One step closer to in-vivo detection of low-grade gliomas
A pilot test confirms that spectral imaging can be a way towards intraoperative and label-free detection of intrinsic brain tumors.
Low-grade gliomas – a group of slow-growing brain tumors – are extremely difficult to detect. Their exact demarcations are hard to retrieve, even using surgical microscopes.
A way to detect these tumors in-vivo would therefore be a breakthrough for surgeons. In that respect, imec’s compact, high-accuracy hyperspectral cameras hold a lot of promise. As was demonstrated by intraoperative pilot tests at the University Hospital of Leuven.
An imec SNAPSCAN VNIR was mounted on a surgical microscope to generate clinical data. That data could then be fed into a deep-learning neural network. Resulting in the knowledge to enable surgeons to discriminate between healthy and anomalous tissue.
The next step is to move to video-rate hyperspectral imaging for real-time tumor classification.
