Lung Nodule Detection from low dose CT scan using Optimization on Intel Xeon and Core processors with Intel Distribution of OpenVINO Toolkit

With the advancement of AI in the field of medical imaging, medical diagnosis is getting faster and viable for medical practitioners especially for cancer diagnosis. Earlier Deep Learning solutions had to be deployed on High Performance Computing devices like GPU for achieving real time performance. But with Optimization on Intel Core and Xeon processors with Intel Distribution of OpenVINO Toolkit (Open Visual Inference and Neural Network Optimization), it is possible to deploy Deep Learning models with accelerated performance, than running Tensorflow / Caffe models on CPU machines. In this paper we describe the proposed work wherein we ported our DetectNet Deep Learning Model with NVIDIA specific custom layer for lung nodule detection trained on LIDC dataset, using Intel Distribution of OpenVINO, and deployed the same in Intel Core/Xeon processors with accelerated performance.