Workflow and hardware for intraoperative hyperspectral data acquisition in neurosurgery

Area of Science:

• Neurosurgery
• Medical Imaging
• Biomedical Engineering

Background:

• Complete removal of malignant brain tissue is crucial in neurosurgery.
• Differentiating tumor from healthy brain tissue and identifying functional areas are key challenges.
• Hyperspectral imaging (HSI) offers potential for detailed tissue characterization via spectral analysis.

Purpose of the Study:

• To adapt and evaluate a hardware system for intraoperative hyperspectral data acquisition in neurosurgery.
• To compare intraoperative HSI performance with a laboratory setup, analyzing illumination and signal-to-noise ratio (SNR).
• To validate the feasibility and benefits of intraoperative HSI for wavelength-related data acquisition.

Main Methods:

• Development and adaptation of a hardware system for intraoperative HSI.
• Comparison of illumination and SNR between intraoperative and laboratory setups.
• Calculation of pixel- and wavelength-specific SNR for spectral data (550-1000 nm).
• Validation of the intraoperative HSI system and generation of a classification map.

Main Results:

• Significant differences in illumination and SNR were observed between intraoperative and laboratory setups.
• Intraoperative HSI achieved a minimum SNR of 25 dB in relevant image areas.
• The adapted intraoperative HSI system was validated, and recorded data were successfully evaluated.
• A classification map demonstrated the visualization of raw data properties.

Conclusions:

• Intraoperative hyperspectral imaging is feasible and beneficial for wavelength-related data acquisition in neurosurgery.
• The developed HSI system aids in differentiating brain tissue and potentially assessing tissue function.
• Further optimization of hardware and image processing is needed for enhanced tissue quality and function specification.