*4.2.1.2 Ultrasound*

Cheung et al. demonstrated the use of a radiation-free three-dimensional ultrasound system for the assessment of spinal curvature in 29 scoliosis patients [6]. Similarly, Kowalski et al. used an ultrasound-based volume projection imaging method to compare the lumbar lordosis and thoracic kyphosis angle in patients with scoliosis as well as normal subjects or other people with spinal disorders [50]. In this volume projection imaging method, the 3D representation of the spinal anatomy was generated using the ultrasound images together with the corresponding 3D spatial information (see **Figure 5**). The structure of the spine anatomy was reconstructed from image data ranging from 16 to 96 MB in size [6]. The results of this feasibility study showed good intra- and interrater reliability with ICCs larger than 0.92 (p < 0.001). The results also showed that the spinal curvature obtained by the new method had a good linear correlation with the X-ray Cobb method (r2 = 0.8; p < 0.001).

Although these results suggest that the ultrasound volume projection imaging method can be a promising approach for the assessment of spinal deformity, there were still a number of factors that contributed to errors. For example, the ultrasound system and its data were susceptible to the distortion of the electromagnetic field, leading to a system offset/counteract or transient jitter in the spatial and orientation data. Therefore, precaution must be taken especially if the supporting frame is made of metal. The additional limitations of using the ultrasound volume projection imaging method were as follows: (a) heavy to carry around,

*Posture and Back Shape Measurement Tools: A Narrative Literature Review DOI: http://dx.doi.org/10.5772/intechopen.91803*

**Figure 5.**

*Illustration of 3-D ultrasound system for the measurement of spinal deformity [6].*

(b) expensive, (c) relatively dependent on the skilled operator [51, 52], (d) only measures the spinal curvature and not the whole back and (d) time-consuming for the assessment of the whole spine. Therefore, this suggests that it is not an appropriate tool for clinical practice.

In summary, the main disadvantage of all tactile posture measurement systems is the error produced due to electromagnetic and patient interference during data acquisition process. This is because it is difficult for patients to maintain a static standing position for a long time.

### *4.2.2 Non-tactile tools of measurement of spinal curvature*

In the following section, non-surface measuring systems, such as 3D radiographic imaging systems and inertial measuring units, will be discussed. This is followed by various surface measurement tools, such as Moiré topography, integrated shape imaging system, laser triangulator system and the Kinect sensor system.

### *4.2.2.1 Non-surface measuring systems*

### *4.2.2.1.1 3D radiographic imaging*

Cheriet et al. demonstrated the use of biplanar X-ray images for the reconstruction of the three-dimensional spine and rib cage [7]. These images are useful in evaluating patients with spinal deformities like scoliosis. In this method, the reconstruction of images is based on a direct linear transformation technique (DLT), which requires the explicit calibration of an object with known 3D coordinates (see **Figure 6**). This method produced accurate 3D reconstruction of six manually identified anatomical landmarks per vertebra (centres of superior and inferior vertebral endplates and the tips of both pedicles). Similarly, the absolute differences between the Cobb angle obtained with the standard DLT and the explicit calibration methods were as low as 0.3 ± 0.42°. The absolute differences of the frontal and sagittal balance were 0.15 ± 0.15°and 0.37 ± 0.25°, respectively.

Using 3D X-rays for clinical or research purposes has the same motion and radiation issues as the use of 2D X-rays. Additionally, most of these tools are complex to set up, are heavy and only can be applied in laboratory environments.

### **Figure 6.**

*Biplanar X-ray (posterior anterior (PA) and lateral view) acquisition system with calibration apparatus (Cheriet et al. [7]).*

### *4.2.2.2 Inertial sensors*

The recent advancement and application of electronic systems and sensors, namely, accelerometers, gyroscopes, flexible angular sensors, electromagnetic tracking systems and sensing fabrics, have enhanced the quality of clinical practice. Godfrey [53] and Fathi [8] all reported the use of sensors in the evaluation of human posture. The following section reviews their clinical applications, together with their problems and limitations.

