Development and implementation of a quality control protocol for B-mode ultrasound equipment

Abstract

Purpose

Quality assurance (QA) of ultrasound (US) equipment is currently required in only a few countries around the world. In Greece, no national or other norms exist for regulating the use of US equipment. However, to obtain accreditation for the radiology department of a Greek hospital, the establishment and implementation of a quality control (QC) protocol and a QA programme for US equipment was required.

Materials and methods

A literature review regarding US QC/QA procedures was performed. The information collected was used as a guide to create a QC/QA protocol and to obtain an appropriate US QC phantom. Drafting and testing of the initial protocol lasted 6 months. Its final version was implemented for 18 months in two US systems and five US transducers.

Results

The QC tests included in the protocol evaluate mechanical and electrical safety, image display, uniformity, penetration depth, distance accuracy, greyscale display, anechoic object imaging, geometric distortion, and axial/lateral resolution. The only QC test that failed was the test for uniformity since intense non-uniformities were observed that led to the replacement of two linear transducers.

Conclusion

US imaging is considered safe and, where appropriate, is preferred over imaging modalities that use ionizing radiation. However, the lack of QC/QA implies that US image quality is not routinely monitored. Therefore, the possibility of malfunctions that may go undetected and lead to wrong diagnosis cannot be excluded. A QC/QΑ programme can contribute to the elimination of such errors and ensure that performance is maintained over time.

Introduction

Ultrasound (US) imaging is a real-time imaging method that enables investigation of the morphology of internal organs and tissues but also of the blood flow within vessels. It is currently utilized for the diagnosis of numerous medical conditions not only by radiologists but also by many other medical specialties. Indeed, gynaecologists, cardiologists, and urologists are some typical examples of specialties that routinely use US in the everyday clinical practice [1]. It is particularly suitable for the management of chronic diseases that require frequent screening, as it is radiation free, comfortable for patients, and much cheaper than other imaging modalities [2].

Brightness mode (B-mode) US imaging is based on the emission of sound waves by a transducer (utilizing the piezoelectric effect), the transmission of these sound waves into the human body, and the detection of sound waves that are reflected back to the US transducer (echoes) at the interface of two media of different acoustic impedances [2, 3]. Blood flow imaging is based on the Doppler effect, which describes the frequency shift of returning echo signals that is caused by the flow of blood [2, 4]. Since US is based on sound waves, US imaging is generally considered safe compared to other imaging methods that involve ionizing radiation. However, US can potentially have adverse biological effects on the body, as it can heat the tissues or cause cavitation (the formation of a bubble within a biological structure) [2]. This is a concern mainly when high power and prolonged imaging times are used, and the possible long-term effects of US are still unknown. For this reason, the United States Food and Drug Administration (FDA) recommends that US imaging users should consider ways to minimize exposure while maintaining diagnostic quality and apply the principles of As Low As Reasonably Achievable (ALARA) at all times [5].

The fact that US imaging is considered safe may be the reason that US is not regulated in the same way as other imaging modalities, whether they use ionizing radiation (such as computed tomography [CT], X-ray, and mammography units) or not (such as magnetic resonance imaging [MRI]). However, these imaging modalities are regulated not only to protect patients and personnel but also to ensure that images are of high enough quality to enable a certified radiologist to make an accurate diagnosis, considering that all imaging methods have limitations and the skills of individual radiologists may vary. Currently, requirements or recommendations concerning the use of US exist in a few countries only, and no international norms have been issued. Consequently, fears of ‘malpractice’ due to insufficient education and training have arisen [1]. Furthermore, for US equipment, the necessity of implementing quality assurance (QA) standards for imaging modalities using X-rays or MRI is not widely recognized [6].

One of the very few countries in which QA in US imaging is either required or recommended is the United Kingdom (UK), where simple quality control (QC) tests are routinely performed by sonographers [7]. In the UK, experience with these QC tests has shown that inadequate image quality can lead to an incomplete or even wrong diagnosis. Therefore, QC tests are needed to ensure that image quality is maintained at high levels so that an accurate diagnosis can be made [8, 9]. Image quality deterioration is a gradual process, and the sonographer most probably will perceive this degradation only when it has reached a level where it hinders diagnosis. Periodic QC tests offer the advantage of identifying US equipment performance changes and image deterioration before they become so serious that they affect diagnosis in the everyday clinical routine [6]. In cases of malfunction, QC tests facilitate the identification of the cause of the problem [10]. The importance of QC/QA has also been recognized by several medical and scientific associations, which have proposed and described procedures for evaluating US performance [10, 11].

