Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

High School Poster

HS Poster

Poster 46 - Correlating Hyperspectral Imaging with Mass Spectrometry of Prostate Tissue for Molecular Biomarker Discovery

Saturday, October 26, 2024
10:00 AM - 12:00 PM EST
Location: Exhibit Hall E, F & G

Introduction: Along with being used in fields such as remote sensing, agriculture, and forensic science, hyperspectral imaging (HSI) has become a powerful tool in noninvasive detection of diseases [1]. HSI increases the range of wavelengths used to capture images, and with these images, spectral details can be highlighted to help detect pathogens. Currently, HSI has been used to detect brain tumors, prostate cancers, and measure blood oxygenation [1]. However, HSI is limited in that molecular components of the tissue are still unknown. Mass spectrometry (MS) is a widespread technique that specializes in observing and identifying molecular composition of tissues [2]. This technique allows for the creation of profiles for different tissues that can determine different proteins, lipids, peptides and metabolites in a sample. With the functions of both techniques, a combination of the two can allow for a more comprehensive understanding of tissue pathology. Combining hyperspectral imaging with mass spectrometry could possibly allow for a more advanced methodology in disease detection that could be used in medicine. Hyperspectral imaging will be used for identification of diseased regions and mass spectrometry will be used for molecular discovery. A combination of information from both techniques could provide insight to a more profound understanding as to what molecular mechanisms, if any, could serve as a biomarker of disease. The objective of this study is to test whether registration can be used to correlate hyperspectral images with mass spectrometry data for early cancer detection.

Materials and

Methods: A fresh human prostate tissue after prostatectomy was obtained for this study. The prostate was then cut into thin slices (4-6 mm) and each section was scanned with hyperspectral imaging. Then each slice was sectioned into 51 small regions and the tissue from each region was sent for mass spectrometry analysis. Hyperspectral imaging and mass spectrometry data were then analyzed. Registration by software Fiji ImageJ was performed to match hyperspectral images between difference slices using a free-form deformable registration method [3]. With HSI, MS, and registered data, comparisons were then made. Specific regions of interest (ROI) were determined by MS data and hyperspectral curves were plotted for each ROI. MS data allowed for determination of which regions were possibly infected, and plotting hyperspectral curves for each ROI allowed us to determine whether abnormal patterns could be recognized from hyperspectral imaging. Registration allowed for a more accurate depiction of where specific ROIs were on the hyperspectral images.

Results, Conclusions, and Discussions: After capturing the hyperspectral images of the prostate and conducting analysis on the mass spectrometry data, Regions 3, 7, 11, 13, 19, 21, 23, 31, 33, 39, 43 were determined to be specific ROIs. ROI 31 was identified by MS to have a high concentration of prostate specific antigen (PSA). Plotting the hyperspectral curves of each region based on wavelength and intensity, ROI 31 (light orange) was determined to have the highest intensity at the highest wavelength (Figure 1). Additionally, Regions 39 (light yellow) and 43 (light green) that followed in highest intensity at the highest wavelength had similar protein composition to that of Region 31 as determined by MS data analysis.

With the correlation of mass spectrometry and hyperspectral imaging data, possible detection for early cancer may be feasible. MS can provide a more detailed insight to molecular composition that can assist hyperspectral imaging in detecting cancer and HSI can provide information in determining possible similarities in molecular composition between prostate regions. In the future, a larger number of experiments will be needed to further develop and support the preliminary findings of this study.

Acknowledgements (Optional):

References (Optional): [1] G. Lu, and B. Fei, “Medical hyperspectral imaging: a review,” J Biomed Opt, 19(1), 10901 (2014).
[2] E. R. Amstalden van Hove, D. F. Smith, and R. M. Heeren, “A concise review of mass spectrometry imaging,” J Chromatogr A, 1217(25), 3946-54 (2010).
[3] T. Eckhard, J. Eckhard, E. M. Valero et al., “Nonrigid registration with free-form deformation model of multilevel uniform cubic B-splines: application to image registration and distortion correction of spectral image cubes,” Appl Opt, 53(17), 3764-72 (2014).