*2.2.2. Linear mechanical 3D scanners*

and the total scan angle needed to cover the desired anatomy. Typically, these parameters can be adjusted to optimize scanning time, image quality and the size of the volume imaged [11-16]. The most common integrated 3D transducers using the wobbling technique are used

The 3D image resolution will not be isotropic. The resolution in the 3D US image will degrade in the axial direction away from the transducer due to the increasing US beam spread in the lateral and elevational directions of the acquired 2D US images. Since the acquired 2D images used to generate a 3D image are arranged as a fan, the distance between the acquired US images increases with increasing axial distance. Increasing axial distances result in decreasing spatial sampling resulting in further loss of spatial resolution in the elevational direction of the

(b) 

(d) 

**Figure 1.** Schematic diagrams of 3D US mechanical scanning methods. (a) A side-firing TRUS transducer is mechanical‐ ly rotated and the acquired images have equal angular spacing. The same approach is used in a mechanically-wob‐ bled transducer. (b) A rotational scanning mechanism using an end-firing transducer, typically used in 3D TRUS guided prostate biopsy. The acquired images have equal angular spacing. (c) A linear mechanical scanning mechanism, in which the acquired images have equal spacing. (d) The mechanically tilting mechanism, but integrated into a 3D US

transducer. The US transducer is "wobbled" inside the housing of the transducer.

for abdominal and obstetrical imaging [17-19].

4 Advancements and Breakthroughs in Ultrasound Imaging

Figure 1 

(a) 

(c) 

acquired 2D US images of the reconstructed 3D image [20].

Linear scanners mechanisms use an external motorized fixture to move the conventional 2D transducer across the skin of the patient. The 2D transducer can be fixed to be perpendicular to the surface of the skin or at an angle for acquiring Doppler images. The spacing between the acquired 2D images is adjustable but constant during the scan so that the acquired 2D images are parallel and uniformly spaced (see Fig. 1c). The velocity of the transducer as it is being scanned is adjusted to obtain 2D images with an appropriate spatial interval for generating high quality 3D images [10].

The predefined spacing between the acquired 2D US images allows 3D images to be recon‐ structed while the 2D US images are being acquired. In the direction parallel to the acquired 2D US images the resolution of the reconstructed 3D US image will be the same as the original 2D US images. However, in the direction of the 3D scanning, the resolution of the reconstructed 3D image will be equal (if spatial sampling is appropriate) to the elevational resolution of the acquired 2D US images. Thus, the resolution of the 3D US image will be poorest in the 3D scanning direction due to greater spread of the US beam in the elevational direction [21].

This scanning approach is not typically used in interventional applications; however, it has been successfully implemented in many vascular B-mode and Doppler imaging applications, particularly of for carotid arteries [11, 22-30] and tumor vascularization [25, 31-33].

#### *2.2.3. Endo-cavity rotational 3D scanners*

The endo-cavity rotational 3D scanning approach has been used extensively in 3D US-guided prostate interventional procedures. In this approach an external fixture or internal mechanism is used to rotate an endo-cavity transducer (*e.g*., a transrectal ultrasound (TRUS) probe, see Fig. 1b) about its long axis. Endo-cavity transducers using an end-firing approach are typically used for prostate biopsy. When these types of conventional transducers are rotated by the motorized fixture, the set of acquired 2D images will be arranged as a fan (Fig. 1b), intersecting in the center of the 3D US image, resulting in an image as shown in Fig. 2. To obtain a 3D image of the prostate as in Fig. 2, an end-firing transducer is typically rotated by 180° [16].

Endo-cavity transducers using a side-firing 1D array are typically used in prostate brachy‐ therapy, cryotherapy and focal therapy. When using these types of conventional transducers, the acquired images will also be arranged as a fan, but intersect at the axis of rotation of the transducer (see Fig. 1a). The side-firing transducer is typically rotated from 80° to 110° to obtain a 3D TRUS image of the prostate [16, 34, 35]. Figure 2 shows that endo-cavity scanning transducer used to image the prostate for 3D US-guided therapy [6, 9, 11, 25, 34, 36-39]

For scanning systems used for 3D US-guided prostate biopsy, the end-firing transducer is rotated by at least 180° about a fixed axis that perpendicularly bisects the transducer array. In this approach, the resolution of the 3D image will not be isotropic. Since the spatial sampling is highest near the rotation axis of the transducer and the poorest away from the axis of rotation of the transducer, thus the resolution of the 3D US image will degrade as the distance from the rotational axis of the transducer is increased. In addition, the axial and elevational resolution will decrease as the distance from the transducer is increased, as discussed above. The combination of these effects will result in a 3D US image resolution that is best near the transducer and the rotational axis, while being poorest away from the transducer and rota‐ tional axis.

the sensing technology: Ascension – Bird sensor [3] Polhemus – Fastrack sensor [41] and

3D Ultrasound Imaging in Image-Guided Intervention

http://dx.doi.org/10.5772/55230

7

The most successful free-hand 3D US scanning approach used in interventional procedures makes use of magnetic field sensors, as well as applications such as echocardiography, obstetrics, and vascular imaging [3, 4, 41-51]. To track the transducer during generation of a 3D US image, a small receiver is mounted on the transducer containing three orthogonal coils allowing six-degrees-of-freedom sensing. The small receiver mounted on the transducer measures the strength of the magnetic field in three orthogonal directions, which is generated by a time-varying 3D magnetic field transmitter placed near the patient. The position and orientation of the transducer is calculated by continuously measuring the strength of the three

Since magnetic field sensors are small and unobtrusive devices, they allow the transducer to be tracked without the need for bulky mechanical devices, and without the need to keep a clear line of sight as required by optical tracking methods. Since magnetic field sensors are sensitive to electromagnetic interference or ferrous (or highly conductive) metals located nearby, geometric tracking errors can occur leading to distortions in the 3D US image. Thus, metal beds used in procedures, or surgical rooms can cause significant distortions. However, modern magnetic field sensors have been produced to be less susceptible to these sources of error, particularly ones that use a magnetic transmitter placed between the bed and the patient.