In Greece, QA in US is still in its infancy, and it is not included in the requirements installed by national authorities for the licensing of radiology departments, in contrast to the strict requirements in place for systems using ionizing radiation and for MRI equipment. In addition, no regulations exist for US equipment used by other medical specialties. However, within the context of the activities required to obtain accreditation from the Hellenic Accreditation System (E.SY.D) for any radiology department in Greece, a QA programme in US is needed. Because of this requirement, a QC/QA protocol needs to be established to routinely evaluate the performance of the B-mode US equipment in our hospital. The purpose of this study was to describe the QC protocol and the QA programme created for monitoring the performance of US equipment, as well as their implementation.

Materials and methods

To draft the US QC protocol and establish the QC tests to be included in this, a literature review was initially performed to investigate the existing literature. The literature search was done by the Medical Physics (MP) Unit of the Konstantopoulio General Hospital in Athens, Greece. The MP Unit is responsible for the QC/QA procedures of all medical imaging equipment in the hospital. The search engines used were PubMed, Google Scholar, and ScienceDirect. The goal of the review was to identify: (a) reports provided by national or international professional and scientific societies that describe QA/QC procedures for B-mode US, (b) recommendations of international organizations, and (c) articles or protocols proposed by other hospitals that perform US QA procedures.

After the literature search, the US QC protocol was produced. Drafting the US QC protocol and testing the QC tests that were included lasted 6 months (from January to June 2018). During this time, we also familiarized ourselves with the use of the US equipment and how to best use it together with the QC phantom to execute the QC tests. The US QC protocol was finalized in July 2018 and was applied for a period of about 18 months in two US systems and five US transducers by the medical physics team of the hospital.

The phantom used for the US QC/QA implementation was the Gammex Sono 410 SCG (Gammex Inc, Middleton, USA). This is a multipurpose accreditation phantom that is used to evaluate uniformity, artefacts, geometric accuracy, sensitivity, noise, contrast resolution, distortion, and other relevant technical parameters in US imaging [12]. It is designed to meet all the American College of Radiology (ACR) requirements for biannual or annual system performance evaluation and routine QC tests. This model contains a tissue-mimicking material for which the speed of sound is 1540 ± 10 m/s and the attenuation coefficient is 0.7 ± 0.05 dB/cm/Hz. The phantom contains ten anechoic cylinders (to simulate air-filled cysts). Eight of them have a diameter of 4 mm, and they are positioned at depths ranging from 2 to 16 cm (every 2 cm). These objects can be used to assess the maximum depth of visualization and to determine the farthest anechoic object that can be visualized above the electronic noise. Two more anechoic structures, one with a diameter of 2 mm and one with a diameter of 1 mm, are located at a depth of 2 cm. It also contains two greyscale targets comprised of four cylinders of 8 mm in diameter (one anechoic, one at − 6 dB, one at + 6 dB, and one at + 12 dB relative to the background material), one located at a depth of 4 cm and the other at a depth of 11 cm. Finally, the phantom contains a number of strings of 0.1 cm in diameter located along the vertical midline: one each at a depth of 1, 2, 3, 5, 9, 11, and 15 cm and three more sets of two strings located on either side of the midline (2 or 4 cm in horizontal distance) at depths of 2, 7, and 13 cm. For the curvilinear and curved transducers, the phantom is positioned upside down to utilize the specially designed bottom surface, which facilitates scanning with those transducers. To ensure an optimally working phantom and transducer, coupling gel is always used, as in the clinical practice.

The study was submitted and approved by the ethics committee of the hospital. As there were no patients included in the study, informed consent was waived. The US QC protocol was established, and the first results of QC testing using the protocol are described in detail below.

Results

The literature review identified a limited number of references relevant to our objective [6–19]. As far as reports from professional societies are concerned, the American Association of Physicists in Medicine (AAPM) (Report No. 1 of the Ultrasound Task Group [10]) and the Institute of Physics and Engineering in Medicine (IPEM) in the UK (Report No. 102 [11]) were the only professional societies reporting routine QC test procedures for US. Recommendations were also found in the guidelines of the American Institute of Ultrasound in Medicine (AIUM) [13] and the British Medical Ultrasound Society (BMUS) [7, 14]. The information contained in the literature was studied to decide which tests should be included in our US QC protocol, given the US QC phantom available. The QC tests included in the protocol, along with the action limits and the frequency of tests, are presented in Table 1. The QC tests were performed in shorter intervals than those suggested in Table 1 to gain experience with performing these tests and to investigate whether shorter QC intervals would possibly be required for the early detection of image quality problems.

