Deep Learning & AI

AI tools built for tissue pathology

Convolutional neural networks (CNNs) are driving the second paradigm shift in AI for image analysis. Conventional image analysis and machine learning are powerful but limited by how well you manually create the rules of your algorithms.

What is the difference between artificial intelligence (AI) and deep learning?

Deep learning is part of a broader family of artificial intelligence (AI) and machine learning methods. Deep learning is based on the latest technology and a much more advanced approach as it learns from underlying features in data using deep neural networks.

Artificial intelligence
Any technique that enables computers to mimic human behaviour.
Machine learning
The ability to learn without being directly programmed.
Deep learning
The learning of underlying features in data using deep neural networks.
Convolutional neural networks

How AI at Visiopharm is optimized for tissue pathologists and researchers

With convolutional neural networks (CNNs), a type of deep learning algorithms that are taught to recognize specific patterns, the underlying features and rules are learned directly from the images. Specifically, these neural networks use a cascade of layers with feature extraction and
transformations, with each successive layer using the output from the previous layer as the input, thus forming a hierarchical representation of the image. This means that the rules of the algorithm are learned automatically.

The deep learning technology in Visiopharm’s AI image analysis platform has been specifically
developed for the field of histopathology, so you are able to apply, train AND create high-quality
deep learning algorithms to obtain breakthrough results in your field of work.

With our platform, you get the power of state-of-the-art deep learning to solve your problems –
without having to write a single line of code.

Solve your most challenging problems with AI module

The AI module facilitates automatic cell and nuclei identification on H&E and IHC stained tissue sections – saving time and staining of multiple sections.
Obtain objective and reproducible results with our deep learning algorithms for any quantitative cell type analyses in combination with e.g. tumor and stroma segmentation.
Most cancer applications involve cell identification (classification + segmentation) and tissue compartmentalization. Manually solving this task is time-consuming, and highly challenging for the untrained eye, and often leads to subjective results. It is no longer the case with deep learning.

Analysis of your tissue samples with AI should not be a barrier, it should be your enabler.

Author your own AI APPs

Develop image analysis protocols and workflows without the need for AI expertise. Use our unique authoring capabilities to tailor an existing APP  to suit your needs, or to build your solution that delivers precise and accurate information without needing any programming experience.

The modular design of the software allows you to combine APPs and build a solution that answers your specific questions.

  • Independently customizable: Use different classifiers and magnifications for each step
  • Reuse and share APPs: Save your configuration and use them as often as you want
  • Report outputs at any stage of the workflow (e.g. tissue area, cell counts, proliferation index)
  • Easily modify APPs without rewriting an entire workflow
Vimeo

By loading the video, you agree to Vimeo's privacy policy.
Learn more

Load video

For advanced users

Build, visualize, and edit your deep learning networks

AI for the experts

The AI platform makes it a lot easier to solve challenging image analysis problems in tissue-based research. The AI Architect module allows expert users to customize and control deep learning topologies such as training parameters, network architecture, and more.

This module gives access to the latest breakthroughs in AI and Deep Learning, representing the most comprehensive and configurable platform for artificial intelligence-based image analysis for digital pathology.

More than just deep learning – A deeper look into our technology

Understanding the biology of tissues calls for a complete toolbox of artificial intelligence techniques. Along with a host of machine learning techniques, the Visiopharm software is equipped with three types of deep learning classifiers. No matter which type of objects you need to classify, the full selection of AI tools is available at the click of a button.

Success
Your message has been successfully sent!