Medical laboratory report image classification method and apparatus

Abstract

The present invention provides a medical laboratory report image classification method and apparatus. The medical laboratory report image classification method comprises: calculating an image feature of a given medical laboratory report image; and determining a type and a format of a medical laboratory report corresponding to the given medical laboratory report image based on the calculated image feature by using a trained classification model. According to the medical laboratory report image classification method and apparatus provided by the present invention, a type and a format of a medical laboratory report are automatically determined according to an image feature, thereby saving the process of manually identifying a type and a format of a medical laboratory report, and improving medical laboratory report identification efficiency.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image classification method and device for medical laboratory sheets. Background technique [0002] With the improvement of medical level, the development of the Internet and the rise of big data technology, medical big data will soon become a hot spot in the industry in the near future and in the future. The electronicization of medical test sheets (or test reports) is one of the key technical means to realize medical big data. The electronicization of medical test sheets refers to the process of using imaging devices (such as scanners, cameras, and smartphones) to take pictures of paper-based medical test sheets and automatically identify their contents. For different test items (such as blood routine, urine routine, liver function, six plasma items or eight serum items, etc.), there may be obvious differences in the format and content of the medical labora...

AI Technical Summary

Problems solved by technology

For different test items (such as blood routine, urine routine, liver function, six plasma items or eight serum items, etc.), there may be obvious differences in the format and content of the medical laboratory test sheet, which brings great challenges to the electronicization of fully automatic medical laboratory test sheets. great difficulties and challenges

[0003] At present, most of the existing electronic technologies and systems for medical test sheets are semi-automatic. They can recognize the content of medical test sheets, but cannot automatically judge their format. Therefore, users need to specify the type and format of medical test sheets in advance.

This electronic method of medical test sheets can guarantee the accuracy of identification, but cannot guarantee the speed of processing

When the number of medical test orders to be identified is huge, the processing efficiency of these systems becomes a serious bottleneck

Images