Table 1.

Quality assurance protocol for ultrasound equipment

Procedure	Action limit	Frequency
Mechanical integrity	Visually observed problem	Daily
Image display	Visually observed problem	6 months
Image uniformity	> 4 dB	3 months
Penetration depth	Decrease of 0.6 cm from baseline	6 months
Distance accuracy	Deviation from nominal value: 1.5 mm for vertical distances 2 mm for horizontal distances	6 months
Anechoic cysts-Distortion	Systematic deviation from baseline: < 90% or > 1.15	6 months
Axial–lateral resolution	≤ 1 mm for frequencies > 4 MHz  ≤ 2 mm for frequencies ≤ 4 MHz	6 months
Gray scale display	> 1 dB from baseline	6 months

QC protocol description

The protocol created for our radiology department consisted of the evaluation of the following parameters:

Mechanical integrity

A visual inspection of the equipment must be performed οn a daily basis. It is important that the mechanical integrity and electrical safety are ensured prior to the examinations. It must be confirmed that the transducers and power cords are in good condition, without cracks, discoloration, and damage to the cables. The video monitor and the control panel must be clean, with the switches and lights fully operating. The wheels and the wheel locks must be functional, not only to rotate freely but also to lock securely. The filters must be free of dirt and dust. Finally, all the accessories must be in good condition and fastened securely to the US unit [8–10, 13–17].

Image display

Image display is evaluated by monitoring a greyscale pattern, such as the pattern of the Society of Motion Picture and Television Engineers (SMPTE). The first, the last, and the number of visible greyscale steps are recorded. It must be confirmed that there is no visually observable image distortion and that the letters and numbers appear clearly in the text [10, 13–15, 17].

Image uniformity

To evaluate the image uniformity, the phantom has a uniform region that needs to be scanned and monitored. Then the image on the monitor must be examined for the existence of streaking. In case of streaking, the grey level of the non-uniformity must be evaluated. If it is greater than 4 dB, the service company must be contacted so that the malfunction can be fixed. Additionally, the type of non-uniformity must be identified, since horizontal non-uniformities and signal dropout usually signify lost elements in the transducer, while vertical non-uniformities are most probably due to time gain compensation problems. Therefore, by identifying the type of non-uniformities, the cause of the malfunction can be determined [8–10, 13, 15–17].

Penetration depth

The QC test for penetration depth is a procedure with which the maximum depth that can be visualized is evaluated. For this test, the power of the transmitter is set to its maximum level (100%). The focal zone must be adjusted so that the US beam is focused on the greatest depth possible, and the receiver gain should be appropriately increased to allow visualization of echoes from deeper structures. The penetration depth is the distance from the surface of the phantom to the deepest visible structure inside the phantom or, alternatively, the distance from the surface of the phantom to the depth where the tissues’ echoes fade into noise [1, 6, 8–10, 13, 15–17]. This measurement should be performed during real-time image acquisition, as in this way electronic noise can be distinguished from the phantom’s visible texture pattern: when the transducer is held stationary over the phantom, the texture pattern is constant, whereas the electronic noise pattern is changing [10].

Figure 1 shows images from the measurements of penetration depth. In Fig. 1a, for a linear array operating at 9.4 MHz, the string located at a depth of 7 cm is clearly visible. Figure 1b shows the measurement from the surface to this depth. Finally, Figure 1c shows that with a curvilinear transducer operating at 3.6 MHz, the most distant structure in the phantom is located at a depth of 16 cm (the phantom is upside down), and the bottom of the phantom is also visible. However, the penetration depth extends to only 13.86 cm, since beyond this depth the image noise prevails.

Fig. 1.

