Raman Studies And Detections Of The Breast Disease

Breast biopsy is usually recommended on most patients diagnosed with a breastirregularity such as a lump or an abnormal finding seen on a mammogram or a breastultrasound. These screening methods have some limitations. As a result patients oftencomplain of having undergone an unnecessary surgical intervention. For this, weconstructed a mathematical model using Raman spectra results obtained from normaland malignant tissues and based on various biochemical characteristics such ascalcification, protein structure and fat structure. This algorithm allowed to identifyearly pathological changes. Most importantly it allowed to distinguish the nature ofthe breast tissue (normal, benign and malignant) and explore the inner link betweenbreast cancer and premalignant lesions. In the study, the main purpose of using thefresh organization and Raman spectrometer to construct the model is to prepare toidentify the property of the breast lesions by nonsurgical methods and metastasis ofaxillary lymphy nodes during the operation for the next study.1. Ex vivo diagnosis and analysis of breast disease by using portable RamanspectroscopyIn this study, a portable Raman spectroscopy was used to identify breast tissuesby comparing the mean Raman spectra of normal, malignant, and benign tissues. Forthis, we constructed a mathematical model using Raman spectra results obtained fromnormal and malignant tissues based on their various biochemical characteristics, suchas, protein and fat. The application of Raman spectroscopy for breast pathologicaldiagnosis can be used for reducing unnecessary surgical biopsy, significantly savingmedical resources.2. Raman Spectra Exploring Breast Tissues: Comparison of PrincipalComponent Analysis and Support Vector Machine-Recursive Feature EliminationRaman spectroscopy is one of the advanced methodologies that could distinguishthe characteristics of normal, malignant and benign breast biopsy samples. However,the Raman signal of breast tissue is often too weak and complicated to provideobvious spectral information for robust, accurate and rapid diagnosis. Comparison ofprincipal component analysis (PCA) is one of most common analysis methods which use to simplify the Raman spectral data to improve malignancy detection sensitivity.In this paper, Raman data analysis of breast tissue using support vectormachine-recursive feature elimination (SVM-RFE) method was applied and its resultswere compared with PCA. Mahalanobis distance and spectral residuals were used asdiscriminating criteria for evaluating these two methods. The comparison shows thatSVM-RFE based on the selection of characteristic peaks better reflects the nature ofbiopsy and it produces better discrimination, accuracy and sensitivity for normal,malignant and benign breast tissues. The Raman spectroscopy combined withSVM-RFE opens great future in the clinical applications of mammary diseasediagnosis.3. Identification of microcalcifications in breast cancer using shell-isolatednanoparticle-enhanced Raman spectroscopy (SINERS)Different breast diseases present different chemical compositions ofmicrocalcifications. Raman spectroscopy has the advantage of identifying componentsof samples. But the label-free Raman spectra of biotissues usually have the strongfluorescence from bio samples. For the aim to obtain a high-quality Raman spectrumof biotissues, we mixed the microcalcifications with the shell-isolated nanoparticles,which is the Raman signal amplification substrate with a composition of a goldnanoparticle core and an ultrathin silica or alumina shell. We used the shell-isolatednanoparticles for the SERS study of microcalcifications of breast tissue. They caneffectively increase the SERS signal of microcalcifications compared with thespectrum without SHINERS.4. The study of breast lesions using shell-isolated nanoparticle-enhancedRaman spectroscopy (SHINERS)Section the frozen breast tissue using serial sectioning technique, make clear thelesions type and precision lesion locations by HE Staining Technique, In this study,Raman spectroscopy was used to identify the different type of breast slices (normal,fibroadenoma, precancerous lesion, DCIS, IDC) with and without shell-isolatednanoparticles, Analysed the different chemical compositions among the different typeof breast lesions and the action of shell-isolated nanoparticles.