A Deep Learning-Based Smartphone App for Real-Time Detection of Retinal Abnormalities in Fundus Images
This prject trained a keras modal and implent this model into an iOS app. There are 3 step to achive this:
- We need to train a keras modal and save it as an .h5 file. (In Python) 2.Conver this .h5 model to .mlmodel model using coremltools.(In Python) 3.Implent this .mlmodel in the iOS app.(Using Swift language, X-Code and OpenCV toolkit) Tips: iOS app development can only been done on Mac, windows system is not supported.
Citation Haoran Wei, Abhishek Sehgal and Nasser Kehtarnavaz . A Deep Learning-Based Smartphone App for Real-Time Detection of Retinal Abnormalities in Fundus Images. SPIE conference of Real-Time Image Processing and Deep Learning, Baltimore, 2019.
一个基于keras深度学习模型的iOS APP，用于眼底糖尿病检测 项目需要如下三步操作： 1.训练一个keras模型，并保存为.h5模型（python实现）； 2.将.h5模型转换为.mlmodel模型（python中实现）； 3.将.mlmodel模型用在iOS app（用swift实现，在x-code中编写代码，使用Mac机器）。注意：iSO开发只能用Mac，不能用window系统。
详情及引用请参考： Haoran Wei, Abhishek Sehgal and Nasser Kehtarnavaz . A Deep Learning-Based Smartphone App for Real-Time Detection of Retinal Abnormalities in Fundus Images. SPIE conference of Real-Time Image Processing and Deep Learning, Baltimore, 2019.