The measurement of the maximum depth of penetration is shown for a linear array transducer operating at 9.4 MHz (a, b) and a curvilinear array transducer operating at 3.6 MHz (c)

Distance accuracy

This test evaluates the accuracy of the electronic calipers of the US software used to measure distances between visualized structures within the human body and dimensions of organs and structures. For the purpose of this QC test, the QC phantom incorporates strings located at specific horizontal and vertical distances. By measuring the distances of the strings within the phantom image with the use of calipers, the distance accuracy can be evaluated both in the horizontal and in the vertical plane [1, 6, 8–10, 13, 15–18]. Figure 2 shows examples of distance measurements in the vertical and horizontal planes.

Fig. 2.

Examples of distance measurements with a curvilinear array transducer operating at 3.6 MHz: a in a vertical direction and b in a horizontal direction

Imaging of anechoic cyst distortion

This test evaluates the imaging of anechoic cysts, and the QC phantom contains spherical anechoic cysts of different dimensions located at various depths within the phantom. The smallest size of anechoic cysts that can be visualized is recorded, and additionally the ratio of the height and the width of a cyst is measured. Since the cysts are spherical, they should be visualized as circles, so their width and height should be the same. A consistent deviation of this ratio from unity shows that image distortion exists [10, 17]. Figure 3 shows an example of diameter measurements in an anechoic structure positioned at a depth of 2 cm to determine the distortion in terms of the height-to-width ratio. For all linear transducers, the anechoic structure of 1 mm in diameter located at a depth of 2 cm was always visible. For the curvilinear transducers, the 4- and 8-mm anechoic structures located at a depth of 14 cm were measured. Recall that since for curvilinear transducers, the phantom is positioned upside down, the 1- and 2-mm anechoic structures located at a depth of 16 cm were not visible.

Fig. 3.

Example of anechoic structure imaging and height-to-width ratio measurements for a linear array transducer operating at 8 MHz

Greyscale display

The phantom contains four different greyscale target cysts. These cysts represent metastases, as these are usually slightly hyperechoic or hypoechoic compared with the surrounding tissue. Their display and detection on the monitor can verify whether the US equipment is able to monitor masses that have a slightly different composition from that of the surrounding normal tissue [13, 15, 18]. Since the US units tested do not offer a region of interest (ROI) tool, a direct measurement of the structures’ attenuation in dB is not possible. In this case, the dual display test was employed. The left and right images were activated in the monitor. In the left image, an image of the greyscale targets was acquired with the receiver gain routinely used for the QC tests. The zoom function was used to focus on the greyscale targets. Switching to dual mode, this image was kept on the left side of the screen, and a new image similar to the one before was acquired (shown on the right side of the screen). On the active image (right side), the gain was increased or decreased until the greyscale of the background matched the greyscale of the selected target on the left image. The difference of the new gain and the gain with which the first image was acquired is the attenuation in dB of the selected target with respect to the background material.

An example of this procedure is shown in Fig. 4, in which the image shown in Fig. 4a was first acquired using a receiver gain of 0 dB. After switching to dual mode, this image was shown to the left of Fig. 4b. The second image shown to the right of Fig. 4b was imaged with increased gain (12 dB) so as to match the grey level of the brightest cyst in the left image, which thus has a 12-dB attenuation relative to the background (i.e., hyperechoic). In all QC tests, the greyscale intensities were practically constant as no variation was observed.

Fig. 4.

Example of the dual display test method used to evaluate the greyscale targets obtained with a linear array transducer operating at 10.7 MHz

Axial-lateral spatial resolution

Axial or lateral resolution is the ability to visualize separately two different small structures that are located very close to each other. Some QC phantoms contain strings that are located close to each other, so the resolution can be easily evaluated. However, if the phantom does not offer this possibility, which is the case for the phantom used in our study, the resolution can be determined by calculating the full width at half maximum (FWHM) from the profile of a single string whose diameter is 0.1 mm, using image processing programs such as ImageJ (http://imagej.nih.gov/ij, version 1.53a) [1, 6, 9, 10, 15–18]. However, as shown in Fig. 5, this test may suffer from inaccuracies since the relationship between greyscale and signal attenuation is not well defined and, additionally, the background signal variation is too high to obtain a reliable level for the background signal. Recall that axial resolution depends on the spatial pulse length or pulse duration, and in general the higher the resolution the better the axial frequency. Also, imaging of anechoic spherical structures of 1 mm in diameter is an indirect way of verifying that the spatial and lateral resolution is better than 1 mm [10].

Fig. 5.

