Hyperspectral Medicine

HELICoiD Demonstrator for Intraoperative Brain Cancer Detection Using Hyperspectral Images

This paper describes the development of a demonstrator within the HELICoiD (HypErspectraL Imaging Cancer Detection) European project, in order to achieve a high-resolution system capable of discriminating between healthy and tumour tissues in neurosurgical operations.

The HELICoiD demonstrator comprises two hyperspectral cameras and an illumination system, covering the spectral range from 400 nm to 1700 nm. A data processing unit manages the demonstrator, and a high-performance platform accelerates the hyperspectral image processing.

Each element is installed in a customized structure designed for surgical environments. This intraoperative demonstrator can help neurosurgeons in the brain tumour resection, avoiding the excessive extraction of healthy tissue. The highly accurate system detects borders of the tumour, avoiding accidental leaving of small tumour tissues. The results are provided by the hyperspectral image classification algorithms in real time, allowing for precise discrimination between healthy and tumourous tissues.

Himar Fabelo, Gustavo M. Callicó, Aurelio Vega, Minerva Alemán, Adrian de Pablo, Roberto Sarmiento
Institute for Applied Microelectronics (IUMA)
University of Las Palmas de Gran Canaria (ULPGC)
Las Palmas de Gran Canaria (Spain)