Endoscopic In Vivo Hyperspectral Imaging for Head and Neck Tumor Surgeries Using a Medically Approved CE-Certified...

Area of Science:

• Surgical Oncology
• Medical Imaging
• Biomedical Engineering

Background:

• Accurate in vivo visualization of malignant tumors is critical for successful head and neck cancer surgery.
• Inadequate tumor margin assessment and incomplete resection negatively impact patient outcomes.
• Hyperspectral imaging (HSI) shows promise for intraoperative tumor visualization but faces regulatory and workflow integration challenges.

Purpose of the Study:

• To develop and evaluate an endoscopic hyperspectral imaging (HSI) workflow for in vivo tumor visualization during head and neck cancer surgery.
• To optimize the HSI workflow for minimal intraoperative time and surgeon adaptation.
• To enable real-time data processing for classification and visualization of HSI data.

Main Methods:

• Endoscopic HSI was applied to 12 head and neck cancer patients during surgery.
• Workflow optimization focused on reducing procedure time and surgeon learning curve.
• Principal component analysis and k-means clustering were used for HSI data evaluation, validated against expert annotations.

Main Results:

• The complete HSI workflow averaged 2-3 minutes, with individual measurements under 10 seconds.
• The system achieved 79% accuracy, 72% sensitivity, and 84% specificity in tissue differentiation.
• Surgical staff demonstrated proficiency after just two procedures.

Conclusions:

• An optimized HSI workflow facilitates accurate, real-time in vivo tissue differentiation during head and neck cancer surgery.
• This approach improves intraoperative evaluation of tumor margins, potentially leading to more complete resections.
• The enhanced visualization provided by HSI can contribute to improved patient outcomes in head and neck cancer surgery.