Sustainably Automate the Manual Cell-Staining process using revolutionary AI-based Solution

In culturing biological cells, process for staining cells requires toxic reagents which kill cells in this process. Open-Source software-AI Connect for Scientific Data(AiCSD) leverages innovative virtual staining AI-models and streamlines complex manual work by securely connecting scientific devices like imaging microscope to on-premise AI pipelines for image analysis and virtual staining. AiCSD leverages containerized microservices based architecture, Open-Source EdgeX Foundries microservices that automatically detect, manage, and process images securely received from OEM equipment, Open-Source tool BentoML to build, configure and deploy these pipelines and Open-Source Intel® Distribution of OpenVINO™ toolkit to optimize and run models on intel architecture. Presentation will illustrate how this revolutionary solution-AiCSD was architected to build a sustainable cell-staining solution for biopharma industry. We will also cover how this flexible solution can be extended to other use cases including retail, industrial and agriculture. Presentation will include several unique distributed edge solution challenges and learnings in deployment, security, project execution etc. Learning Objective: 1) Describe the manual cell staining process and disadvantages of using toxins during culturing of biological cells. 2) Explain this innovative AI-based solution to automate the cell staining process, through image processing with images captured via microscope using the open-source microservice-based containerized solution- AI Connect for Scientific Data(AiCSD). 3) Illustrate how AiCSD can automatically transfer, process, and compare scientific data (like images) using AI-assisted pipelines for other custom use cases. 4) Describe the AiCSD microservice-based containerized architecture and integration with open source EdgeX, BentoML and Intel® Distribution of OpenVINO™ toolkit. 5) Share the learnings and challenges involved in developing this distributed container based microservice solution.

Track

Data Science and AI

About the session

The session is approved.

The presenter will allow another presenter.

There are 4 people interested in this session.

Edit Session