-aided diagnosis, can solve the problems such as subarachnoid hemorrhage segmentation is difficult to reflect, bleeding segmentation performance is not ideal, and complete segmentation is difficult, so as to ensure the feasibility of optimization and reduce diagnosis The time required to improve the effect of accurate diagnosis

Active Publication Date: 2022-05-13
深圳市铱硙医疗科技有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The performance of hemorrhage segmentation other than brain parenchyma is not ideal, and the segmentation of subarachnoid hemorrhage is also difficult to reflect
The main reason is limited data, or the feature extraction of subdural hemorrhage close to the skull is different from that of general cerebral parenchymal hemorrhage, and it is usually difficult to complete segmentation by methods

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
  • Intracranial hemorrhage automatic sketching method and device based on head CT (Computed Tomography) plain-scan image
  • Intracranial hemorrhage automatic sketching method and device based on head CT (Computed Tomography) plain-scan image
  • Intracranial hemorrhage automatic sketching method and device based on head CT (Computed Tomography) plain-scan image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0100] The present invention provides a method for automatic delineation of intracranial hemorrhage based on head CT plain scan images, such as figure 1 shown, including:

[0101] Step 1: Process the original image of the head, read the processed CT value, adjust the required window width and level, preprocess the CT value within the window width and level, and convert it into a single-channel matrix;

[0102] Step 2: Input the processed original image and preprocessing results into the preset sketching model;

[0103] Step 3: Predict the sketched data based on the preset sketching model, and output the predicted value;

[0104] Step 4: Based on the prediction value, obtain a lesion delineation image of the output prediction image.

[0105] In this example, when the doctor diagnoses intracranial hemorrhage, the main window width and window level used can be the brain window [40,80], and the subdural window level will be used when diagnosing subdural hemorrhage, which can be ...

Embodiment 2

[0111] On the basis of Embodiment 1, step 1: process the original image of the head, read the processed CT value, and adjust the required window width and level, and preprocess the CT value within the window width and window level, Convert to a single-channel matrix, including:

[0112] Read the CT value of the original image of the head, filter the value that meets the first condition in the CT value and keep it, and remove the remaining value to obtain the coordinate information of the skull;

[0113] Set the required window width and window level, and then extract the first value from the CT value of the original head image;

[0114] The extracted first value is normalized, and the coordinate information of the skull is masked in the matrix corresponding to the set window width and level, and converted to 0 to construct a single-channel matrix.

[0115] In this embodiment, normalization is performed on the extracted value to convert the value between 0 and 1, and the previ...

Embodiment 3

[0121] On the basis of Embodiment 1, in step 2, before inputting into the preset sketching model, it also includes:

[0122] Construct the basic intracranial hemorrhage segmentation model and determine the segmentation network evaluation index set;

[0123] Based on the evaluation index set, evaluate the basic intracranial hemorrhage segmentation model;

[0124] According to the evaluation results, determine the basic level of the basic intracranial hemorrhage segmentation model;

[0125] When the basic level is the normal level, the basic intracranial hemorrhage segmentation model is optimized to obtain a preset sketching model;

[0126] When the basic level is a special level, the basic intracranial hemorrhage segmentation model is saved, and the saved model is the preset sketching model.

[0127] In this embodiment, the construction of the basic intracranial hemorrhage segmentation model, such as image 3 As shown, it can be achieved as follows:

[0128] The convolution...

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 provides an intracranial hemorrhage automatic sketching method and device based on a head CT plain-scan image, and the method comprises the steps: 1, processing an original head image, reading a processed CT value, adjusting a needed window width and window level, carrying out the preprocessing of the CT value in the window width and window level, and converting the CT value into a single-channel matrix; 2, inputting the processed original image and the preprocessing result into a preset sketching model; 3, predicting the sketching data based on a preset sketching model, and outputting a predicted value; and 4, based on the predicted value, obtaining a focus sketching image of the output predicted image. Intracranial hemorrhage is accurately segmented and sketched, the time required by diagnosis is shortened, and the detection efficiency and the stability of accurate diagnosis are improved.

Description

technical field [0001] The present invention relates to the technical field of AI medical image-aided diagnosis, in particular to an automatic delineation method and device for intracranial hemorrhage based on head CT plain scan images. Background technique [0002] According to statistics, since 2017, stroke has become the main cause of death in my country, and its incidence rate ranks first in the world. Among the various types of stroke, cerebral hemorrhage and non-traumatic parenchymal hemorrhage are the most fatal stroke types. The diagnosis of cerebral hemorrhage usually uses plain CT, and doctors diagnose the cause of hemorrhage based on the location of hemorrhage and the volume of intracranial hemorrhage on CT images. Intracranial hemorrhage often includes multiple subcategories in addition to parenchymal hemorrhage. These include, subdural hemorrhage, and epidural hemorrhage, intraventricular hemorrhage, subarachnoid hemorrhage mainly caused by ruptured aneurysm. ...

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): G06T11/00G06T7/149G06N3/04
CPCG06T11/00G06T7/149G06T2207/10081G06T2207/30016G06N3/045
Inventor 王思伦刘涵川
Owner 深圳市铱硙医疗科技有限公司
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