Related Experiment Video
Updated: Sep 25, 2025
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
Classification of tissue biopsies by Raman spectroscopy guided by quantitative phase imaging and its application to
Almog Taieb1, Garry Berkovic1,2, Miki Haifler1,3
1Department of Biomedical Engineering, Faculty of Engineering, Tel Aviv University, Tel Aviv, Israel.
This study introduces a label-free optical method combining quantitative phase imaging and Raman spectroscopy for precise tissue analysis. This approach significantly enhances bladder cancer urothelium classification accuracy by improving spectral analysis. Keywords: label-free optical measurement, quantitative phase imaging, Raman spectroscopy, bladder cancer classification.
Area of Science:
- Biomedical Optics
- Optical Spectroscopy
- Computational Pathology
Background:
- Accurate tissue biopsy analysis is crucial for disease diagnosis, particularly in oncology.
- Current methods often rely on staining, which can introduce artifacts and alter tissue properties.
- Label-free techniques offer a promising alternative for preserving native tissue characteristics.
Purpose of the Study:
- To develop and validate a multimodal, label-free optical measurement approach for analyzing sliced tissue biopsies.
- To combine quantitative phase imaging (QPI) and localized Raman spectroscopy (LRS) for enhanced tissue segmentation and classification.
- To improve the accuracy of classifying benign and malignant bladder cancer tissues.
Main Methods:
- Automated scanning for label-free QPI of unstained tissue slices.
- Pixel-wise tissue segmentation using a support vector machine based on Haralick texture features from QPI data.
- Location-guided LRS measurements guided by QPI segmentation.
- Sparse multinomial logistic regression (SMLR) for classifying tissue types based on Raman spectra.
Main Results:
- Successful segmentation of urothelium in benign and malignant bladder cancer tissues using QPI.
- Demonstrated improvement in Raman spectra classification accuracy from 85.7% to 94.7% after QPI-based segmentation.
- The multimodal approach effectively integrates structural information from QPI with chemical information from LRS.
Conclusions:
- The presented multimodal label-free optical measurement approach provides a powerful tool for analyzing tissue biopsies.
- Combining QPI for segmentation and LRS for chemical analysis significantly enhances diagnostic accuracy in bladder cancer.
- This label-free methodology holds potential for advancing histopathology and cancer diagnostics.
