© 2022 All Rights Reserved

Disease diagnosis by FTIR spectroscopy

Denise M. Zezell 1*, Pedro A.A. Castro 1, Matheus del Valle 1, Kleber Stancari 1, Moises O. Santos1,2
1 Center for Lasers and Applications, Nuclear and Energy Research Institute IPEN/CNEN, Sao Paulo-SP, Brazil
2 Amazonas State University UEA, Manaus-AM, Brazil


The infrared (IR) microspectroscopy has proved to be an ideal tool for investigating the biochemical composition of biological samples at the microscopic scale, as well as its fast, sensitive, and label-free nature, with the potential to complement current pathological methods, reducing subjectivity in biopsy samples analysis. In this talk, I shall introduce the basis of data pre-processing, which is a very sensitive matter, with imposition of selection criteria to avoid pixels not covered by tissue and/or those that displayed excessively strong scattering effects, water vapour or CO2 presence. After pre-processing, normalization and smoothing, spectral datasets are subsequently converted to pseudocolor images using for instance hierarchical cluster analysis (HCA), which clusters patterns in a dataset based on their spectral similarity, and the most suitable method of clustering is chosen. Principal Component Analysis (PCA) is used for data exploration and dimensionality reduction for prediction models, thus decreasing training time and overfitting. Linear Discriminant Analysis (LDA), Partial Least Squares (PLS), Support Vector Machine (SVM), Random Forest (RF) algorithms among others are commonly trained as classification methods, where their accuracy, sensitivity and specificity are assessed through cross validation tests. I will describe our research in Dentistry and our ongoing research using FTIR of biopsied tissues, such as burned skin, diagnose and molecular differentiation between thyroid and different breast cancer subtypes, and cutaneous tumor tissue from healthy skin, treated by ALA-MEALA Photodynamic Therapy.
ACKNOWLEDGEMENTS: This work was supported by National Institute of Photonics CNPq INCT-465763/2014-6, CNPq PQ-309902/2017-7, FAPESP 17/50332-0 and 17/07519-2, and CNEN. I would like to gratefully thank all students and collaborators who contributed to the results presented in this talk.


Denise Maria Zezell
Nuclear and Energy Research Institute IPEN/CNEN


Ask question