Biomedical Engineering Reference
In-Depth Information
5.5.1.1 Conventional 2D Ultrasound Scanning ....................... 103
5.5.1.1.1 Refraction ....................................................... 103
5.5.1.1.2 Depth Errors .................................................. 103
5.5.1.1.3 Spatial Resolution ......................................... 103
5.5.1.2 Position Measurement.................................................... 105
5.5.1.2.1 Sensor Design-Independent Errors ............ 105
5.5.1.2.1.1 Latency ...................................... 105
5.5.1.2.1.2 Uncertainty in Position
Sensors....................................... 106
5.5.1.2.1.3 Calibration of Sensor............... 107
5.5.1.2.2 Sensor Design-Specific Errors ..................... 107
5.5.1.2.2.1 Electromagnetic Position
Sensors....................................... 107
5.5.1.2.2.2 Optical Position Sensors ......... 107
5.5.1.2.2.3 Acoustic Position Sensors....... 108
5.5.1.2.2.4 Mechanical Arms ..................... 108
5.5.1.3 3D Acquisition Process................................................... 108
5.5.1.3.1 Transducer Motion ....................................... 108
5.5.1.3.2 Tissue Motion ................................................ 109
5.5.1.3.3 Nonuniform Sampling ................................. 109
5.5.1.3.3.1 Fixed Geometry........................ 109
5.5.1.3.3.2 Freehand Geometry................. 109
5.6 General Conclusions.................................................................................. 110
References ........................................................................................................... 111
5.1
Introduction
In this chapter we review image artifacts due to scanner errors relevant to
image registration, and describe methods available for correction. Our dis-
cussion is restricted to the 3D imaging modalities: computed tomography
(CT), magnetic resonance imaging (MRI), emission tomography (ET)
(positron emission tomography [PET] and single-photon emission computed
tomography [SPECT]), and ultrasound (US).
Three types of applications of medical imaging are particularly reliant on
accurate images: image-guided surgery, multimodality image registration, and
quantitative imaging. Generally, these are uses of images requiring calculation
and application of mathematical mapping between image space and physical
space or physical dimensions. Depending on the specific application, the reg-
istration mapping may be limited to determination of pixel dimensions or
encompass position, orientation, and dimensional scaling of the image relative
to real space. Applications in image-guided surgery are an example of the lat-
ter. Even if image registration can sometimes be limited to the relationship
