Abstract
Purpose:
Currently applied calibration approaches lead to satisfying 3D roadmapping overlay and 3D reconstruction accuracies; however, the required calibration times are extensive. The aim of this paper is the introduction of a novel model-based approach for geometric system calibrations, leading to a significant reduction of calibration times.
Methods:
By using physical insight into the system, a physical model is derived which can be exploited to predict geometric calibrations parameters. Model-parameters are estimated using a limited set of phantom-based measurement data. Effectively, the calibration procedure is recast to a parameter identification experiment.
Results:
The potential of the proposed approach is illustrated by virtue of a benchmark object, successful reconstruction of a clinical phantom, and comparison to phantom-based accuracies.
Conclusions:
Accurate models are required to achieve the desired accuracies. Based on the results in this work, the approach seems to be feasible for practical applications; however, to achieve all the desired specifications, future research should focus on enhanced modeling techniques.
Currently applied calibration approaches lead to satisfying 3D roadmapping overlay and 3D reconstruction accuracies; however, the required calibration times are extensive. The aim of this paper is the introduction of a novel model-based approach for geometric system calibrations, leading to a significant reduction of calibration times.
Methods:
By using physical insight into the system, a physical model is derived which can be exploited to predict geometric calibrations parameters. Model-parameters are estimated using a limited set of phantom-based measurement data. Effectively, the calibration procedure is recast to a parameter identification experiment.
Results:
The potential of the proposed approach is illustrated by virtue of a benchmark object, successful reconstruction of a clinical phantom, and comparison to phantom-based accuracies.
Conclusions:
Accurate models are required to achieve the desired accuracies. Based on the results in this work, the approach seems to be feasible for practical applications; however, to achieve all the desired specifications, future research should focus on enhanced modeling techniques.
Original language | English |
---|---|
Pages (from-to) | 6170-6182 |
Number of pages | 13 |
Journal | Medical Physics |
Volume | 42 |
Issue number | 11 |
DOIs | |
Publication status | Published - 7 Oct 2015 |