Introduction

Over the past 122 years since the discovery of x-rays, diagnostic imaging has undergone significant evolution.1,2 During this time, the method of image acquisition has shifted from an analog process to a digital one.3 The way projectional radiographs are acquired, manipulated, stored and viewed has advanced, leading to significant changes for all stakeholders.4 Images are now able to be viewed simultaneously by multiple viewers across differing geographic locations and stored for almost instantaneous retrieval.3 What has remained constant since the inception of diagnostic x-ray imaging is the need for appropriate image quality to provide an accurate diagnosis.

The literature acknowledges that dose and image quality are directly related.5,6 Image quality can be significantly improved by increasing the exposure factors, but this is at the expense of increased radiation dose to the patient.7 Optimization, rather than maximization, of image quality in diagnostic radiography should be the chief goal. An optimized technique results in the clinical question being effectively answered, while not imposing a radiation dose to the patient that is higher than necessary.8 By utilizing an optimized technique, radiographers are able to ensure that their commitment to keeping doses "as low as reasonably achievable" is met, while not compromising the diagnostic quality of the examination.

In the literature, digital radiography is used as an umbrella term for images that are acquired through any mechanism that transforms the incident photon into an electrical charge.9 Under this definition, digital radiography is comprised of both computed radiography (CR) and direct digital radiography (DDR).3 Direct digital radiography systems acquire images by converting the incident x-ray energy into a digital signal almost instantaneously,6 skipping the intermediary storage step that is associated with CR.10 The detector used in DDR systems acts as both the acquisition and conversion device, whereas a CR system has a separate acquisition device (the photostimulable phosphor plate) and conversion device (the processor). In a DDR system, the mechanism by which the energy is transformed into a digital signal depends on the type of detector used, and this is the method by which DDR systems are classified.3 This review will focus only on the optimization of radiographic technique parameters for DDR.

Common across all imaging modalities, not just DDR, is the need for appropriate image quality for diagnosis. When considering the term "image quality", it is important to make the distinction between a visually appealing image and an image of adequate quality. An image of "adequate" quality can effectively answer the clinical question posed,7 regardless of whether the image is visually appealing to the reader or not. Adequate image quality in analog imaging revolved around obtaining images with optimal contrast and density.4 Image contrast and density were almost solely dependent on exposure technique and film-screen combination factors chosen prior to acquisition.11,12 As the radiographic film acted as both the acquisition and display medium, there were limited means of altering the image appearance after exposure.10 The transition from analog to digital imaging saw the decoupling of the acquisition and display mediums.4

In terms of acquisition, DDR detectors have a wider dynamic range than that of film-screen. The dynamic range, also known as latitude, of an acquisition device refers to the range of exposure values over which it is able to produce an adequate image.13 Direct digital radiography detectors do not require tight control of exposure factors in order to produce an image of diagnostic quality, as was the case in film imaging, due to their wide dynamic range.12 Another advantage of the wide dynamic range of the DDR detector is their ability to represent structures of varying attenuation in a single image.3 In terms of image display, digital radiography images are able to be manipulated after their creation by way of post-processing. Optimal contrast and brightness is no longer reliant on the use of a specific film-screen combination or set of radiographic technique parameters.10

Digital radiography technology has given rise to many avenues for dose reduction; no longer bound by a certain exposure requirement for optimal image quality, the new limiting factor is image noise.14,15 There are a number of sources that are responsible for image noise,16 yet regardless of its origin, all noise leads to degradation of image quality. Noise is the result of statistical fluctuations in signal intensity received by the detector, and is represented in the resultant image as fluctuations in brightness, leading to a mottled appearance.5,17 Visual appreciation of image noise is very subjective18 and what constitutes an acceptable level of noise depends on both the preference of the observer and on the clinical question being asked.12,19

Image quality research in medical imaging is performed using a variety of methods, one or a combination of test objects, phantoms, and clinically acquired images.7 Test objects measure a specific quality of an imaging system under ideal conditions, but it is difficult to link these results to performance in clinical use.7 Imaging phantoms used for research fall into one of two broad categories: geometric or anthropomorphic. Geometric phantoms consist primarily of geometric shapes, whereas anthropomorphic phantoms are designed to be analogous to human tissue and accurately represent the anatomical structure of the body.20 As images of test objects alone are unable to be directly linked to clinical performance,7 only studies using phantoms and/or clinically acquired images will be included for this review.

Subjective and objective measures of image quality exist, summarized well in Martin et al.7 Subjective measures of image quality, such as visual grading analysis, performed on clinical images by appropriately credentialed individuals are useful as they allow more direct assessment of clinical utility of the resultant image.21,22 The most common objective measure of image quality is the signal to noise ratio, which describes the strength of a signal in the presence of background noise.23 For the purpose of this review, any method of image quality evaluation will be considered, provided that it is applied in an appropriate context.

