Fluorescence intrinsic characterization of excitation-emission matrix using multi-dimensional ensemble empirical mode decomposition

Excitation-emission matrix (EEM) fluorescence spectroscopy is a noninvasive method for tissue diagnosis and has become important in clinical use. However, the intrinsic characterization of EEM fluorescence remains unclear. Photobleaching and the complexity of the chemical compounds make it difficult to distinguish individual compounds due to overlapping features. Conventional studies use principal component analysis (PCA) for EEM fluorescence analysis, and the relationship between the EEM features extracted by PCA and diseases has been examined. The spectral features of different tissue constituents are not fully separable or clearly defined. Recently, a non-stationary method called multi-dimensional ensemble empirical mode decomposition (MEEMD) was introduced; this method can extract the intrinsic oscillations on multiple spatial scales without loss of information. The aim of this study was to propose a fluorescence spectroscopy system for EEM measurements and to describe a method for extracting the intrinsic characteristics of EEM by MEEMD. The results indicate that, although PCA provides the principal factor for the spectral features associated with chemical compounds, MEEMD can provide additional intrinsic features with more reliable mapping of the chemical compounds. MEEMD has the potential to extract intrinsic fluorescence features and improve the detection of biochemical changes.

Support Vector Machine on fluorescence landscapes for breast cancer diagnostics

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.

Application of supervised self-organizing maps in breast cancer diagnosis by total synchronous fluorescence spectroscopy

Data from total synchronous fluorescence spectroscopy (TSFS) measurements of normal and malignant breast tissue samples are introduced in supervised self-organizing maps, a type of artificial neural network (ANN), to obtain diagnosis. Three spectral regions in both TSFS patterns and first-derivative TSFS patterns exhibited clear differences between normal and malignant tissue groups, and intensities measured from these regions served as inputs to neural networks. Histology findings are used as the gold standard to train self-organizing maps in a supervised way. Diagnostic accuracy of this procedure is evaluated with sample test groups for two cases, when the neural network uses TSFS data and when the neural network uses data from first-derivative TSFS. In the first case diagnostic sensitivity of 87.1% and specificity of 91.7% are found, while in the second case sensitivity of 100% and specificity of 94.4% are achieved.

Fluorescence spectroscopy: an adjunct diagnostic tool to image-guided core needle biopsy of the breast

We explored the use of a fiber-optic probe for in vivo fluorescence spectroscopy of breast tissues during percutaneous image-guided breast biopsy. A total of 121 biopsy samples with accompanying histological diagnosis were obtained clinically and investigated in this study. The tissue spectra were analyzed using partial least-squares analysis and represented using a set of principal components (PCs) with dramatically reduced data dimension. For nonmalignant tissue samples, a set of PCs that account for the largest amount of variance in the spectra displayed correlation with the percent tissue composition. For all tissue samples, a set of PCs was identified using a Wilcoxon rank-sum test as showing statistically significant differences between: 1) malignant and fibrous/benign; 2) malignant and adipose; and 3) malignant and nonmalignant breast samples. These PCs were used to distinguish malignant from other nonmalignant tissue types using a binary classification scheme based on both linear and nonlinear support vector machine (SVM) and logistic regression (LR). For the sample set investigated in this study, the SVM classifier provided a cross-validated sensitivity and specificity of up to 81% and 87%, respectively, for discrimination between malignant and fibrous/benign samples, and up to 81% and 81%, respectively, for discriminating between malignant and adipose samples. Classification based on LR was used to generate receiver operator curves with an area under the curve (AUC) of 0.87 for discriminating malignant versus fibrous/benign tissues, and an AUC of 0.84 for discriminating malignant from adipose tissue samples. This study demonstrates the feasibility of performing fluorescence spectroscopy during clinical core needle breast biopsy, and the potential of this technique for identifying breast malignancy in vivo.

