Quantitative analysis method of substantia nigra hyperechoic intensity by transcranial ultrasound

An analysis method and hyperechoic technology, which is applied in the field of quantitative analysis of substantia nigra hyperecho intensity, can solve problems such as identification and segmentation of staying images, quantification of non-stantia nigra hyperecho intensity, etc., so as to improve the level of diagnosis and treatment and high diagnostic sensitivity Effect

CN113538380BActive Publication Date: 2022-04-22TONGJI HOSPITAL ATTACHED TO TONGJI MEDICAL COLLEGE HUAZHONG SCI TECH
4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2022-04-22

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a quantitative analysis method of substantia nigra hyperecho intensity by transcranial ultrasound, which comprises the following steps: 1) obtaining the original TCS picture of the patient and the corresponding labeled picture after interpretation by the clinician, and randomly dividing them into a training set and a training set. Test set; 2) Deeply learn the adversarial network on the training set and perform network training to establish a model for analyzing substantia nigra echo images; 3) use the test set to verify the model for analyzing substantia nigra echo images; The predicted image is first converted into a grayscale image, and then converted into a binary image; 5) Morphological analysis of the grayscale value is performed to extract the images of the left and right substantia nigra hyperechoic areas in the midbrain; 6) Extract feature parameters; 7) Calculate Obtain the substantia nigra hyperechoic total intensity value. The quantitative analysis method of substantia nigra hyperecho intensity of the transcranial ultrasound of the present invention can realize quantitative interpretation and analysis of substantia nigra hyperecho, makes the change of substantia nigra echo more intuitive, and assists clinicians to improve the ability of identifying and diagnosing PD.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of quantitative analysis of hyperechoic intensity of substantia nigra, in particular to a method for quantitative analysis of hyperechoic intensity of substantia nigra by transcranial ultrasound. Background technique

[0002] In recent years, with the scientific development of computer technology and artificial intelligence, there have been researches on the application of computer-aided diagnosis (CAD) systems in medical imaging. The emergence of this technology has greatly improved the efficiency and accuracy of clinicians' interpretation of images, avoided the influence of inexperience and short time on the accuracy of interpretation results, and achieved better and more accurate transcranial ultrasound. Quality echo image analysis, faster identification and quantification of abnormal substantia nigra echo.

[0003] Transcranial midbrain sonography (TCS) is a convenient, repeatable, non-invasive examinat...

Examples

Embodiment Construction

[0039] The implementation of the present invention will be described in detail below in conjunction with the examples of implementation, but they do not constitute a limitation of the present invention, and are only examples. At the same time, the advantages of the present invention will become clearer and easier to understand.

[0040] A method for quantitative analysis of substantia nigra hyperechoic intensity of transcranial ultrasound of the present invention comprises the following steps:

[0041] 1. Selection of image specimens: A total of 600 patients who had undergone TCS in clinical practice were selected. Then extract their original TCS images and corresponding annotated images interpreted by clinicians. In order to ensure the universality of the training effect, the original pictures of 400 patients and the corresponding labeled pictures interpreted by clinicians were randomly selected as the training set; the pictures of the remaining 200 patients were used as the...