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

Lesion detection method, apparatus and device

A detection method and CT image technology, applied in image data processing, instrumentation, calculation, etc., can solve problems such as high system performance and time consumption, and low efficiency of lesion detection methods

Inactive Publication Date: 2019-01-08
NEUSOFT MEDICAL SYST CO LTD
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Since the existing processing method requires two convolution calculations, which contain repeated calculation content, and the convolution calculation itself is a processing process that consumes a lot of system performance and time, the current lesion detection method is inefficient.

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
  • Lesion detection method, apparatus and device
  • Lesion detection method, apparatus and device
  • Lesion detection method, apparatus and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] see figure 1 , which is a flow chart of a lesion detection method provided in an embodiment of the present application, the method comprising:

[0050] S101: Obtain a computed tomography CT image.

[0051] In the embodiment of the present application, before lesion detection, first obtain a CT image as a lesion detection object; wherein, the CT image can be a CT image of the patient's organ to be detected, such as a lung CT image, a thyroid CT image, a brain CT image, etc. Images, etc. Specifically, the embodiment of the present application can detect pulmonary nodules in lung CT images, detect thyroid nodules in thyroid CT images, and detect cerebral ischemia and necrosis in brain CT images. etc., and no specific lesion detection object is limited here.

[0052] In practical applications, the CT image used as the lesion detection object may be a CT image acquired in advance and stored in a database, or may be a CT image acquired after performing a CT scan on a patien...

Embodiment 2

[0071] For the location of the lesion center point, lesion range, and lesion type information detected after lesion detection, doctors need to further diagnose the patient based on the above test results. In order to facilitate doctors to understand the patient's test results more intuitively, and To determine the follow-up treatment more accurately, refer to image 3 , the embodiment of the present application further provides the following steps on the basis of the above method embodiment 1 S101-S105:

[0072] S301: Construct a three-dimensional model of the CT image according to the lesion center position, lesion range, and lesion type information of the CT image.

[0073] After detecting the lesion center point position, lesion range and lesion type information of the CT image through the first embodiment of the above method, a three-dimensional model of the CT image is constructed based on the above detection results.

[0074] Taking the construction of the 3D model of 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

A lesion detection method, apparatus and device are disclosed. The method comprises: a CT image is processed by a self-encode convolutional neural network model, feature extraction is conducted on theCT image by a convolution layer and a pooling layer of the encoder network, a lesion candidate region of the CT image is obtained, the lesion candidate region of CT image is processed by the upper sampling layer and convolution layer of the decoder network, and the lesion center position of the CT image is obtained; the CT image which has determined the lesion center position is processed by theprobability output layer, and the lesion range and the lesion type information of the CT image are obtained. Since the calculation of the decoder network is based on the convolution processing resultof the encoder network, and the application does not need to carry out repeated convolution calculation processing, thus saving the convolution calculation processing time and improving the lesion detection efficiency to a certain extent on the premise of ensuring the lesion detection accuracy.

Description

technical field [0001] The present application relates to the field of medical image processing, in particular to a lesion detection method, device and equipment. Background technique [0002] Cancer is currently the most threatening disease to human health and life in society. Its commonly used diagnostic technology is computerized tomography (CT) technology, which is usually used to diagnose whether various organs have lesions and determine the type of lesions ( That is, benign and malignant), it is helpful for the early diagnosis and treatment of various cancers, and thus can reduce the mortality of patients. [0003] For example, lung cancer is one of the fastest-growing cancers in morbidity and mortality. Lung cancer generally evolves from pulmonary nodules. Doctors can help early diagnosis and treatment of lung cancer by analyzing the benign and malignant of lung nodules in patients. [0004] In actual clinical application, the full convolutional neural network model ...

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): G06T7/00
CPCG06T7/0012G06T2207/10081G06T2207/20081G06T2207/20084G06T2207/30096
Inventor 陈俊强陈磊刘允康雁
Owner NEUSOFT MEDICAL SYST CO LTD
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