Conferences and supporting programme
Embedded Deep Learning Healthcare Collaboration System
Deep Learning has been proven to be very successful in image classification and detection tasks, also for medical diagnosis. Deploying deep learning networks from the training environment to embedded platforms for inference might be a complex task. There are a number of deep learning frameworks widely used in the industry, such as Caffe, TensorFlow etc. Typically the training of the deep learning networks is performed in data centers while the inference might take place on embedded platforms, optimized for performance and power consumption. Such platforms are typically limited both from software and hardware perspective, so usually it is not recommended to use original training framework for inference. Additional complications of the deployment process include supporting various layer types and networks that are getting more and more complex. Obviously, ensuring the accuracy of the transforms networks is not trivial. Thanks to Intel OpenVINO toolkit, it’s easier to deploy optimized Deep Learning system on Intel platforms. OpenVINO is a toolkit to fast-track development of high performance computer vision and deep learning into vision applications. It enables deep learning on hardware accelerators and easy heterogeneous execution across Intel platforms to meet various performance, power and cost requirement. This paper shows an Deep Learning medical inference system using Intel OpenVINO toolkit. Deep learning inference can be run optimally on Intel platforms from edge to cloud.
--- Date: 26.02.2019 Time: 11:30 - 12:00 Location: Conference Counter NCC Ost