オックスフォード・インストゥルメンツー事業部ページ
拡張

Imaris 9.9 - Segment with Machine Learning

Imaris 9.9 adds a new pixel classification segmentation method powered by Labkit. The addition of machine learning pixel classification with an intuitive and interactive training mode broadens the diversity of images for analysis as it enables some electron microscopy segmentation and shape recognition. In addition, Imaris 9.9 is the most open and flexible Imaris version ever. We further facilitate the connection between open-source software packages and Imaris by enabling the ability to directly import label images or position data as Surfaces or Spots, respectively.

Request Pricing/Free Trial

Machine Learning Pixel Classification using Labkit and Imaris

Inside the Imaris Surface Creation Wizard you can now select the Labkit pixel classification segmentation method. Training your pixel classifier with a few brush strokes leads to smart segmentation of your data. This method can be applied to fluorescent data as well as some greyscale images.

All other parts of Imaris Surface creation workflow, such as filtering, editing or object classification are available with this method. It makes the workflow very smooth and opens new great image analysis and visualization options.

Import of Label Images as Imaris Surfaces

All microscopists know that some segmentation problems may be best solved with a specific open source software using machine learning or other tools. Those labelled images (from ilastik, StarDist, Labkit and many more) can now be imported directly into Imaris and visualized as Surfaces (together with raw data) with immediate access to all Imaris statistics calculated from the segmentation.

Surfaces imported as labelled images can be visualized using Imaris’ choice of materials, including semi-transparent ones, with color-coding or even further classified in Imaris with the built-in machine learning object classifier.

Import of Spots and Tracks from Open Source tools into Imaris

Tracking objects like single nuclei in developing embryos can be a challenging task. Software packages like MaMuT – a tracking and track-editing framework for large, multi-view images can help to analyse the data.

Imaris 9.9 provides a direct import of the resulting Spots and Tracks for users who prefer to work in both MaMuT (or TrackMate) as well Imaris. By importing the data into Imaris users have access to Imaris’ 3D rendering with object color-coding and multiple visualization modes. Overlaying tracking results with the raw, multi-channel microscopy data unlocks the possibility to create stunning snapshots and animations.

Imaris Workflows with Machine Learning and Open Source tools

Using Pixel Classification in Imaris – Labkit Segmentation

 
 
 
 
 
 

Segmenting Cell Nuclei in StarDist and importing them into Imaris

 

Additional Resources

The Imaris Learning Center hosts a wide range of tutorial videos, how-to articles and webinars to guide you through the many features of Imaris. We have provided some links below which will get you started on some of our most recent developments.