An inertial measurement unit (IMU) is an electronic device that primarily contains accelerometers, gyroscope and magnetometer sensors. All these sensors are based on measuring and converting the global position of human body segment, momentum/inertia or changes of path length. An accelerometer is a sensor which measures a specific force and acceleration. In this context, an accelerometer is used to determine the orientation of the spinal segment in relation to the Earth's gravitational field. A gyroscope sensor measures the rate of change of angles. Using these sensors, a three-dimensional (3-D) position together with displacement data is calculated by combining inertial sensors orientation data, together with its known distance between the sensors [54, 55].

Kent et al., in their randomised controlled study, used dorsaVi's hardware (which contains two IMU movement sensors) (see **Figure 7**) to measure posture and movement in subacute and chronic low back pain patients (n = 58) [56]. The results not only demonstrated that the procedure was suitable for posture measurement but also demonstrated its applicability in providing postural biofeedback. Similarly, Fathi and Curran demonstrated the effective application of wireless IMU sensors to detect the curvature of the spine with 85–95% accuracy in ankylosing spondylitis patients [8].

Other portable, non-invasive sensors used in the assessment of posture are e-textiles. Many studies [57, 58] have reported the use of textile sensors to detect the curvature of the spine. The specially designed fabric contains an inductive sensor, a circuit board and a piezoelectric actuator (a component of a machine responsible for moving and controlling the piezoelectric system) (see **Figure 8**). Any change in posture and spinal movement is calculated by a change in the length or position of the sensors together with the percentage of change in electrical resistance.

Sardini et al. compared the e-textile output data with an optical motion system (Vicon) [58]. The trials performed on four subjects obtained on different days demonstrated that the wireless wearable sensor described in this paper is capable of producing reliable data compared with the data obtained with the optical system.

*Posture and Back Shape Measurement Tools: A Narrative Literature Review DOI: http://dx.doi.org/10.5772/intechopen.91803*

### **Figure 7.**

*ViMove wearable motion-sensor system with IMU sensors and surface EMG electrodes (Kent et al. [56]).*

### **Figure 8.** *E-textile with inductive sensors [58].*

As the above IMU and e-textile tools were low-cost, portable and easy to use, it might be appropriate to use these for monitoring movement. The reliability of the above tools for measuring spinal curvatures or other back parameters has not yet been reported. The potential limitation of the IMU and e-textile tools is that their interaction with metal in the environment could affect the sensor data extraction due to its capacity to distort electromagnetic waves. In addition, these tools do not provide back surface and whole-body data.

### *4.2.2.3 Surface measuring systems*

Berryman et al. detail that back surface observation and measurement methods have been widely used by both clinicians and researchers for the evaluation of posture and spinal curvature in patients with spinal disorders [59]. The following section aims to review both the qualitative and quantitative studies that describe skin surface measurement tools.

### *4.2.2.3.1 Moiré topographic methods*

Moiré topography and rastereo photography systems are the most valuable and widely used non-radiographic tools in the measurement of posture/back surface. Additionally, these instruments are also used for screening three-dimensional

spinal deformities and furthermore for quantifying the progression of the 3D spinal curvature.

The above topographical systems work on the basis of projecting a structured light onto the back surface. Based on the reflection of the structured light from the subject, Moiré topography images are produced (see **Figure 9**). The contour map image helps to visualise back asymmetry and record the spatial information of the subject's three-dimensional back shape and posture. The quantification of Moiré fringes typically involves the derivation of quantitative angular and/or linear measures by comparing the left and right side back surfaces.

Numerous authors [60, 61] have described the use of the Moiré topography method to evaluate back shape and spinal deformity. The main limitation of the Moiré topography method is that the measurement depends on the absolute order of Moiré fringes.

A Moiré pattern is a low-frequency line image produced from two highfrequency line images or grids. For example, by projecting a high-frequency grid onto an object and viewing the reflection of this projected pattern through another high-frequency grid is called Moiré fringes [62]. The formation of the Moiré fringes depends on a patient's position. A slight change in the patient's position or movement can produce considerable changes in the Moiré topogram. Thus, a direct inspection of Moiré fringes may be misleading. Further Stokes and Moreland states that the data analysis is a complex procedure, requiring much expertise [63]. Additionally, Nissinen et al. also reported that the correlation of Moiré topographs with X-rays is poor and ranges from r = 0.24–0.45 [64].
