| Full text | |
| Author(s): |
Raniero, L.
[1]
;
Canevari, R. A.
[1]
;
Ramalho, L. N. Z.
[2]
;
Ramalho, F. S.
[2]
;
dos Santos, E. A. P.
[1]
;
Bitar, R. A.
[2]
;
Jalkanen, K. J.
[1, 3]
;
Martinho, H. S.
[4]
;
Martin, A. A.
[1]
Total Authors: 9
|
| Affiliation: | [1] Univ Vale Paraiba, UniVap, IP&D, Lab Biomed Vibrat Spect LEVB, Inst Res & Dev, BR-12244000 Sao Jose Dos Campos, SP - Brazil
[2] Univ Sao Paulo, Dept Patol & Med Legal, Fac Med Ribeirao Preto, BR-14049900 Ribeirao Preto, SP - Brazil
[3] Tech Univ Denmark, Dept Phys, Quantum Prot QuP Ctr, DK-2800 Lyngby - Denmark
[4] Univ Fed ABC, Ctr Ciencias Nat & Humanas, BR-09210170 Santo Andre, SP - Brazil
Total Affiliations: 4
|
| Document type: | Journal article |
| Source: | THEORETICAL CHEMISTRY ACCOUNTS; v. 130, n. 4-6, p. 1239-1247, DEC 2011. |
| Web of Science Citations: | 21 |
| Abstract | |
In this work, Raman spectra in the 900-1,800 cm(-1) wavenumber region of in vivo and ex vivo breast tissues of both healthy mice (normal) and mice with induced mammary gland tumors (abnormal) were measured. In the case of the in vivo tissues, the Raman spectra were collected for both transcutaneous (with skin) and skin-removed tissues. To identify the spectral differences between normal and cancer breast tissue, the paired t-test was carried out for each wavenumber using the whole spectral range from both groups. Quadratic discriminate analysis based on principal component analysis (PCA) was also used to determine and evaluate differences in the Raman spectra for the various samples as a basis for diagnostic purposes. The differences in the Raman spectra of the samples were due to biochemical changes at the molecular, cellular and tissue levels. The sensitivity and specificity of the classification scheme based on the differences in the Raman spectra obtained by PCA were evaluated using the receiver operating characteristic (ROC) curve. The in vivo transcutaneous normal and abnormal tissues were correctly classified based on their measured Raman spectra with a discriminant proportion of 73%, while the in vivo skin-removed normal and abnormal tissues were correctly classified again based on their measured Raman spectra with a discriminant proportion of 86%. This result reveals a strong influence due to the skin of the breast, which decreased the specificity by 11%. Finally, the results from ex vivo measurements gave the highest specificity and sensitivity: 96 and 97%, respectively, as well as a largest percentage for correct discrimination: 94%. Now that the important bands have been experimentally determined in this and other works, what remains is for first principles molecular-level simulations to determine whether the changes are simply due to conformational changes, due to aggregation, due to changes in the environment, or complex interactions of all of the above. (AU) | |