Surgical skill rating method and system based on interpretable artificial intelligence

A surgical and artificial intelligence technology, applied in the field of artificial intelligence, to achieve the effect of intuitive rating results, reducing the amount of parameters, and improving the speed

Pending Publication Date: 2022-03-11
翁莹
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] This old-fashioned traditional training method is obviously not enough to meet the high-efficiency and high-precision requirements of modern medicine. Therefore, a more scientific and autonomous method is urgently needed to help novice doctors find irregular operations in their own operations, so as to make targeted improvements

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
  • Surgical skill rating method and system based on interpretable artificial intelligence
  • Surgical skill rating method and system based on interpretable artificial intelligence
  • Surgical skill rating method and system based on interpretable artificial intelligence

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] In order to help doctors understand their own defects in surgical operations and display them in a scientific and intuitive way, such as figure 1 Shown, the present invention proposes a kind of surgical operation skill rating method based on explainable artificial intelligence, comprises steps:

[0038] S1: Obtain the rectified video of the operation, and extract the feature atlas under each channel (a total of N channels) of the rectified video through the expanded 3D convolutional network;

[0039] S2: Through the global average pooling layer, each feature map is averaged in the time dimension and the time series average feature is obtained;

[0040] S3: Calculate the classification score of each feature region in the rectified video according to the time series mean feature;

[0041] S4: Obtain a class activation map by associating the classification score with the corresponding feature region in the rectified video, and obtain a saliency map with the surgical traje...

Embodiment 2

[0057] In order to have a better overall understanding of the technical content of the present invention, this embodiment uses the system structure method to carry out a modular description of the functional division of the technical solution of the present application, such as figure 2 As shown, a surgical skill rating system based on explainable artificial intelligence, including:

[0058] The feature extraction module is used to extract the feature atlas under each channel of the rectified video through the expanded three-dimensional convolutional network according to the rectified video of the operation;

[0059] The mean value processing module is used to perform mean value processing on the time dimension of each feature map through the global average pooling layer and obtain the time series mean value feature;

[0060] A classification and scoring module, used to obtain the classification and scoring of each feature region in the rectified video according to the time s...

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 surgical operation skill rating method based on interpretable artificial intelligence, which relates to the field of artificial intelligence, and mainly comprises the following steps: acquiring a rectification video of an operation, and extracting a feature image set under each channel of the rectification video through an expansion three-dimensional convolution network; performing equalization processing on each feature map in a time dimension through a global average pooling layer, and obtaining a time sequence average feature; the classification score of each feature region in the rectified video is solved according to the time sequence mean value feature; and associating the classification score with the corresponding feature region in the rectified video to obtain a category activation map, and obtaining a saliency map with the operation trajectory and the rating of each trajectory segment according to the category activation map. According to the method, the popularity saliency map with the operation track and the corresponding score is obtained, so that a doctor can clearly know defects and deficiencies existing in the operation of the doctor in the operation, and targeted training is performed.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a surgical skill rating method and system based on interpretable artificial intelligence. Background technique [0002] At present, the skill training for doctors' surgical operations is still in the old-fashioned model, relying on the surgical experience of senior doctors to guide and correct the intraoperative operations of novice doctors. However, due to the influence of personal subjective cognition, different doctors have deviations in their understanding of intraoperative operations, and some of their understandings may deviate from the correct operation, which leads to non-standard guidance for novice doctors. Novice doctors still have to find their own mistakes and correct them in actual operations again and again. [0003] This old-fashioned traditional training method is obviously not enough to meet the high-efficiency and high-precision require...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/46G06V20/40G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08G06Q10/06G16H30/20
CPCG06Q10/06393G06N3/08G16H30/20G06N3/045G06F18/2414
Inventor 翁莹陈珂张一名
Owner 翁莹
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