Molecular Imaging with Individualized Priors

Vasilis Ntziachristos

Technical University of Munich, Germany, and Department of Electrical and Computer Engineering,
Northeastern University, Boston, MA, USA

Background and Goals

Prof. Ntizachristos is one of the world leaders in the field of tomographic fluorescence imaging. In his previous position at Mass General Hospital and in his present position as Director of the Institute for Biomedical Engineering at the Technical University of Munich his group builds novel imaging systems, devises and implements novel forward modeling and reconstruction algorithms, and invents and tests an ever expanding range of applications. He also holds an Adjunct Research Professor position and Northeastern University where he is the PI of the NIH grant listed above. He and Prof. Brooks with collaborators have had a productive collaborative relationship for the past five years.

Prof. Ntziachritos' current projects which are of direct relevance to this collaboration with CIBC include:

• Photo-acoustic molecular imaging: Photo-acoustic imaging is an emerging multi-modality imaging technology which offers the promise of higher resolution than can be achieved with either modality alone. The idea is to use focused laser light to excite a small region of tissue being interrogated. The consequent absorption of energy causes rapid local heating, and the mechanical expansion creates a focal source of an acoustic signal. This signal is observed by transducers and a tomographic reconstruction is then applied. Prof. Ntziachritos' approach is to combine this modality with the use of molecular probes to enhance contrast.
• Multi-modality fluorescence molecular tomography: Fluorescence molecular tomography (FMT) reconstructs the distribution of fluorophores in vivo, generally in small animals, by making multiple measurements with an array of optodes while illuminating from many angles and then using this coclletion of data to perform a tomographic reconstruction. (The publications cited above concern algorithms for improving such reconstructions.) FMT is an inherently low-resolution modality because of the scattering of light in tissue, even with the specificity introduced by careful use of the fluorescence phenomeon. An area of great current interest is the use of anatomical images such as from CT to build higher resolution prior models for FMT reconstructions. Prof. Ntziachritos has built combined CT-FMT instruments similar in concept to PET-CT or SPECT-CT devices. These techniques require the same kind of image-to-reconstruction-to-visualization pipeline as many of the translational collaborations described in this proposal.
• Catheter-based fluorescence and ultrasound imaging of plaque: In a project being carried out both in Munich and in Boston (Mass General Hospital and Northeastern), Prof. Ntziachritos is investigating development of a novel imaging device which would use fluorescent dyes molecularly targeted to specific cell types in inflammatory cardiovascular plaque to add molecular specificity to vessel imaging methods such as intravascular ultrasound (IVUS). Current efforts are focused on instrument development but we expect in the near term to begin to aquire both IVUS and optical measurements from the same probe. Since the optical probe has a narrow angular field of view and rotates while being "pulled back" inside the vessel during acquistion, this becomes a challenging multi-modality imaging problem, where the IVUS image can be used as a prior for reconstructing the more specific fluorescence data.

In all of these applications Prof. Ntziachritos' team will require computational steps that follow the proposed CIBC imaging pipeline: extraction of models from images, forward simulation to predict measurements (often using finite element methods), comparison of predicted to actual measuremnts to drive a solution to an inverse estimation problem, and adequate visualization all through the process.

Collaborator Publications

1. D. Hyde, E. Miller, D.H. Brooks, and V. Ntziachristos. "A statistical approach to inverting the born ratio". IEEE Trans Med Imag, 26(7):893–905, July 2007.
2. D. Hyde, E. Miller, D. Brooks, and V. Ntziachristos. "New techniques for data fusion in multimodal FMT-CT imaging". In Proc. Int. Symp. on Biomed Imaging (ISBI09), Paris, France, May 2008.
3. D. Hyde, R. De Kleine, S. MacLaurin, E. Miller, D. Brooks, T. Krucker, and V. Ntziachristos. "Hybrid FMTCT imaging of amyloid-beta plaques in a murine Alzheimer's disease model". NeuroImage, 44(4):1304– 1311, February 2009.
4. D. Hyde, R. Schulz, , E. Miller, D. Brooks, and V. Ntziachristos. "Performance dependence of hybrid X-ray CT - FMT on the optical forward problem". JOSA A, 24(4):919–923, February 2009.
5. D. Hyde, E. Miller, D. Brooks, and V. Ntziachristos. "Data specific spatially varying regularization for multimodal Fluorescence Molecular Tomography". IEEE Trans. Med. Imag., 2009.
6. D. Hyde, E. Miller, D. Brooks, and V. Ntziachristos. "Differential equation-driven regularization for joint FMT-CT imaging". In Proc. Int. Symp. on Biomed Imaging (ISBI09), Boston, MA, USA, 2009.
7. D. Hyde, E.L. Miller, D.H. Broks, and V. Ntziachristos. "A statistical method for inverting the born ratio". In Proceedings of the International Symposium on Biomedical Imaging (ISBI), Arlington, VA, USA, April 2006.
8. D. Hyde, R. DeKlein, S.A. MacLaurin, D. Brooks, E.L. Miller T. Krucker, , and V. Ntziachristos. "Hybrid FMT-CT method for in vivo imaging of amyloid plaques in a mouse model for alzheimers disease". In Society for Molecular Imaging Meeting, Providence, RI, USA, September 2007.
9. D.E. Hyde, R. Schulz, E. Miller, D.H. Brooks, and V. Ntziachristos. "Incorporation of structural apriori information in fluorescence molecular tomography". In OSA Biomedical Optics Topical Meeting, St. Petersburg, FL, USA, March 2008.