Surgical instrument image intelligent segmentation method and system based on multi-scale feature fusion
A multi-scale feature and surgical instrument technology, applied in image analysis, image enhancement, image data processing, etc., can solve the problems of inability to automate, identify and track surgical instruments, achieve important clinical research significance and application value, and high accuracy , Improve the effect of diagnosis and treatment
Pending Publication Date: 2021-12-07
HEFEI UNIV OF TECH
View PDF0 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] Aiming at the deficiencies of the existing technology, the present invention provides a method, system, storage medium and electronic equipment for intelligent segmentation of surgical instrument images based on multi-scale feature fusion, which solves the problem that the existing technology cannot automatically and precisely identify surgical instruments Technical Issues with Tracking
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
[0054] First aspect, such as figure 1 As shown, the embodiment of the present invention provides a method for intelligent segmentation of surgical instrument images based on multi-scale feature fusion. Such as figure 2 As shown, the method first constructs a lightweight network architecture for multi-scale feature fusion, the lightweight network architecture includes a pre-trained encoder and decoder, and the encoder includes a parallel first convolutional neural sub-network and a second convolutional neural sub-network; the method comprising:
[0055] S1. Preprocessing the image of the surgical instrument to acquire a high-resolution image and a low-resolution image, where the resolution of the high-resolution image is twice that of the low-resolution image;
[0056] S2. Input the high-resolution image into the first convolutional neural sub-network, and obtain the large-scale image feature information output by the last convolutional layer in each hidden layer; input the ...
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 provides a surgical instrument image intelligent segmentation method and system based on multi-scale feature fusion, and relates to the field of surgical instrument image segmentation. The method comprises the following steps: firstly, constructing a lightweight network architecture with multi-scale feature fusion, wherein the lightweight network architecture comprises a pre-trained encoder and a pre-trained decoder; inputting a high-resolution image obtained through preprocessing into a first convolutional neural sub-network of the encoder, inputting a low-resolution image into a second convolutional neural sub-network of the encoder, and respectively obtaining large-scale image feature information and small-scale image feature information; fusing the large-scale image feature information and the small-scale image feature information with the same dimension by adopting a cascade mode; inputting the final image feature information into the decoder, and connecting each piece of fused feature information to each decoding unit in a layer-skipping manner to execute up-sampling operation, thereby obtaining an intelligent segmentation result of a surgical instrument image. Compared with a traditional deep learning method, the obtained surgical instrument image segmentation result has high accuracy, and the reasoning time of models is shortened.
Description
technical field [0001] The invention relates to the technical field of surgical instrument image segmentation, in particular to a method, system, storage medium and electronic equipment for intelligent segmentation of surgical instrument images based on multi-scale feature fusion. Background technique [0002] With the continuous development of science and technology, the proportion of minimally invasive surgery in hospitals at all levels continues to increase. Compared with traditional surgical operations, minimally invasive surgery has limited field of view and narrow cavity space, which increases the difficulty of minimally invasive surgery and lengthens the learning curve for surgeons. At the same time, compared with doctors in tertiary hospitals, doctors in primary hospitals have limited minimally invasive diagnosis and treatment capabilities, and their minimally invasive surgical skills are not high, which increases the risk of postoperative complications and secondary...
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 IPC IPC(8): G06T7/10G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06T7/10G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/20021G06N3/045G06F18/253
Inventor 王浩丁帅汪家欣杨善林
Owner HEFEI UNIV OF TECH