Advisor(s)

Max Diem

Contributor(s)

David E. (David Edward) Budil, John R. Engen, Mary Jo Ondrechen, Nora V. Laver

Date of Award

4-2011

Date Accepted

4-25-2011

Degree Grantor

Northeastern University

Degree Level

Ph.D.

Degree Name

Doctor of Philosophy

Department or Academic Unit

College of Science. Department of Chemistry and Chemical Biology.

Keywords

cervical cancer, cytopathology, human papillomavirus, infrared micro-spectroscopy, principal component analysis, spectral cytopathology

Subject Categories

Cancer - Spectroscopic imaging, Cancer - Cytopathology Infrared spectroscopy, Cervix uteri - Cancer

Disciplines

Biochemistry | Cancer Biology | Medicinal-Pharmaceutical Chemistry

Abstract

Spectral Cytopathology (SCP) is a novel application of Fourier transform infrared micro-spectroscopy (IR-MSP) for use as a diagnostic tool for the objective and reproducible screening for cervical cancers. SCP couples IR-MSP with unsupervised multivariate statistical methods of analysis, and is capable of detecting spectral patterns related to intrinsic molecular changes within individual exfoliated cells that are unique to maturation state, hormonal influence, viral infection, and disease. Spectral differences within individual cells may reveal biochemical changes that may be precursors to cancer before they are reflected in the morphology of the cell.

Clinical cervical samples were obtained in collaboration with the Department of Pathology at Tufts Medical Center. Infrared spectral maps of 16 mm2 were collected from cells deposited onto IR microscope slides. Spectroscopic data was processed using the PapMap algorithm, which results in one IR spectrum for each cell that describes its discrete biochemical composition. Spectral data is pre-processed analyzed by unsupervised multivariate methods, specifically principal component analysis (PCA).

Spectral patterns related to squamous differentiation, menopausal status, and hormonal influence were identified in cervical cells by SCP. Cervical samples that had a cytological diagnosis of normal or low grade dysplasia, were successfully correlated by SCP with their clinical diagnosis. The morphologically abnormal cells and morphologically normal cells from two confirmed low grade dysplasia samples had similar spectral patterns and were not differentiated by SCP. SCP was also correlated against human papillomavirus (HPV) DNA analysis in a study which included 49 clinical samples. This study reports a sensitivity of 88% for detecting infections with high-risk strains of HPV. These results suggest that SCP could provide a potent adjunct diagnostic tool for cytopathologists.

Document Type

Dissertation

Rights Holder

Jennifer M. Schubert


View Full Text

Click button above to open, or right-click to save.

Share

COinS