- Home
- Products
- Software
- ZEN Intellesis
ZEISS ZEN Intellesis for Image Segmentation in Microscopy
Use the Power of Deep Learning to Easily Segment Your Images
As enjoyable as microscopic images are - their real value is in the data they provide. Image segmentation is currently one of the biggest challenges in microscopy and the foundation for all subsequent image analysis steps. ZEN Intellesis uses established machine-learning techniques powered by Python, such as pixel classification and deep learning to easily create robust and reproducible segmentation results, even for non-experts. You can now train the software once and then ZEN Intellesis can segment a batch of hundreds of images automatically. You save time and minimize user bias.
For additional information and technical details please use Open Application Development - Machine Learning *1
Highlights
Use Deep Learning to Segment Your Images
To get reliable data from images, you often need to segment different classes of objects in multiple images. With ZEN Intellesis you can now use your expertise to train the software on a few images. All time-consuming segmentation steps on the hundreds of similar images will then be done by the powerful machine learning algorithms (including Deep Learning) of the software module ZEN Intellesis. The Python-powered tools offers pixel classification with real multichannel feature extraction and segmentation by pre-trained networks *2, even complex multidimensional, multi-modal data can be analyzed, regardless of their origin. The software even allows to import and use your own deep learning models *3
Enjoy Smooth Workflow Integration
The software module ZEN Intellesis makes machine learning easy to use: You simply load your image, define your classes, label pixels, train your model and perform the image segmentation. For common tasks like Nucleus Segmentation one can even import a pretrained model, which can be used to segment and analyze full datasets. Moreover, it is even possible to train your own network and import into ZEN. (The creation of customized pre-trained networks is also available as a paid service). Intellesis is fully integrated into the Image Analysis Framework in ZEN Blue and ZEN core Imaging Software and support scripting. This integration ensures that all valuable Metadata stay connected and are available for further processing steps.
Analyze Multi-Modal Images from Many Sources, in Many Formats
With the software module ZEN Intellesis you can easily segment multidimensional images from a wealth of different imaging sources, as diverse as
- Widefield Microscopy
- Superresolution Microscopy
- Fluorescence Microscopy
- Label Free Microscopy
- Confocal Microscopy
- Light Sheet Microscopy
- Electron Microscopy
- X-ray Microscopy
Together with ZEN Connect one can combine images from different microscopes from the same sample and use those results to extract even more valuable information by using ZEN Intellesis. The software will by use image features from all modalities at once to segment the structures of interest. By importing OME-TIFF or TXM images directly or when using the 3rd party import function ZEN Intellesis can also be used to segment all Bio-Formats compatible images.
Application Example Life Sciences
1. Scratch Assay



2. Spines and Dendrites


3. Drosophila



Application Example Materials Analysis




How to Segment Your Images
Download
ZEISS ZEN Intellesis - Flyer
Segmentation and Classification by Machine Learning
pages: 4
file size: 871 kB
ZEISS ZEN Intellesis for Life Science
Image Segmentation for 2D and 3D Datasets
pages: 2
file size: 1149 kB
ZEISS ZEN Intellesis for Materials Science
Your Imaging Software for Machine Learning
pages: 2
file size: 2262 kB
Advanced Segmentation for Industrial Materials
Learn about machine learning – a solution to the segmentation problem with ZEISS ZEN Intellesis for initial data generation or analysis and processing of a segmented image.
pages: 11
file size: 4794 kB
ZEISS ZEN Intellesis
Machine Learning Approaches for Easy and Precise Image Segmentation
pages: 8
file size: 5151 kB
Results 1 - 5 of 5