Customized software solution

Customize your software! ZEISS ZEN core can be expanded using the software's own macro environment (OAD - Open Application Development) and Python. The ZEN library for advanced analysis and control of the software is freely available on github.com. This means that you can solve even the most complex analysis tasks either using Open Source code and your own programming or as a service from ZEISS. We would be happy to advise you further.

Standard software doesn't meet all your needs? We offer you a customized software solution for complex image analyses tailored precisely to your requirements and wishes.

• Robot control & robot loading
• Connection to external software
• Analyses running in the background
• Integration into comprehensive workflows
• Control of external systems and lighting

Smith & Nephew is a British international company that manufactures medical devices and innovative products for wound care and arthroscopy, trauma and clinical therapy as well as orthopaedic reconstruction.

Situation
Smith & Nephew was looking for software to evaluate the layer thickness and porosity of coatings on medical implants in accordance with ASTM F1854. The standard ZEISS ZEN core solutions could only achieve results up to a certain point and did not fully meet the requirements.

Our solution
ZEISS has tackled the problem and developed a customized software solution for the company. With the extension of the ZEISS ZEN core software with a customized module for AI-based coating thickness detection and porosity measurement in an automated workflow and a user-defined report, all needs and requirements of Smith & Nephew could be met.

The benefits

• Automation through AI enables an increase in productivity
• Human influences are minimized

Electromobility is being increasingly expanded and focused on, with lithium-ion batteries playing a key role in the automotive industry. Not only are capacity and longevity important, but above all the safety of the battery must be guaranteed. To check the battery for defects, neural networks can help with automatic defect detection on a microscopic scale.

As part of a project, Aalen University has examined a prismatic lithium-ion battery (NMC) for plug-in electric vehicles in more detail with the help of AI modules from the ZEN core software suite. An AI model was trained to recognize and evaluate the microstructure of the battery. Thus, defects such as cracks, kinks, inclusions etc. can be localized.

The results of the analyses can be illustrated using a heat map shown in this image. Blue colorations represent minor or no deviations from the expected structure. The higher the proportion of red in this visualization, the more the result deviates from the learned structure and signals a defect. These analyses can ensure the safety and compliance with quality standards of the lithium-ion battery.¹