Laparoscopic surgery stage identification method and system based on dual-granularity time convolution

A recognition method and double-grained technology, applied in neural learning methods, character and pattern recognition, image data processing, etc., can solve problems such as difficult to accurately distinguish transition frames between stages, and achieve good recognition effect, improved recognition effect, and good generalization effect of ability

Pending Publication Date: 2022-04-19
SOUTH CHINA UNIV OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Using the visual and timing information of surgical videos, it can solve the problem that in the field of deep learning, the

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
  • Laparoscopic surgery stage identification method and system based on dual-granularity time convolution
  • Laparoscopic surgery stage identification method and system based on dual-granularity time convolution
  • Laparoscopic surgery stage identification method and system based on dual-granularity time convolution

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0084] Example 1

[0085] as Figures 1 through 4 As shown, the present embodiment discloses a method based on two-particle size time convolution laparoscopic surgical stage identification method, the specifics of which are as follows:

[0086] 1) First, the laparoscopic minimally invasive surgical process is recorded through a miniature camera mounted at the tip of the surgical instrument, and each complete surgical process is stored as a video. Each video is then sliced using ffpmeg, saving a picture every 5 frames, in frame number order. Abnormal images, including those with full-scale blur, large-scale phantoms, extreme lighting, and incomplete shooting, were then eliminated and made into datasets for laparoscopic surgery and split into training, validation, and testing sets at a 40:8:32 ratio. Finally, OpenCV was used to perform image enhancement operations such as center flipping, random cropping, and scrambling order of laparoscopic surgery pictures.

[0087] 2) The processe...

Example Embodiment

[0118] Example 2

[0119] See Figure 5 As shown, the present embodiment discloses a laparoscopic surgical stage recognition system based on two-particle size time convolution, comprising the following functional modules:

[0120] Data acquisition module for collecting laparoscopic surgical videos, downsampling each video, retaining several images at each stage of each video, making a laparoscopic surgical dataset, and arranging them in the format of "address / video serial number / frame sequence number" to form a video sequence;

[0121] Data processing module, which is used to input the video sequence in the laparoscopic surgical dataset into the first part of the two-particle time convolutional network, that is, the double-particle time convolution module, model the long-distance time context information, generate the initial prediction result, and use the cross-entropy loss function to calculate the degree of difference between the initial prediction result and the actual data; th...

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 discloses a laparoscopic surgery stage identification method and system based on dual-granularity time convolution. The method comprises the following steps: 1) constructing a laparoscopic surgery data set; 2) performing preliminary feature extraction on the picture sequence by using a dual-granularity time convolution module of the dual-granularity time convolution network, and outputting an initial prediction result of each frame of image; 3) correcting an initial prediction result output by the dual-granularity time convolution module by using a single-granularity time convolution module of the dual-granularity time convolution network; and 4) mapping a prediction result to an interval of (0, 1) to obtain a final operation stage identification result. The laparoscopic surgery stage recognition is realized by using the dual-granularity time convolution network, higher precision and better generalization ability under different backgrounds are realized, different types of surgery stages can be accurately detected, and the visual and time sequence information of the surgery video is utilized to realize the recognition of the laparoscopic surgery stage. The problem that in the deep learning field, operation stage categories can be recognized, but stage transition frames are difficult to distinguish accurately can be solved.

Description

technical field [0001] The present invention relates to the technical field of laparoscopic minimally invasive surgery image processing and neural network, in particular to a laparoscopic surgery phase recognition method and system based on double granularity time convolution. Background technique [0002] Laparoscopic minimally invasive surgery is a common minimally invasive surgical procedure that not only provides substantial medical benefits to the patient, but also provides the opportunity for the physician to record video of the procedure due to the need for a camera during the procedure. Reviewing and analyzing the surgical process through surgical video can improve the technical quality of surgeons and improve the safety of patients. However, manual retrieval of surgical videos is a very tedious and time-consuming task. With the increasing maturity of computer-aided technology, automated surgical stage recognition methods can better help doctors monitor and optimize ...

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): G06T7/00G06N3/08G06N3/04G06K9/62G06V10/44G06V10/764
CPCG06T7/0012G06N3/08G06T2207/20081G06T2207/10016G06N3/045G06F18/2415
Inventor 吴秋遐韦喆艺
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap