Health and Life Sciences AI Suite#
The Health and Life Sciences AI Suite is a collection of AI applications, libraries, and benchmarking tools to showcase the capabilities of Intel® Core™ Ultra processors within simulated medical situations.
The suite provides multi-modal AI pipelines accelerated with the OpenVINO™ toolkit, enabling high-performance deployment on Intel® Core™ Ultra platforms using CPU, integrated GPU (iGPU), and NPU.
[!Note] This software is for reference and evaluation purposes. The included applications are not intended for direct use in clinical or diagnostic environments and are not validated as such.
This suite offers reference applications demonstrating how multiple AI workloads can run simultaneously on a single platform and demonstrate how Intel hardware may be applied in the areas of: Multi-Modal Patient Monitoring, NICU Warmers, and Surgical Instruments (endoscopic video).
The apps provide simulated workflows that are consistent with those that would be required within a real-life scenario, in a hospital or a general health care facility settings.
For full setup, usage, and contribution guide, see:
Health and Life Sciences AI Suite Samples#
Demonstrates the consolidation of multi-modal AI workloads on a single edge device.
Demonstrates the concurrent execution of five AI models in the context of monitoring newborns patients in warmer units.
Demonstrates how different Intel accelerators may be applied through DL Streamer to an endoscopic polyp detection task.
Tools and Libraries#
Deep Learning Streamer (DL Streamer)
Open-source GStreamer framework-based media analytics framework used by video-centric applications.
DL Streamer Pipeline Server
Python code-based, containerized microservice for video analytics pipelines built using DL Streamer.
OpenVINO™ Toolkit
Optimizes and deploys AI models across Intel® hardware in cloud, edge, and on premises environments.
OpenVINO™ Model Server
Hosts AI models via networks, e.g. enabling Multi-Modal Patient Monitoring to share a single monitoring model.