Skip to main navigation Skip to main content Skip to page footer

Visualization

Getting the best out of your data

The goal of a 3-D histology reconstruction is to usually to gain more insight into the morphology of a structure, or to better understand the spatial context. This starts with a proper visualization of the resulting data set. 

During the reconstruction, the color histology images are transformed and slightly warped such that neighboring slices fit together - this is the key result of the reconstruction process, and also part of our scope of delivery.

 

Movie mode

Straight forward through your data

Quite often, it is already helpful to be able to jump back and forth between neighboring images without the usual reorientation - giving the opportunity to observe structures emerging, moving and vanishing along the stack.

Virtual Slice Images

Different orientation? No problem!

Use reslicing of the 3D data set to generate virtual histology sections(*). What if the specimen had been sliced in a different direction? Give your students the possibility to see how fiber bundles look like when sliced in different ways - it will for sure help to improve their understanding!

Volume rendering

Going 3D

Use a volume renderer of your choice to visualize the 3-D data set. Based on gray-level images - which are part of our scope of delivery - you can generate 3-D views of your data set. Use Look-Up-Tables (LUTs) to focus on different structures of your data set, and boost your understanding of the spatial relations of structures in your data.

For the volume renderings we provide, we typically use so-called physically based rendering techniques, resulting in stunning, photorealistic images.

Structure Segmentation

Would you like to cut out a structure from your tissue sample and visualise it separately? From simple colour-based approaches to AI-supported manual segmentation, we will find the right approach for your data.