Functional nuclear imaging systems like PET or SPECT provide unique information that is used extensively in diagnosis, but it has also proven very useful for image-guided interventions. In the case of SPECT and radio-guided surgeries, 1D gamma detectors called gamma probes are routinely used during interventions to localize hotspots in conjunction with pre-operative SPECT images, or more recently, intra-operative SPECT images. As the tissue is being manipulated during surgery, these SPECT images quickly lose their validity, necessitating either new scans, which is in most cases unfeasible, or requiring the surgeon to do a mental update of the available imagery. In this paper, we present a novel 1D-3D registration procedure for functional nuclear imaging that registers tracked intra-operative 1D probe readings to a pre- or intra-operatively acquired 3D functional image. This procedure allows incorporating prior knowledge during radio-guided surgeries, enabling rapid updates of the visualization in the case of tissue deformation without the overhead of an additional complete scan. We show results using phantom data as well as patient data. © 2011 Springer-Verlag.
CITATION STYLE
Vetter, C., Lasser, T., Wendler, T., & Navab, N. (2011). 1D-3D registration for functional nuclear imaging. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6891 LNCS, pp. 227–234). https://doi.org/10.1007/978-3-642-23623-5_29
Mendeley helps you to discover research relevant for your work.