Fellow Detail

I work at the centre of a collaboration between biophysicists, medical doctors and statisticians working together to translate mid infrared spectroscopy as a real time diagnostic clinical tool for detecting the pre-cancerous stages of oesophageal adenocarcinoma (cancer of the gullet).

This highly interdisciplinary project involves direct contact with patients and physicians at the hospital. Currently I collect fresh biopsies taken for diagnostic purposes by the doctors at University College London Hospital (UCLH), record their spectra using Attenuated Total Reflectance - Fourier Transform Infrared (ATR-FTIR) spectroscopy; a non destructive technique allowing each biopsy to be staged in the classical histological manner subsequently for cross validation. This technique provides a biochemical snapshot of the sample, by plotting absorbance against wavelength, each pre-cancerous stage is shown to exhibit a unique spectrum.

In order to build the diagnostic model from the spectral biopsies I have integrated two biophysical techniques, ATR-FTIR spectroscopy and FTIR microspectroscopy to get the maximum chemical information from a combination of fresh and sectioned/stained biopsies. Statistical techniques such as hierarchical clustering, principle component analysis, partial least squares discrimination are then applied to build the diagnostic model based on the histological classification of the pre-cancerous stage