Registration of photoacoustic tomography vascular images: Comparison and analysis of automatic registration approaches

Abstract

Image registration is crucial in the clinical application of photoacoustic tomography (PAT) for vascular growth monitoring. Aiming to find an optimized registration scheme for PAT vascular images acquired at different times and with varying imaging conditions, we compared and analyzed different commonly used intensity-based and feature-based automatic registration schemes. To further improve the registration performance, we proposed a new scheme that combines phase correlation with these commonly used intensity-based registration methods and compared their performances. The objective evaluation measures: peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM), root mean square error (RMSE), and quantitative visual perception (jump percentage P), as well as subjective evaluation using mean opinion score (MOS), were combined to evaluate the registration performance. Results show that the feature-based approaches in this study were not suitable for PAT image registration. And by adding phase correlation as rough registration, the overall registration performance was improved significantly. Among these methods, the proposed scheme of phase correlation combined with mean square error (MSE) similarity measure and regular-step-gradient-descent optimizer provides the best visual effect, accuracy, and efficiency in PAT vascular image registration.

Automated three-dimensional image registration for longitudinal photoacoustic imaging

“…6,[10][11][12][13] To the best of our knowledge, only two research studies deal with the problem of unimodal registration of serial PA images. 14,15 A parametric approach using an intensity-based scheme and gradient descent optimizer was proposed to co-register PAT images of hands. 14 However, one major limitation of parametric registration algorithms is the requirement to choose the anticipated deformation field parametrization (examples are rigid, affine, or B-splines).…”

“…14,15 A parametric approach using an intensity-based scheme and gradient descent optimizer was proposed to co-register PAT images of hands. 14 However, one major limitation of parametric registration algorithms is the requirement to choose the anticipated deformation field parametrization (examples are rigid, affine, or B-splines). This can be problematic in the case of complex and unknown deformations, i.e., the breast repositioning within the recording aperture.…”

“…Several image registration algorithms have been proposed for PA imaging but most of them aimed at registering PA images with conventional imaging modalities such as MRI 6 , 10 – 13 To the best of our knowledge, only two research studies deal with the problem of unimodal registration of serial PA images 14 , 15 . A parametric approach using an intensity-based scheme and gradient descent optimizer was proposed to co-register PAT images of hands 14 .…”

Unsupervised deep learning enables real-time image registration of fast-scanning optical-resolution photoacoustic microscopy

“…The scale-invariant feature transform (SIFT) and speeded-up robust features (SURF) methods are commonly used to extract coordinates of feature points from image pairs [20] , [21] . Then the deformed images can be corrected based on the coordinates.…”