A Multi-Modal Approach for Exploring Sarcoma and Carcinoma Using FTIR and Polarimetric Analysis

Area of Science:

• Biomedical Optics
• Spectroscopy
• Computational Pathology

Background:

• Accurate cancer diagnosis is crucial for effective treatment.
• Novel diagnostic methods are needed to detect subtle cancerous features.
• Distinguishing between sarcoma and carcinoma requires precise tissue characterization.

Purpose of the Study:

• To evaluate Fourier transform infrared (FTIR) spectroscopy and Mueller matrix polarimetry (MMP) for ex vivo tissue analysis.
• To assess the potential of support vector machines (SVM) in classifying sarcoma and carcinoma.
• To explore combined optical techniques for automated pathology.

Main Methods:

• Ex vivo tissue samples of sarcoma and carcinoma were analyzed using FTIR spectroscopy and MMP.
• FTIR characterized chemical composition (O-H, C-H, N-H bonds) across 400-4000 cm⁻¹.
• MMP analyzed 13 polarimetric parameters in the visible spectrum (400-800 nm).
• Support vector machines (SVM) were employed for computer-assisted classification.

Main Results:

• FTIR spectra revealed distinct chemical composition differences between sarcoma and carcinoma tissues.
• MMP showed significantly higher polarimetric parameter values in sarcoma compared to carcinoma.
• Microscopic analysis confirmed morphological variations correlating with optical findings.
• SVM achieved 90% accuracy, sensitivity, and specificity in classifying tissue types.

Conclusions:

• The combination of optical polarimetry and FTIR holds significant potential for automated pathology.
• SVM analysis of optical data enables accurate classification of sarcoma and carcinoma.
• These advanced optical methods offer a promising approach for objective cancer diagnosis.