Colon Tumor Discrimination Combining Independent Endoscopic Probe-Based Raman Spectroscopy and Optical Coherence...

Area of Science:

• Biomedical Optics
• Medical Diagnostics
• Cancer Research

Background:

• Colorectal cancer is a leading global health concern.
• Current diagnostic methods like colonoscopy often rely on biopsies.
• There is a need for improved, less invasive diagnostic tools.

Purpose of the Study:

• To evaluate the combined diagnostic performance of Raman spectroscopy (RS) and optical coherence tomography (OCT).
• To assess the impact of integrating RS and OCT on machine learning models for colorectal tissue discrimination.
• To determine if combining these optical modalities enhances diagnostic accuracy compared to individual methods.

Main Methods:

• Development of fiber-optic probes for in vivo RS and volumetric OCT measurements.
• Clinical study conducted in an ex vivo setting using the developed probes.
• Validation of individual modalities (RS with spectral preprocessing, OCT with texture analysis) using linear discriminant analysis.
• Integration of RS and OCT data using Bayes rule for combined analysis.

Main Results:

• Raman spectroscopy (RS) achieved an Area Under the Curve (AUC) ROC score of 0.77.
• Optical coherence tomography (OCT) achieved an AUC ROC score of 0.86.
• The combined RS and OCT approach resulted in a significantly improved AUC ROC score of 0.93.

Conclusions:

• The integration of Raman spectroscopy and optical coherence tomography offers a synergistic benefit for colorectal cancer diagnosis.
• Combining these optical modalities with machine learning models substantially enhances the accuracy in discriminating cancerous from healthy tissue.
• This hybrid approach holds promise for improving the diagnostic yield and reducing unnecessary biopsies in colorectal cancer screening.