Model based and empirical spectral analysis for the diagnosis of breast cancer

We explored the use of both empirical (Partial Least Squares, PLS) and Monte Carlo model based approaches for the analysis of fluorescence and diffuse reflectance spectra measured ex vivo from freshly excised breast tissues and for the diagnosis of breast cancer. Features extracted using both approaches, i.e. principal components (PCs) obtained from empirical analysis or tissue properties obtained from model based analysis, displayed statistically significant difference between malignant and non-malignant tissues, and can be used to discriminate breast malignancy with comparable sensitivity and specificity of up to 90%. The PC scores of a subset of PCs also displayed significant correlation with the tissue properties extracted from the model based analysis, suggesting both approaches likely probe the same sources of contrast in the tissue spectra that discriminate between malignant and non-malignant breast tissues but in different ways.

Diagnosis of breast cancer using fluorescence and diffuse reflectance spectroscopy: a Monte-Carlo-model-based approach

We explore the use of Monte-Carlo-model-based approaches for the analysis of fluorescence and diffuse reflectance spectra measured ex vivo from breast tissues. These models are used to extract the absorption, scattering, and fluorescence properties of malignant and nonmalignant tissues and to diagnose breast cancer based on these intrinsic tissue properties. Absorption and scattering properties, including beta-carotene concentration, total hemoglobin concentration, hemoglobin saturation, and the mean reduced scattering coefficient are derived from diffuse reflectance spectra using a previously developed Monte Carlo model of diffuse reflectance. A Monte Carlo model of fluorescence described in an earlier manuscript was employed to retrieve the intrinsic fluorescence spectra. The intrinsic fluorescence spectra were decomposed into several contributing components, which we attribute to endogenous fluorophores that may present in breast tissues including collagen, NADH, and retinol/vitamin A. The model-based approaches removes any dependency on the instrument and probe geometry. The relative fluorescence contributions of individual fluorescing components, as well as beta-carotene concentration, hemoglobin saturation, and the mean reduced scattering coefficient display statistically significant differences between malignant and adipose breast tissues. The hemoglobin saturation and the reduced scattering coefficient display statistically significant differences between malignant and fibrous/benign breast tissues. A linear support vector machine classification using (1) fluorescence properties alone, (2) absorption and scattering properties alone, and (3) the combination of all tissue properties achieves comparable classification accuracies of 81 to 84% in sensitivity and 75 to 89% in specificity for discriminating malignant from nonmalignant breast tissues, suggesting each set of tissue properties are diagnostically useful for the discrimination of breast malignancy.

Synchronous fluorescence spectroscopic characterization of DMBA-TPA-induced squamous cell carcinoma in mice

While initially confined to the epidermis, squamous cell carcinoma can eventually penetrate into the underlying tissue if not diagnosed early and treated. The noninvasive early detection of the carcinoma is important to achieve a complete treatment of the disease. Of the various non-invasive optical techniques, the synchronous fluorescence (SF) technique is considered to provide a simplified spectral profile with more sharp spectral signatures of the endogenous fluorophores in complex systems. The potential use of the SF technique in the characterization of the sequential tissue transformation in 7,12-dimethylbenz(a)anthracene-12-O-tetradecanoylphorbol-13-acetate (DMBA-TPA)-induced mouse skin tumor model in conjunction with simple statistical analysis is explored. The SF spectra show distinct differences during the earlier weeks of the tumor-induction period. Intensity ratio variables are calculated and used in three discriminant analyses. All the discriminant analyses show better classification results with accuracy greater than 80%. From the observed differences in the spectral characteristics and the ratio variables that resulted in better classification between groups, it is concluded that tryptophan, collagen, and NADH are the key fluorophores that undergo changes during tissue transformation process and hence they can be targeted as tumor markers to diagnose normal from abnormal tissues using the SF technique.

