A hepatic echinococcosis focus recognition method and system based on deep learning

Abstract

The invention discloses a hepatic echinococcosis focus recognition method and system based on deep learning. The method comprises the steps that S1, an echinococcosis CT image set is acquired from anechinococcosis case; S2, training and verifying a liver segmentation model based on the common data set to perform liver segmentation and liver volume quantification; S3, obtaining a segmented liver region from the enveloped insect CT image set through a liver segmentation model, training and verifying a focus recognition model based on a liver segmentation result, and respectively endowing cysticechinococcosis lesions and vesicular echinococcosis lesions with different target tags in training and verification; And S4, obtaining a segmented liver region from the one-pack worm CT image throughthe liver segmentation model, and inputting the liver region as a VOI region into the focus recognition model on the flat scan CT to obtain a recognition result. According to the method and the system provided by the invention, feature information such as liver placeholder can be mined, flat scanning CT images of the echinococcosis can be marked manually, and various echinococcosis can be identified and classified by utilizing a convolutional neural network model.

Description

technical field [0001] The invention relates to the technical field of liver echinococcosis identification, in particular to a method and system for identifying liver echinococcosis lesions based on deep learning. Background technique [0002] Hydatid disease is a serious zoonotic parasitic disease that spreads across all continents of the world. The number of people and patients threatened by echinococcosis in my country ranks first in the world, and the infection rate of hermaphrodite echinococcosis in the hardest-hit Sanjiangyuan area of ​​Qinghai Province is 8.93-12.38%. The environment in this area is harsh, medical resources are scarce, and the level of doctors is not homogeneous. [0003] The current screening method for echinococcosis is relatively simple, mainly by portable B-ultrasound. This screening method has the following disadvantages: (1) It is highly dependent on the doctor's experience. If the doctor has insufficient experience or a low level, it is diffi...

AI Technical Summary

Problems solved by technology

[0003] The current screening method for echinococcosis is relatively simple, mainly by portable B-ultrasound

This screening method has the following disadvantages: (1) It is highly dependent on the doctor's experience. If the doctor has insufficient experience or a low level, it is difficult to guarantee the accuracy of the screening results, and different doctors may draw different conclusions

(2) The efficiency of this screening method is low due to factors such as environment and climate. Taking Qinghai Province as an example, due to the limitation of medical level, most of the screening work is done by doctors in relatively low-altitude areas. Most of the high-incidence areas are in areas with high altitude, hypoxia, harsh living environment, and economically backward areas, and screening personnel cannot carry out screening work for a long time

Images