An example of the measurement of lateral resolution. a The image was viewed using ImageJ, and a horizontal line was drawn passing through a string of 0.1 mm in diameter. b In the generated profile, a rectangular was drawn, starting from the background to the top of the profile, which intersected the profile, resulting in a full width at half maximum (FWHM) of 0.55 mm

The results (pass/fail) of the QA protocol in routine testing performed during the past year on the US units are presented in Table 2 and analytically in Figs. 6, 7, 8, 9, 10, 11, in which the first letter stands for the manufacturer (S for Siemens and T for Toshiba, with the remaining letters identifying the transducer type).

Table 2.

Results from the QC are found for the five types of transducers and the two US machines in the department

US equipment		Type	Visual inspection	Image display	Image uniformity	Penetration depth	Distance Accuracy	Anechoic imaging	Axial and lateral resolution	Gray scale display
Siemens Acuson X-700	VF-10.5	L	✓	✓	✕	✓	✓	✓	✓	✓
	VF-12.4	L	✓	✓	✓	✓	✓	✓	✓	✓
	4C1	C	✓	✓	✓	✓	✓	✓	✓	✓
Toshiba SSA-660A	PVT 704AT	L	✓	✓	✕	✓	✓	✓	✓*	✓
	PVT 375 BT	C	✓	✓	✓	✓	✓	✓	–	✓

✓: pass, ✕: fail, ✓*: pass (based on the visualization of 1 mm anechoic structure), since for Toshiba US unit resolution could not be measured using ImageJ

US ultrasound, L linear, C curvilinear

Fig. 6.

The number (#) and intensity (dB) of the non-uniformities that were observed in the two linear transducers are shown. The sudden drop to zero is due to the replacement of the transducers

Fig. 7.

The penetration depths and respective limits are shown. Open symbols (data) and black dashed lines (lower and upper limits) refer to the primary Y-axis (left). The dark symbols and the thicker black lines refer to the secondary Y-axis (right)

Fig. 8.

The vertical distance measurements at different depths are shown for all US units and transducers

Fig. 9.

The horizontal distance measurements at different depths are shown for all US units and transducers

Fig. 10.

The ratios of the anechoic structures’ diameter measurements (height/width) are shown. For the linear transducers, the anechoic structures located at a depth of 2 cm were measured. For the curvilinear structures, the anechoic structures of 4 and 8 mm in diameter located at a depth of 14 cm were measured

Fig. 11.

The resolution measurements are shown. Note that for the S-4C1 transducer operating at 3.6 MHz the tolerance in resolution is 2 mm

Discussion

A protocol for the routine QC of US equipment operating in B-mode was created and implemented in the radiology department of the Konstantopoulio General Hospital, Athens, Greece. As shown in Table 2, the only malfunction identified was the existence of non-uniformities in two linear transducers. Recent studies report that non-uniformities, along with mechanical integrity problems, are the most common malfunctions observed in US equipment [8, 9]. It should be noted that in our case, the streaking observed in both US transducers was gradually becoming worse. Indeed, although initially the non-uniformities were rare and less than 4 dB, they became worse within a time interval of only 3 months. At this point, the small cyst (of 1 mm in diameter) could no longer be visualized when it was located below the malfunctioning elements of the transducer. This implies that in the clinical practice, the sonographer or the medical professional executing the US would possibly miss small cysts in the patient near the surface. This finding supports the necessity of evaluating the uniformity more than once a year. Dudley et al. [19] have even proposed a simple uniformity test for US phased arrays to detect element failures and, therefore, the existence of non-uniformities. This finding also proved that the implementation of this QC protocol contributed to repairing the malfunction on time and before it could create problems for diagnosis. The two defective linear transducers were eventually replaced by new transducers.

The penetration depth was approximately 7 cm for the linear transducers; when the curvilinear transducers were used, the penetration depth was approximately 14 cm. These results are in accordance with what was expected, given the frequency of each transducer. The linear transducers operated at higher frequencies than the curvilinear transducers, so the linear transducers had a smaller maximum penetration depth. During repeated QC tests, no decrease in the penetration depth was observed. The calipers were found to work properly, and the distance and dimension measurements were accurate even when small-sized visualized structures were measured. All the relevant QC tests performed showed that the equipment was working properly as all deviations observed were within the limits.

