Visual Quality Enhancement in Optoacoustic Tomography Using Active Contour Segmentation Priors

Area of Science:

• Biomedical Imaging
• Medical Image Analysis
• Optoacoustic Tomography

Background:

• Biomedical image segmentation is crucial for anatomical characterization and pathology detection.
• Segmentation enhances tomographic reconstruction by addressing heterogeneities in excitation fields and tissue properties.
• Optoacoustic tomography (OPT) imaging performance can degrade due to optical and acoustic property discontinuities if not properly segmented.

Purpose of the Study:

• To analyze the efficacy of active contour models for boundary segmentation in cross-sectional optoacoustic tomography.
• To improve optoacoustic image reconstruction accuracy by employing segmentation-derived tissue models.
• To validate the proposed segmentation and modeling approach in phantom and small animal studies.

Main Methods:

• Active contour models were utilized for boundary segmentation of cross-sectional optoacoustic images.
• A two-compartment model for acoustic and optical tissue parameters was constructed from segmented masks.
• The model was integrated into image reconstruction routines to enhance accuracy.

Main Results:

• The active contour model approach enabled efficient segmentation of anatomical structures in optoacoustic images.
• The derived two-compartment model successfully accounted for tissue property variations.
• Improved accuracy in image reconstruction was demonstrated in both tissue-mimicking phantoms and small animal experiments.

Conclusions:

• Active contour models provide a robust method for segmenting optoacoustic images, even with low intrinsic contrast.
• Segmentation-based modeling significantly enhances the accuracy of optoacoustic tomography image reconstruction.
• The validated approach holds promise for improved diagnostic capabilities in optoacoustic imaging.