Infant excrement sampling image classification processing system and method

Abstract

The invention discloses an infant excrement sampling image classification processing system which comprises an image acquisition subsystem used for inputting infant excrement images; and an image classification subsystem which is used for carrying out classification training on the infant excrement image input by the image acquisition subsystem to obtain a classification training result. In the sampling image classification processing system, firstly, an infant excrement image is obtained, related color, texture and other characteristic information are extracted from the early infant excrementimage; and the images are classified and tranined based on deep learning method to obtain a classification and training result. Compared with the prior art, the method has the advantages that the accuracy of image classification and recognition is greatly improved, the efficiency of disease screening is greatly improved, and a good auxiliary effect is achieved for screening work of early infant diseases.

Description

technical field [0001] The invention relates to a computer system capable of realizing data statistics and processing in the infant disease screening process, and in particular to a system and method for classifying and processing infant feces sampling images. Background technique [0002] In the prior art, some data statistics and processing methods are usually involved in the screening process of some infant diseases. Taking the screening of early infant biliary atresia disease as an example: biliary atresia is the most serious infant digestive tract disease, Clinically, jaundice is the first symptom. When jaundice appears, the feces also turn light yellow. As the jaundice worsens, it gradually becomes whiter or even clay-like grayish white. Therefore, the color of infant feces images has become an important basis for distinguishing the disease. Existing screening methods include: screening using a stool color card and screening based on the saturation (S) value in the HSV...

AI Technical Summary

Problems solved by technology

The disadvantages of using the stool color card are as follows: First, there are errors in the conclusions drawn by the parents after comparing the color of the color card. Abnormal biliary obstruction stools, this is missed in up to 30% of cases

Second, the awareness and vigilance of medical staff at all levels is not high for biliary atresia. At present, there are a large number of pediatric patients in mainland China, and doctors are scarce, so it is impossible to use limited manpower and material resources to achieve good results in the screening of biliary atresia and other diseases. grades

[0005] It can be seen that in the existing image data statistics and processing methods, there are deficiencies in using the saturation (S) value in the HSV model of stool images for screening: first, the saturation (S) value in the HSV model alone There will be certain errors in the screening, and the second is that the accuracy of identification and judgment is not high

Images