Hough space parametrization: ensuring global consistency in intensity-based registration.

Area of Science:

• Medical imaging
• Computer vision
• Image processing

Background:

• Intensity-based image registration is challenging with limited overlapping information.
• Lack of strong structures in overlap regions hinders global structural consistency.

Purpose of the Study:

• To develop a novel global regularization method for image registration.
• To leverage information beyond overlapping regions for improved registration accuracy.

Main Methods:

• Utilized Generalized Hough Transform for a novel global regularization term.
• Learned a parametrization of target anatomy in Hough space using prior data.
• Applied this parametrization as regularization for registering partial images without prior data.

Main Results:

• Demonstrated good performance on synthetic and real medical images.
• The proposed method effectively addresses registration challenges with insufficient overlap information.
• Ensured global consistency even with limited local information.

Conclusions:

• The proposed Generalized Hough Transform-based regularization enhances image registration.
• This method is valuable for medical image analysis, especially with partial or information-limited overlaps.
• The approach shows potential for improving the accuracy and reliability of image registration techniques.