Hepatocellular carcinoma (HCC) is the fifth most common diagnosed malignancy and the third most frequent cause of cancer related deaths worldwide [52]. Incidence is particularly high in Asia and sub-Saharan Africa due to the large incidence of hepatitis B and C, both of which are complicated by hepatic cirrhosis, which is the greatest risk factor for HCC. Recently, increasing trends in HCC have been reported from several Western countries [53]. Further‐ more, the liver is the second most common site of metastatic cancer arising in other organs. When feasible, surgical resection or liver transplant is the accepted standard therapeutic approach, and currently has the highest success rate of all treatment methods for primary and metastatic liver cancer. Unfortunately, only 15% of patients are candidates for surgery [54, 55]. Patients who do not qualify for surgery usually are offered other therapeutic solutions such as chemotherapy and radiotherapy, but unfortunately have variable limited success rates. Minimally invasive percutaneous techniques, such as radio-frequency (RF) and microwave (MW) ablation of malignant tissue in the liver is a rapidly expanding research field and treat‐ ment tool for those patients who are not candidates for surgical resection or transplant. In some cases this acts as a bridge to liver transplantation [54, 56]. Due to low complications rates and shorter recovery times, the indications for these minimally invasive procedures are constantly increasing. However, these methods have a higher local recurrence rate than surgical resec‐

tion, mostly due to insufficient or inaccurate local ablation of cancerous cells [56, 57].

Northern Digital – Aurora sensor [4].

components of the local magnetic field.

**3.1. Clinical problem**

**3. 3D Ultrasound-guided focal liver ablation**

3D rotational scanning with an end-firing transducer is most sensitive to the motion of the transducer and patient since the axis of rotation is in the center of the 3D US image. Any motion during the 3D scan will cause a mismatch in the acquired 2D US images, resulting in artifacts in the center of the 3D US image. Artifacts in the center of the 3D US image will also occur if the axis of rotation is not accurately known; however, proper calibrations can remove this source of potential error. Thus, for interventional applications such as 3D US-guided prostate biopsy or brachytherapy, the rotational scanning mechanism is typically supported by a stabilization apparatus [16, 34, 40].

**Figure 2.** The 3D US of the prostate displayed using the multi-planar reformatting approach: (a) An end-firing TRUS prostate cube-view 3D image, allowing the sides to be translated and angles to reveal the desired anatomy. (b) A 3D TRUS image acquired using a side-firing transducer using the mechanical rotation approach.

#### *2.2.4. Free-hand scanning with position sensing*

Some 3D US-guided interventional procedures are making use of 3D scanning techniques that do not require a mechanical scanning device. In this approach, the user holds and manipulates a conventional US transducer to cover the patient's anatomy being investigated. Since construction of a 3D US image requires that the position and orientation of the conventional transducer be known, free-hand scanning requires a method to track the positions and orientations of the transducer as it is being moved. All methods to accomplish this task require a sensor to be mounted on the transducer to allow measurement of the conventional 2D transducer's position and orientation as it is moved over the body.

Over the past 2 decades, several approaches for free-hand scanning have been developed: tracked 3D US with articulated arms, free-hand 3D US with acoustic sensing, free-hand 3D US with magnetic field sensing, and image-based sensing (speckle decorrelation). The method used most commonly is the magnetic field sensing approach with several companies providing the sensing technology: Ascension – Bird sensor [3] Polhemus – Fastrack sensor [41] and Northern Digital – Aurora sensor [4].

The most successful free-hand 3D US scanning approach used in interventional procedures makes use of magnetic field sensors, as well as applications such as echocardiography, obstetrics, and vascular imaging [3, 4, 41-51]. To track the transducer during generation of a 3D US image, a small receiver is mounted on the transducer containing three orthogonal coils allowing six-degrees-of-freedom sensing. The small receiver mounted on the transducer measures the strength of the magnetic field in three orthogonal directions, which is generated by a time-varying 3D magnetic field transmitter placed near the patient. The position and orientation of the transducer is calculated by continuously measuring the strength of the three components of the local magnetic field.

Since magnetic field sensors are small and unobtrusive devices, they allow the transducer to be tracked without the need for bulky mechanical devices, and without the need to keep a clear line of sight as required by optical tracking methods. Since magnetic field sensors are sensitive to electromagnetic interference or ferrous (or highly conductive) metals located nearby, geometric tracking errors can occur leading to distortions in the 3D US image. Thus, metal beds used in procedures, or surgical rooms can cause significant distortions. However, modern magnetic field sensors have been produced to be less susceptible to these sources of error, particularly ones that use a magnetic transmitter placed between the bed and the patient.
