banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Fully adaptive FEM based fluorescence optical tomography from time-dependent measurements with area illumination and detection
Rent this article for


Image of FIG. 1.
FIG. 1.

Tissue phantom with fluorescent target.

Image of FIG. 2.
FIG. 2.

Instrumentation for acquiring frequency domain fluorescence measurements in a homodyne mode. Numbered components include: 1: neutral density filter (OD-3), 2: holographic band rejection filter, 3: interference bandpass filter, 4: linear polarizer, 5: image intensifier, and 6: linear polarizer. Optical filters can be moved in and out of the filter box assembly to acquire measurements at excitation and emission wavelengths. Phantom surface image on the CCD camera is expanded to depict measurement data acquisition by raster scanning the CCD pixels.

Image of FIG. 3.
FIG. 3.

Excitation source fluence: real (left image) and imaginary (right image) components.

Image of FIG. 4.
FIG. 4.

Forward mesh evolution on the illumination surface. Meshes after (a) 0, (b) 2, (c) 4, and (d) 5 adaptive refinements are depicted.

Image of FIG. 5.
FIG. 5.

Raster scanning of the CCD camera pixels is performed to extract the fluorescence measurements on the detection plane. Field of view of the camera system is . Pixels in the field of view are numbered and treated as individual detector locations.

Image of FIG. 6.
FIG. 6.

Experimentally observed and simulated real and imaginary components of fluorescence fluence at the measurement surface, plotted against CCD detector points for (a) and (b) target depths of , and (c) and (d) . In (e) and (f) we plot the RMSE defined in (10) for a sequence of adaptively refined meshes used for solution of coupled diffusion equations, for target depths of 1 and .

Image of FIG. 7.
FIG. 7.

Reconstructed and true images for 1 and deep fluorescent targets are presented. True targets are depicted by the black wireframe, while the reconstructed targets are represented by colored blocks. The top 10% of the contour levels of reconstructed fluorophore distribution are depicted. (a),(b) True and recovered targets ( plane view); (c),(d) true and recovered target ( plane view). Left column depicts the deep target case, while the right column represents the deep target.

Image of FIG. 8.
FIG. 8.

Parameter mesh adaptation for: deep fluorescent target [(a) (c) (e)], (b) deep target [(b) (d) (f)]. Top row [(a) (b)] depicts the solutions obtained on the initial coarse mesh; middle row [(c) (d)] depicts the solutions obtained on the mesh after one adaptive refinement; bottom row [(e) (f)] depicts the solutions obtained on the final (5th) adaptively refined mesh.

Image of FIG. 9.
FIG. 9.

Top row: Reduction of the objective function with Gauss-Newton iterations. Bottom row: Behavior of the norm of the residual as a function of iterations. Parts (a) and (c) correspond to the deep fluorescent target, parts (b) and (d) to the deep target. The lines are broken whenever mesh refinement occurs between two iterations.


Generic image for table

Summary of the reconstructed images for the 1 and deep fluorescent targets. “No. iterations” denotes the number of Gauss-Newton iterations; the final misfit between prediction and scaled observation ; and indicate the centroids of the true and recovered targets; is the number of parameter unknowns in the final parameter mesh; is the recovered fluorescence absorption coefficient in the targets.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Fully adaptive FEM based fluorescence optical tomography from time-dependent measurements with area illumination and detection