安卓平台上的医用图像信息采集系统的开发设计

为了探讨在安卓平台上构建医用图像采集系统的开发个案,分析通过以智能手机、平板电脑为核心安卓设备通过拍照获得化验单数据后进行文本识别并提交智慧医疗系统的解决方案。本文首先通过二值化算法形成低阈值图像数据,使用卷积神经元网络算法对文本进行逐一识别,使用K-means算法对识别后的单字文本进行字段记录值的整合并形成元数据库服务于其他智慧医疗系统模块。在使用9000组数据对神经元网络进行前期训练的前提下,该系统的识别准确率达到了99.5%以上。本系统具有一定的可行性,对未来智慧医疗的系统开发有实践意义。

Development and Design of Medical Image Information Acquisition System on Android Platform

This paper focuses on the case study of building a medical image acquisition system on Android platform,and analyzes the solution of text recognition and intelligent medical system submission after obtaining test sheet data by taking photos on Android devices with smart phones and tablets as the core.Firstly,the binary algorithm is used to form low threshold image data,the convolution neural network algorithm is used to recognize the text one by one,and the k-means algorithm is used to integrate the field record values of the recognized single word text and form the Metadatabase to serve other intelligent medical system modules.On the premise of using 9000 sets of data to train the neural network,the recognition accuracy of the system is over 99.5%.This system has certain feasibility and practical significance for the future development of intelligent medical system.