Overview / Usage

At Asociación de Investigacion en Inteligencia Artificial Para la Leucemia Peter Moss we work on open-source technologies to enhance the healthcare/medical industry. In particular, our focus is researching into technologies that could be used for early detection of Leukemia, including Acute Myeloid & Acute Lymphoblastic Leukemia.

The NVIDIA® Jetson Nano™ is a small, powerful computer that is capable of running multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing. Our goal with this project is to utilize the power of Intel® oneAPI AI Analytics Toolkit on an Intel® NUC Kit NUC7i7BNH to accelerate the training process, and TensorRT for high performance inference on the NVIDIA® Jetson Nano™.

Methodology / Approach

• The Intel® oneAPI AI Analytics Toolkit and Intel® Optimization for Tensorflow* are installed on the Intel® NUC Kit NUC7i7BNH to optimize and accelerate the CNN training process.
• The trained model is converted to TFRT, ONNX & TensorRT and deployed to the NVIDIA® Jetson Nano™ for high performance inference on the Jetson's 128-core Maxwell GPU.

Technologies Used

• Intel® oneAPI AI Analytics Toolkit
• Intel® Optimization for Tensorflow*
• Intel® NUC Kit NUC7i7BNH
• TFRT
• ONNX
• TensorRT
• NVIDIA® Jetson Nano™
• 128-core Maxwell GPU

Repository

https://github.com/AMLResearchProject/ALL-Jetson-Nano-Classifier

Other links

• Docker Hub