Use of a multiseparation fiber optic probe for the optical diagnosis of breast cancer

We explore the effects of the illumination and collection geometry on optical spectroscopic diagnosis of breast cancer. Fluorescence and diffuse reflectance spectroscopy in the UV-visible spectral range are made with a multiseparation probe at three illumination-collection separations of 735, 980, and 1225 microm, respectively, from 13 malignant and 34 nonmalignant breast tissues. Statistical analysis is carried out on two types of data inputs: (1) the fluorescence and diffuse reflectance spectra recorded at each of the three illumination-collection separations and (2) the integrated fluorescence (at each excitation wavelength) or diffuse reflectance over the entire spectrum at all three illumination-collection separations. The results show that using the integrated fluorescence intensities recorded at a single excitation wavelength at all three illumination-collection separations can discriminate malignant from nonmalignant breast tissues with similar classification accuracy to that using spectral data measured at several excitation wavelengths with a single illumination-collection separation. These findings have significant implications with respect to the design of an optical system for breast cancer diagnosis. Examining the intensity attenuation at a single wavelength rather than spectral intensities at multiple wavelengths can significantly reduce the measurement and data processing time in a clinical setting as well as the cost and complexity of the optical system.

Autofluorescence characterization for the early diagnosis of neoplastic changes in DMBA/TPA-induced mouse skin carcinogenesis

BACKGROUND AND OBJECTIVE

Squamous cell carcinoma (SCC), the second most common skin cancer, usually remains confined to the epidermis for some time but eventually penetrates the underlying tissues, if left untreated. The non-invasive early detection of the SCC is important for appropriate therapeutic strategies. In this study, we aim to characterize the tissue transformation in DMBA/TPA induced mouse skin tumor model using autofluorescence excitation emission matrix (EEM) in conjunction with a multivariate statistical method for early detection of the neoplastic changes.

STUDY DESIGN/MATERIALS AND METHODS

The fluorescence EEM from experimental group (n = 40; DMBA/TPA application), control group (n = 6; acetone application), and the blank group (n = 6; no application of DMBA/TPA or acetone) were measured every week using a spectrofluorometer coupled with a fiber optic bundle. The EEM was recorded at excitation wavelengths from 280 to 460 nm at 10 nm intervals and the fluorescence emission was scanned from 300 to 750 nm. The fluorescence emission characteristics corresponding to different fluorophores were extracted from the EEM and the spectral data were used in a multiple/linear discriminant statistical algorithm.

RESULTS

The changes in the fluorescence emission intensity were observed as early as the 1st week of tumor initiation by DMBA. Morphological changes as well as differences in the gross appearance of the skin surface were observed during the entire tumor initiation and promotion period of 15 weeks. The statistical analysis was performed for each excitation wavelength in the EEM and better classification accuracy was obtained for 280 and 410 nm excitations, corresponding to tryptophan and endogenous porphyrins, respectively. The statistical analysis of the combination wavelengths resulted in 11.6% increase in the overall classification accuracy when compared to the highest classification accuracy obtained with single wavelength analysis.

CONCLUSION

The intensity ratio mapping using the combination of emission intensities of key fluorophores such as tryptophan, collagen, NADH, and endogenous porphyrins from the measured EEM in conjunction with a simple multivariate statistical analysis can be used as a potential tool for the discrimination of early neoplastic changes with improved classification accuracy. Tryptophan and endogenous porphyrins may be used as biomarkers for the discrimination of early neoplastic changes when single wavelength excitations are used.

Three-dimensional Total Synchronous Luminescence Spectroscopy Criteria for Discrimination Between Normal and Malignant Breast Tissues

Specimens of malignant and normal female human breast tissues were analyzed after surgery by means of synchronous luminescence spectroscopy. Measurements were performed in the ranges of excitation wavelengths from 330 to 650 nm and synchronous wavelengths from 30 to 120 nm to obtain ordinary and first derivative three-dimensional total synchronous luminescence spectra (3d-TSLS) of each specimen. Arithmetic mean of these spectra has been calculated for normal and malignant specimens and analyzed to establish criteria for tissue differentiation. Spectral domain volumes (volumes below luminescence intensity surface) and mean spectral slopes have been calculated and also analyzed as tissue discrimination criteria. The obtained results are discussed in view of the possible relevance of synchronous luminescence spectroscopy in discrimination between normal and malignant breast tissue.