Definiens Cognition Network Technology®

Definiens’ revolutionary, patented Definiens Cognition Network Technology® is the foundation of the company’s leading position in the field of image intelligence.


Definiens Cognition Network Technology® is the brainchild of the pioneering development team around Nobel Laureate Prof. Gerd Binnig, Definiens’ founder and Head of Research. The team’s goal was to find a more effective method for extracting intelligence from digital data of any kind, including text, images, numbers, and tables. Definiens Cognition Network Technology® is the result of their intense development efforts.

The human mind is unique in its ability to make sense of objects, patterns, and context, and in its ability to extract intelligence from this data. Conventional computer-based methods fall embarrassingly short of this natural knack for learning and cognition.

To emulate the human mind’s cognitive powers more faithfully, Definiens took a radical departure from conventional approaches. Using patented segmentation and classification processes, our team has developed a robust methodology for rendering knowledge within a semantic network.

Principles of Definiens Cognition Network Technology

This innovative new technology examines pixels not in isolation, but in context. It builds up an image iteratively, recognizing groups of pixels as objects. Just like the human mind, it uses the colors, shapes, textures, and sizes of objects as well as their context and relationships to draw the same conclusions and inferences that an experienced analyst would draw.

When applied to image analysis, the outcome of this semantic and context-based segmentation approach is an unsurpassed accuracy and consistency in the detection of structures of interest – despite all the challenges that biomedical image analysis entails:

  • biological and phenotypic variation,
  • variability in staining and illumination, and
  • heterogeneous or ambiguous structures.

This has a major impact on the end data and intelligence that you can receive from image analysis: more accurate segmentation directly results in superior measurements and readouts.