TNO

About

Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek (TNO; English: Netherlands Organisation for Applied Scientific Research) is an independent research organisation in the Netherlands that focuses on applied science.

The organisation also conducts contract research, offers specialist consulting services, and grants licences for patents and specialist software. TNO tests and certifies products and services, and issues an independent evaluation of quality. Moreover, TNO sets up new companies to market innovations.

Wound recognition and -analysis with Smart Glasses and Artificial Intelligence.

TNO and Gemvision are pilot-testing algorithms developed for object and character recognition, being used to recognize wounds with multispectral imaging. This project is currently in a pilot phase involving companies and hospitals.

The Goal: The complete solution for wound care with wound care video-calling, wound recognition via Smart Glasses and wound analysis with Artificial Intelligence. This could greatly benefit healthcare professionals and make accurate analysis about the wound healing process.

By making estimates of wounds with digital techniques, these can be better evaluated and measures can be taken more quickly to better nurse wounds. Think, for example, of pressure ulcers. This saves the patient a lot of suffering and offers a faster recovery, in addition, this saves the nursing institution a lot of costs and the hospital can work much more efficiently (cost-saving). Broader applications are therefore envisaged for the medical world, especially in telemedicine.

More and better optical sensors are developing. For example, high-frame rate snapshot multispectral cameras are being developed. These cameras can see light at more wavelengths than the human eye. It has already been shown that this type of technology can be used for better tissue differentiation during surgery, and for example also that it can say something about the perfusion status of tissues. The latter property in particular is interesting for wound diagnostics. Interpreting hyperspectral information is not trivial, however, there may also be a role for the mixed reality setup. How can this technology be developed into an AR application? This is what we are doing.

Example

The image below shows a first person view of a wound being recognised. The first step is de localisation and blurring of the surrounding image. This is done for privacy reasons. Only the wound matters in this case. When de wound is identified, the surface is mapped and analysed. This can then be compared to previous stages and form a conclusion on how the healing progress is.