<  Back to the Polytechnique Montréal portal

Data consistency and classification model transferability across biomedical Raman spectroscopy systems

Fabien Picot, François Daoust, Guillaume Sheehy, Frédérick Dallaire, Layal Chaikho, Théophile Bégin, Samuel Kadoury and Frédéric Leblond

Article (2020)

[img]
Preview
Published Version
Terms of Use: Creative Commons Attribution .
Download (1MB)
Cite this document: Picot, F., Daoust, F., Sheehy, G., Dallaire, F., Chaikho, L., Bégin, T., ... Leblond, F. (2020). Data consistency and classification model transferability across biomedical Raman spectroscopy systems. Translational Biophotonics, 3(1), p. 1-11. doi:10.1002/tbio.202000019
Show abstract Hide abstract

Abstract

Surgical guidance applications using Raman spectroscopy are being developed at a rapid pace in oncology to ensure safe and complete tumor resection during surgery. Clinical translation of these approaches relies on the acquisition of large spectral and histopathological data sets to train classification models. Data calibration must ensure compatibility across Raman systems and predictive model transferability to allow multi-centric studies to be conducted. This paper addresses issues relating to Raman measurement standardization by first comparing Raman spectral measurements made on an optical phantom and acquired with nine distinct point probe systems and one wide-field imaging instrument. Data standardization method led to normalized root-mean-square deviations between instruments of 2%. A classification model discriminating between white and gray matter was trained with one point probe system. When used to classify independent data sets acquired with the other systems, model predictions led to >95% accuracy, preliminarily demonstrating model transferability across different biomedical Raman spectroscopy instruments.

Uncontrolled Keywords

cancer, classification models, medical imaging, Raman spectroscopy

Open Access document in PolyPublie
Subjects: 3100 Physique > 3100 Physique
3100 Physique > 3101 Études atomiques et moléculaires
3100 Physique > 3111 Laser
Department: Département de génie physique
Research Center: Non applicable
Funders: CRSNG/NSERC - Discovery Grant Program, TransMedTech Institute, Canadian Institutes of Health Research & CRSNG/NSERC - Collaborative Health Research Program, Canadian Network for Research and Innovation in Machining Technology
Date Deposited: 26 Jan 2022 15:22
Last Modified: 27 Jan 2022 01:20
PolyPublie URL: https://publications.polymtl.ca/9253/
Document issued by the official publisher
Journal Title: Translational Biophotonics (vol. 3, no. 1)
Publisher: Wiley
Official URL: https://doi.org/10.1002/tbio.202000019

Statistics

Total downloads

Downloads per month in the last year

Origin of downloads

Dimensions

Repository Staff Only