Classification of Soft Tissue Sarcoma Specimens with Raman Spectroscopy as Smart Sensing Technology

Area of Science:

• Oncology
• Biomedical Engineering
• Spectroscopy

Background:

• Intraoperative margin assessment is crucial for soft tissue sarcoma surgery.
• Frozen section analysis is the current gold standard but lacks real-time feedback.
• Current methods delay surgical completion due to lengthy histologic examination.

Purpose of the Study:

• To develop and study Raman spectroscopy for real-time detection and classification of soft tissue sarcoma.
• To assess the feasibility of using Raman spectroscopy for intraoperative margin status evaluation.
• To compare quantitative and machine learning methods for spectral pattern analysis.

Main Methods:

• Acquired Raman spectra from sarcoma and adjacent benign tissues (muscle, fat, dermis) during surgery.
• Developed a quantitative method (QM) and a machine learning method (MLM) to analyze spectral data.
• Compared spectral analysis results with adjacent H&E-stained frozen section findings.

Main Results:

• Both QM and MLM achieved high classification accuracy (>85%) for differentiating tissue types.
• Raman spectroscopy effectively identified sarcoma, muscle, fat, and dermal tissues.
• The methodology demonstrates potential for real-time in vivo tissue assessment.

Conclusions:

• Raman spectroscopy offers a viable, real-time alternative for intraoperative margin assessment in sarcoma surgery.
• This technology could significantly reduce surgical time by providing immediate feedback.
• Further development with a hand-held probe could enable in vivo application in the operating room.