Presentation + Paper
12 March 2024 Innovative approach to soft tissue classification using fiber-optic Raman probe as a smart sensing tool
Soha Yousuf, Mohamed Irfan Karukappadath, Azhar Zam
Author Affiliations +
Abstract
Achieving accurate resections is crucial for confirming the success of surgical procedures during the operation. However, the time-consuming histologic analysis lacks the ability to provide real-time diagnoses, resulting in delays in surgical procedures. As a result, there is an urgent need for real-time assessment of both healthy and cancerous soft tissue. In this study, Raman spectral data from diverse bovine tissue were acquired using a 785 nm fiber-optic Raman probe system and utilized for classification through a Random Forest (RF) classifier. The study entailed a quantitative and experimental analysis, utilizing locally collected bovine samples, including muscle, fat, bone, and bone marrow, with Raman spectra obtained from 1200 sites across 24 samples. The Random Forest analysis demonstrated significant potential for distinguishing between various types of bovine tissue, achieving an average accuracy of approximately 99.5%, specificity of about 99.7%, and sensitivity of about 99.1%. The integration of machine learning techniques with Raman sensing technology shows immense potential in facilitating real-time, intraoperative, in vivo evaluations of soft tissues.
Conference Presentation
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Soha Yousuf, Mohamed Irfan Karukappadath, and Azhar Zam "Innovative approach to soft tissue classification using fiber-optic Raman probe as a smart sensing tool", Proc. SPIE 12855, Advanced Chemical Microscopy for Life Science and Translational Medicine 2024, 1285508 (12 March 2024); https://doi.org/10.1117/12.3003633
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Tissues

Raman spectroscopy

Bone

Muscles

Biological samples

Random forests

Machine learning

Back to Top