Abstract
In optoacoustic tomograophy, detector geometry plays an important
role in determining the image quality. When image reconstruction is
performed without accounting for the effect of detector geometry,
the resolution of the image may be degraded. This is often the case
when flat acoustic detectors with relatively large size are used. When
reconstruction is performed without accounting for the geometry of these
detectors, the resolution in the tangential direction is degraded.
In this work we developed a model-based reconstruction method which
accounts for detector geometry. The model is composed of two steps:
First, the relation between the detected signals and the optoacoustic
image is discretized under the assumption of point detectors. Then, the
effect of the detector is taken into account by temporally convolving its
response with the signals obtained in the first step. The results are saved
in a matrix which is subsequently inverted to obtain the optoacoustic
image. Regularization is applied in the matrix inversion to enable stable
reconstruction.
The method is demonstrated in an ex vivo experiment, where the brain of
a mouse was imaged using cylindrically focused transducers. Significant
resolution enhancement is obtained using the new technique as
compared to the reconstruction obtained by model-based inversion that
does not account for the detector geometry. In the enhanced images, the
contrast of some anatomical features was increased by over a factor of 3.
role in determining the image quality. When image reconstruction is
performed without accounting for the effect of detector geometry,
the resolution of the image may be degraded. This is often the case
when flat acoustic detectors with relatively large size are used. When
reconstruction is performed without accounting for the geometry of these
detectors, the resolution in the tangential direction is degraded.
In this work we developed a model-based reconstruction method which
accounts for detector geometry. The model is composed of two steps:
First, the relation between the detected signals and the optoacoustic
image is discretized under the assumption of point detectors. Then, the
effect of the detector is taken into account by temporally convolving its
response with the signals obtained in the first step. The results are saved
in a matrix which is subsequently inverted to obtain the optoacoustic
image. Regularization is applied in the matrix inversion to enable stable
reconstruction.
The method is demonstrated in an ex vivo experiment, where the brain of
a mouse was imaged using cylindrically focused transducers. Significant
resolution enhancement is obtained using the new technique as
compared to the reconstruction obtained by model-based inversion that
does not account for the detector geometry. In the enhanced images, the
contrast of some anatomical features was increased by over a factor of 3.
| Original language | American English |
|---|---|
| Pages | 8223-104 |
| State | Published - 2012 |
| Externally published | Yes |
| Event | SPIE Photonics West 2012 - San Francisco, United States Duration: 21 Jan 2012 → 26 Jan 2012 https://spie.org/conferences-and-exhibitions/past-conferences-and-exhibitions/photonics-west-2012?SSO=1 |
Conference
| Conference | SPIE Photonics West 2012 |
|---|---|
| Country/Territory | United States |
| City | San Francisco |
| Period | 21/01/12 → 26/01/12 |
| Internet address |