Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics
Abstract
Excitation-emission matrices (EEM) and total synchronous fluorescence spectra (SFS) of normal and malignant breast tissue specimens are measured in UV-VIS spectral region to serve as data inputs in development of Support Vector Machine (SVM) based breast cancer diagnostics tool. Various input data combinations are tested for classification accuracy using SVM prediction against histopathology findings to discover the best combination regarding diagnostics sensitivity and specificity. It is shown that with EEM data SVM provided 67 % sensitivity and 62 % specificity diagnostics. With SFS data SVM provided 100 % sensitivity and specificity for a several input data combinations. Among these combinations those that require minimal data inputs are identified.
Keywords:
Breast cancer / Fluorescence / Support vector machine / Synchronous fluorescenceSource:
Journal of Fluorescence, 2012, 22, 5, 1281-1289Funding / projects:
- Materials of Reduced Dimensions for Efficient Light Harvesting and Energy conversion (RS-MESTD-Integrated and Interdisciplinary Research (IIR or III)-45020)
- Molecular determinants for tumor marker design (RS-MESTD-Basic Research (BR or ON)-173049)
DOI: 10.1007/s10895-012-1070-0
ISSN: 1053-0509
PubMed: 22678149
WoS: 000308064100011
Scopus: 2-s2.0-84869082830
Collections
Institution/Community
VinčaTY - JOUR AU - Dramićanin, Tatjana AU - Lenhardt, Lea I. AU - Zeković, Ivana Lj. AU - Dramićanin, Miroslav PY - 2012 UR - https://vinar.vin.bg.ac.rs/handle/123456789/5020 AB - Excitation-emission matrices (EEM) and total synchronous fluorescence spectra (SFS) of normal and malignant breast tissue specimens are measured in UV-VIS spectral region to serve as data inputs in development of Support Vector Machine (SVM) based breast cancer diagnostics tool. Various input data combinations are tested for classification accuracy using SVM prediction against histopathology findings to discover the best combination regarding diagnostics sensitivity and specificity. It is shown that with EEM data SVM provided 67 % sensitivity and 62 % specificity diagnostics. With SFS data SVM provided 100 % sensitivity and specificity for a several input data combinations. Among these combinations those that require minimal data inputs are identified. T2 - Journal of Fluorescence T1 - Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics VL - 22 IS - 5 SP - 1281 EP - 1289 DO - 10.1007/s10895-012-1070-0 ER -
@article{ author = "Dramićanin, Tatjana and Lenhardt, Lea I. and Zeković, Ivana Lj. and Dramićanin, Miroslav", year = "2012", abstract = "Excitation-emission matrices (EEM) and total synchronous fluorescence spectra (SFS) of normal and malignant breast tissue specimens are measured in UV-VIS spectral region to serve as data inputs in development of Support Vector Machine (SVM) based breast cancer diagnostics tool. Various input data combinations are tested for classification accuracy using SVM prediction against histopathology findings to discover the best combination regarding diagnostics sensitivity and specificity. It is shown that with EEM data SVM provided 67 % sensitivity and 62 % specificity diagnostics. With SFS data SVM provided 100 % sensitivity and specificity for a several input data combinations. Among these combinations those that require minimal data inputs are identified.", journal = "Journal of Fluorescence", title = "Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics", volume = "22", number = "5", pages = "1281-1289", doi = "10.1007/s10895-012-1070-0" }
Dramićanin, T., Lenhardt, L. I., Zeković, I. Lj.,& Dramićanin, M.. (2012). Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics. in Journal of Fluorescence, 22(5), 1281-1289. https://doi.org/10.1007/s10895-012-1070-0
Dramićanin T, Lenhardt LI, Zeković IL, Dramićanin M. Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics. in Journal of Fluorescence. 2012;22(5):1281-1289. doi:10.1007/s10895-012-1070-0 .
Dramićanin, Tatjana, Lenhardt, Lea I., Zeković, Ivana Lj., Dramićanin, Miroslav, "Support Vector Machine on Fluorescence Landscapes for Breast Cancer Diagnostics" in Journal of Fluorescence, 22, no. 5 (2012):1281-1289, https://doi.org/10.1007/s10895-012-1070-0 . .