A deep learning-based type-2 diabetes renal lesion image classification method
A type 2 diabetes and kidney disease technology, applied in the field of deep learning, can solve problems such as limited, different, difficult feature extraction and classification work, and achieve the effect of improving recognition efficiency, reducing work pressure and burden, and reducing the consumption of human resources
Pending Publication Date: 2021-06-11
THE FIRST AFFILIATED HOSPITAL OF GUANGZHOU UNIV OF CHINESE MEDICINE
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The current classification methods based on artificially extracted features mainly have the following limitations: 1) Image quality
The quality of the collected kidney images is easily affected by many other uncontrollable factors such as lighting, lens, machine equipment and experience of image collection personnel; 2) Doctor's personal experience
Doctors usually assess and determine the extent of lesions by visually inspecting images, but the features recognized by human eyes in lesion images are limited, and due to different clinical experiences of doctors, different doctors may give different clinical diagnoses for the same image Results; 3) The characteristics of the pathological image itself
The difference between different stages of the image is very small, which brings certain difficulties to feature extraction and classification
Therefore, the classification of kidney images is a very challenging task
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0033] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, the specific implementation manners of the present invention will now be described in detail with reference to the accompanying drawings.
[0034] Such as figure 1 As shown, the present invention provides a type 2 diabetic nephropathy image classification method based on deep learning, including:
[0035] Extract images of renal lesions in diabetic patients marked with renal lesions, and divide the training set and test set after image preprocessing;
[0036] On the basis of the AlexNet convolutional neural network model, a batch normalization layer is introduced before each convolutional layer and fully connected layer to obtain a deep convolutional neural network BNnet, and the image of kidney lesions of diabetic patients as a training set is input to depth Carry out training in the convolutional neural network BNnet, adjust the network parameters and fu...
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
Login to View More Abstract
The invention relates to a deep learning-based type-2 diabetes renal lesion image classification method; aiming at the urgent demand of intelligent classification and statistics of diabetes renal lesion, the method comprises the following steps: acquiring a renal lesion image of a diabetic patient, and carrying out pre-processing such as marking and normalization on the renal lesion image of the diabetic patient; processing the renal lesion image of the diabetic patient by using a neural network method, automatically detecting the renal lesion category of the diabetic patient, and outputting a detection result. The invention can assist doctors in diagnosis of lesion recognition of the diabetic kidney, greatly improves the recognition efficiency, is particularly suitable for large-scale recognition, screening and statistics of the diabetic kidney lesion, helps to reduce the working pressure and burden of inspection personnel, reduces the consumption of human resources, and the invention can better serve areas with shortage of medical resources.
Description
technical field [0001] The present invention relates to the technical field of deep learning, and more specifically, relates to a method for classifying images of type 2 diabetic nephropathy based on deep learning. Background technique [0002] In recent years, artificial intelligence technology represented by deep learning has developed significantly, and its applications in the fields of pathology and medical imaging diagnosis have also attracted more and more attention. Deep learning is a technology to realize machine learning, and it is a method based on representation learning of data. Its technical core is an artificial neural network that simulates the human brain for analysis and learning, through different levels of feature analysis and extraction, and finally completes the data classification process. In the field of image recognition, deep convolutional neural network is a representative algorithm of deep learning, which is a network computing mode extended from ...
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
Login to View More Patent Type & AuthorityApplications(China)
IPC IPC(8): G06T7/00G06T5/00G06K9/62G16H30/40G06N3/04G06N3/08
CPCG06T7/0012G16H30/40G06N3/084G06T2207/20081G06T2207/20084G06T2207/30084G06T2207/30096G06N3/045G06F18/214G06F18/24G06T5/80G06T5/70
Inventor邱士军谭欣梁艺覃春红李伊凡
OwnerTHE FIRST AFFILIATED HOSPITAL OF GUANGZHOU UNIV OF CHINESE MEDICINE