Alimentary canal anatomical position identification method and device

Abstract

The invention discloses an alimentary canal anatomical position identification method and device. The method comprises the steps that a training set and a test set are acquired; a deep network model based on a deep learning framework is selected to act as the current deep network model, the current deep network model is trained by using the training set, the trained current deep network model is tested by using the test set and the identification accuracy data of the current deep network model are obtained, whether the identification accuracy data meet the preset accuracy requirements is judged, training is completed if the judgment result is yes, or the deep network model obtained by adjustment of the trained current deep network model is determined as the current deep network model, andthe process returns to the step of performing training on the current deep network model by using the training set; and the alimentary canal anatomical image acquired by the capsule endoscopy is inputted to the completely trained deep network model so that the corresponding alimentary canal anatomical position of the alimentary canal anatomical image outputted by the deep network model is obtained. Accurate locating of the capsule in the body can be realized.

Description

technical field [0001] The invention relates to the technical field of medical devices, and more specifically, to a method and device for identifying an anatomical position of the digestive tract. Background technique [0002] The capsule endoscope is used to take images or videos of the inner wall of the human digestive tract while moving in the human digestive tract, and wirelessly transmits to the external signal receiving device. Compared with the traditional medical endoscope, the capsule endoscope has the advantages of simple operation, It has the advantages of no trauma, no pain, no cross-infection, and does not affect the normal work of patients. It has high medical diagnostic value especially for the examination of small intestine diseases. [0003] A large number of pictures of the digestive tract collected by capsule endoscopy have a huge amount of data, and the automatic identification of the anatomical position of the digestive tract can play a positive role in ...

AI Technical Summary

Problems solved by technology

At present, most methods for identifying the anatomical position of the digestive tract rely on the image data sent back by the medical staff to roughly identify the location of the capsule, which is easily affected by the subjective factors of the medical staff and cannot accurately locate the capsule in the body.

Images