Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
Journal Home About Issues in Progress Current Issue All Issues Feature Issues

Three-dimensional reconstructions from low-count SPECT data using deformable models

G.S. Cunningham, K.M. Hanson, and X.L. Battle

Open Access Open Access

Abstract

We demonstrate the reconstruction of a 3D, time-varying bolus of radiotracer from first-pass data obtained by the dynamic SPECT imager, FASTSPECT, built by the University of Arizona. The object imaged is a CardioWest Total Artificial Heart. The bolus is entirely contained in one ventricle and its associated inlet and outlet tubes. The model for the radiotracer distribution is a time-varying closed surface parameterized by 162 vertices that are connected to make 960 triangles, with uniform intensity of radiotracer inside. The total curvature of the surface is minimized through the use of a weighted prior in the Bayesian framework. MAP estimates for the vertices, interior intensity and background count level are produced for diastolic and systolic frames, the only two frames analyzed. The strength of the prior is determined by finding the corner of the L-curve. The results indicate that qualitatively pleasing results are possible even with as few as 1780 counts per time frame (total after summing over all 24 detectors). Quantitative estimates of ejection fraction and wall motion should be possible if certain restrictions in the model are removed, e.g., the spatial homogeneity of the radiotracer intensity within the volume defined by the triangulated surface, and smoothness of the surface at the tube/ventricle join.

©1998 Optical Society of America

Full Article  |  PDF Article
More Like This
4D embryonic cardiography using gated optical coherence tomography

M. W. Jenkins, F. Rothenberg, D. Roy, V. P. Nikolski, Z. Hu, M. Watanabe, D. L. Wilson, I. R. Efimov, and A. M. Rollins
Opt. Express 14(2) 736-748 (2006)

Decomposition of images and objects into measurement and null components

D.W. Wilson and H.H. Barrett
Opt. Express 2(6) 254-260 (1998)

Experimental Cerenkov luminescence tomography of the mouse model with SPECT imaging validation

Zhenhua Hu, Jimin Liang, Weidong Yang, Weiwei Fan, Congye Li, Xiaowei Ma, Xueli Chen, Xiaopeng Ma, Xiangsi Li, Xiaochao Qu, Jing Wang, Feng Cao, and Jie Tian
Opt. Express 18(24) 24441-24450 (2010)

View More...

Supplementary Material (1)

Media 1: MOV (274 KB)     

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)
Fig. 1.
Fig. 1. Raw data for a) diastolic frame and c) systolic frame. Predicted detector Poisson rates for b) diastolic frame and d) systolic frame using MAP solution with α=0.2.

Download Full Size | PDF

Fig. 2.
Fig. 2. BIE canvas used to analyze SPECT data. The triangulated surface box at the far left contains x. It is transformed into a voxellated grid, f, and then by the matrix, H, to produce Hf(x). Finally it is multplied by the intensity, I, and the additive background constant, s, is added, to produce the predicted detector pixel rates g=IHf(x)+s.

Download Full Size | PDF

Fig. 3.
Fig. 3. The L-curve for the diastolic frame.

Download Full Size | PDF

Fig. 4.
Fig. 4. MAP reconstructions of the bolus boundary surface using a) α=3.2, b) α=0.2, and c) α=0.1.

Download Full Size | PDF

Fig. 5.
Fig. 5. Comparison between diastolic and systolic frame reconstructions: a) diastolic frame is wireframe and systolic frame is solid surface, b) diastolic frame is solid surface and systolic frame is wireframe. See http://planck.lanl.gov/~cunning/3D for an interactive Java display of the reconstructions. [Media 1]

Download Full Size | PDF

Fig. 6.
Fig. 6. Cut planes through the reconstructions in Fig. 5. Red lines are for diastolic frame and green lines are for systolic frame. Z-slices are a) -8 mm, b) -4 mm, c) 0 mm, d) 4 mm, e) 8 mm, and f) 12 mm.

Download Full Size | PDF

Equations (5)

Equations on this page are rendered with MathJax. Learn more.

π ( x ) = ∑ i A i ( ∑ j [ tan ( θ ij 2 ) l ij ] 2 ) ,
ϕ x I s = − ln Prob [ data ∣ predicted data ]
= ∑ i [ − k i ln g i + g i ] ,
x MAP ( α ) = arg min x [ ϕ x I s + α π ( x ) ] ,
( ϕ ( x MAP ( α ) ) , π ( x MAP ( α ) ) ) ,
Select as filters
Select Topics Cancel
Top
       
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.
Privacy | Terms of Use