Regarding spatial resolution, since the available phantom does not have a specific region for the evaluation of resolution, axial and lateral resolution were calculated using the ImageJ software. The FWHM of one single string was determined in both axial and lateral planes. Although according to the AAPM report [10], this method is the preferred technique for determining the spatial resolution, we observed that the FWHM values calculated for the same transducer varied across repeated QC tests. This finding suggests that there may be a significant uncertainty when this method is used for determining spatial resolution. It must be noted that due to a malfunction in the Universal Serial Bus (USB) port, the export of Digital Imaging and Communications in Medicine (DICOM) images and the calculation of the spatial resolution using the ImageJ program were not feasible for the Toshiba US unit. For the linear transducer, it can be assumed that the resolution is acceptable, based on the visualization of a 1-mm anechoic structure at a depth of 2 cm. However, for the curvilinear transducer, this indirect method cannot be used since the phantom is positioned upside down and the 1-mm anechoic structure is at a depth of 16 cm and no longer visible. Therefore, the resolution could neither be calculated nor estimated for this transducer. The malfunction in the USB port is the reason that images from the Toshiba unit are not presented.

An important issue to note is that in the literature, no specific values are proposed for the action limits of some tests. As an example, even though the AAPM protocol provides specific action limits for the resolution test, penetration depth, and distance accuracy, it does not provide an action limit for the geometric distortion. The action limit proposed is a systematic deviation from the baseline values or a major distortion, e.g., ± 20% or more. During the performance of this test in our equipment, the results of the anechoic imaging and geometric distortion QC test varied from 0.93 to 1.13. Since the cysts were imaged as circles and the results were consistent for repeated tests and, furthermore, the sonographers (who, in our case, were all certified radiologists) found that the visualization of cysts as circles was satisfactory, it was decided to set the 0.9–1.15 range as a local action limit for the height-to-width ratio.

It should be emphasized that to accurately execute the QA programme (perform the QC tests at certain time intervals, analyse the results of tests, and compare the outcome with previous performance), the medical physicist must be experienced with the operation of the US unit. The main reason is that minor modifications to the settings and parameters during the operation can affect the results of the QC tests, leading to problems in the analysis of the QC test data. For example, when the QA programme was implemented for the first time, the greyscale display test could not be conducted in the Siemens US unit. After contacting the service representatives, we were advised to deactivate the dynamic tissue contrast enhancement (DTCE), which improves the tissue contrast resolution through speckle reduction using an advanced proprietary processing algorithm. Once the DTCE was deactivated, the test could be performed easily and successfully.

One final but important issue that needs consideration is that most action limits refer to the deviation from the baseline values that should normally be acquired when the equipment is bought. If the equipment is old, the baseline values should be determined during QC tests performed after service has been performed to the US equipment. When service has been performed by a field service engineer who is certified by the manufacturer and who follows the manufacturer’s service instruction manual, it can be assumed that the US equipment meets its specifications. Action limits based on deviations from baseline performance values have the purpose of showing performance degradation over time and alerting the user to act before this degradation manifests itself in clinical images. To our knowledge, currently no special provision exists for setting the baseline values in the case that the QA programme is implemented in US units that are not new. Although the equipment has not shown any other degradation during the QC monitoring period than the afore-mentioned problem of non-uniformity, the clinicians claimed that the image quality has deteriorated since the equipment was first bought. While the QC protocol results cannot support these claims, it creates reasonable doubts regarding the previous assumption made, namely that after service the US equipment operates within its specifications, and questions the adequacy of the adjustments and tests performed by the field service engineers. To resolve these issues, it is mandatory that the baseline performance of US equipment is set at the time that it is installed and that reference values of expected performance exist for guidance.

Conclusion

A protocol for the routine QC of US equipment operating in B-mode was created and applied to two US units and five US transducers for 18 months by the MP Unit of the Konstantopoulio General Hospital, Athens, Greece. The problem identified during the repeated QC test was non-uniformity in two linear transducers (one in each US unit). This finding agrees with the recent literature, in which the existence of non-uniformities is recognized as the most common malfunction.

In US imaging, equipment malfunction involves a risk for patient safety since a wrong clinical diagnosis may lead to false or delayed treatment whose impact on a patient’s health could be immense. A QC/QA programme for monitoring the equipment performance in terms of image quality can contribute to the avoidance of such errors. It can also be used to verify that the US equipment is properly serviced so that its performance does not deteriorate over time or after repairs and software or hardware updates.

Funding

The authors did not receive support from any organization for the submitted work. No funding was received to assist with the preparation of this manuscript.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.