Tissue optics in endoscopy of the future
|Funding from Vinnova||SEK 545 147|
|Project duration||September 2016 - December 2017|
Purpose and goal
In BioOptico we develop algorithms for streaming video in medicine and license them to manufacturers of endoscopic cameras. In this project we set out to develop a general framework based on our algorithms to more easily be able to apply them in new endoscopic applications and markets. More specifically it was about machine learning based on convolutional neural networks, trained with medical image data from different endoscopic applications. The project results were very positive, including a developed well working framework for different platforms, validated with clinical data.
Expected results and effects
The project results convinced our American customer within arthroscopy to start a major commercialization project together with us. In the project we also found good IP openings and will file a patent application. We also attracted another American customer within otoscopy, based on the framework and the results. Both these tracks will lead to growth in BioOptico and probably to good and successful medical products within the upcoming years.
Planned approach and implementation
The project consisted of three work packages, one technical where we developed the framework, one clinical where we trained and validated it with clinical data (arthroscopy) and one work package where we investigated the IP landscape. We fulfilled all goals within the project.