There are five radiographic technique parameters available to be manipulated at the time of image acquisition. These are exposure factors (tube current time product and tube voltage), source-to-image distance, a choice of additional beam filtration and a method of scatter reduction. The applied tube voltage directly controls the peak energy of the x-ray beam which is described by kilovoltage (kV).24 The current applied to the x-ray tube and the length of time the current is applied for are described by milliampere-seconds (mAs).24 Additional beam filtration is used to remove low energy photons, and this acts on top of the inherent filtration within the tube housing. It is used to reduce the number of photons that would have sufficient energy to reach the patient, but insufficient energy to add to the diagnostic image, therefore adding only to the overall patient dose.24 Source-to-image distance is the distance between the x-ray source and the image receptor.2 Scattered radiation, which degrades image quality, can be compensated for by use of either an air-gap technique or an anti-scatter grid.11 Manipulation of each of these parameters has a direct impact on patient dose, and on resultant image quality. Traditional selection of technique parameters is due to a combination of governing body recommendations, manufacturer recommendations, and of the personal experience of the performing radiographer.25

Optimization of radiographic technique parameters for improved image quality is a key concern in the pursuit of providing high-level patient care. While the image acquisition technology has evolved and continues to advance, it is evident that limited work has been conducted to optimize technique parameters to suit this new technology.4 To date, a search of PubMed, JBI Database of Systematic Reviews and Implementation Reports, and the Cochrane Database of Systematic Reviews shows that there have been no systematic reviews on optimizing image quality for DDR. This systematic review will synthesize available evidence to highlight areas for improvement upon currently accepted best practice, as well as establish gaps in the literature deserving of further investigation.

Review question

What is the effectiveness of adjusting radiographic technique parameters on image quality in projectional radiographs acquired on a DDR system?

Inclusion criteria

Participants

The review will consider studies that include projectional radiographs acquired on a DDR system of the axial and appendicular skeleton. Projectional radiographs of phantoms, adult or pediatric patients (living or post-mortem) will be considered. Studies using test objects, such as contrast-detail phantoms, to assess detector characteristics in the absence of application to a specific clinical scenario will be excluded from review.

Interventions

This review will consider studies that evaluate the effect of changing any, all, or a combination of the following radiographic technique parameters:

i. Tube voltage within a clinically applicable range: [almost equal to] 40-150 kV

ii. Tube current time product within a clinically applicable range: [almost equal to] 0.1-200 mAs

iii. Additional beam filtration of copper (Cu) or aluminum (Al) in differing thicknesses

iv. Source-to-image receptor distance within a clinically applicable range: >100 cm

v. Use of anti-scatter grid, with a clinically acceptable ratio of 8:1-12:1, or air-gap technique.

Comparators

Evaluations of different ranges or options for each radiographic technique parameter will be compared. Studies need to directly compare either an optimized technique to a currently accepted standardized technique, or at least two different options for optimization of a particular technique parameter to be included.

Outcomes

This review will consider studies that include the following outcomes: evaluation of image quality and patient dose.

Image quality will be evaluated by subjective means; objective evaluation will be included but only in studies that also include subjective evaluation. Subjective image quality evaluation should be performed by individuals appropriately credentialed to make comment or report on diagnostic images. Patient dose will be considered also to ensure that the technique is optimized, but only in studies that also measure image quality.

Types of studies

This review will consider all experimental and quasi-experimental study designs including (but not limited to) randomized controlled trials, non-randomized controlled trials, before and after studies and interrupted time-series studies that meet our inclusion criteria. In addition, analytical observational studies including prospective and retrospective cohort studies, case-control studies and analytical cross-sectional studies will be considered for inclusion. This review will also consider descriptive observational study designs including case series, individual case reports and descriptive cross-sectional studies for inclusion. Studies published in English will be included. Studies published from 1997 will be included, as the first digital flat panel detector was released for use in this year.15

Methods

Search strategy

The search strategy will aim to find both published and unpublished studies. An initial limited search of PubMed and Embase has been undertaken followed by analysis of the text words contained in the title and abstract, and of the index terms used to describe article. This informed the development of a search strategy which will be tailored for each information source. A full search strategy for PubMed is detailed in Appendix I. The reference list of all studies selected for critical appraisal will be screened for additional studies. Authors of included studies will also be contacted to obtain details of other studies worthy of inclusion.

Information sources

The databases to be searched include: PubMed, Embase, Scopus and CINAHL.

The search for unpublished studies will include: ProQuest Repository for Masters and PhD theses. A search in Google Scholar of selected keywords and results from the first 10 pages will be reviewed.

Study selection

Following the search, all identified citations will be collated and uploaded into EndNote X8 (Clarivate Analytics, PA, USA) and duplicates removed. Titles and abstracts will then be screened by two independent reviewers (CS & GM) for assessment against the inclusion criteria for the review. Studies that may meet the inclusion criteria will be retrieved in full and their details imported into JBI System for the Unified Management, Assessment and Review of Information (JBI SUMARI; Joanna Briggs Institute, Adelaide, Australia). The full text of selected studies will be retrieved and assessed in detail against the inclusion criteria. Full-text studies that do not meet the inclusion criteria will be excluded and reasons for exclusion will be provided in an appendix in the final systematic review report. Included studies will undergo a process of critical appraisal. The results of the search will be reported in full in the final report and presented in a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.26 Any disagreements that arise between the reviewers will be resolved through discussion, or with a third reviewer.

Assessment of methodological quality

Selected studies will be critically appraised by two independent reviewers (CS & GM) at the study level for methodological quality in the review using a tailored critical appraisal instrument available for review in Appendix II. Any disagreements that arise will be resolved through discussion or with a third reviewer.

All studies, regardless of their methodological quality, will undergo data extraction and synthesis (where possible).

Data extraction

Data will be extracted from papers included in the review using a tailored extraction tool available in Appendix III by the first author and checked by the author team for accuracy. The data extracted will include specific details about the radiographic technique parameters investigated, method of image quality evaluation, types of examinations investigated, subject used for the investigation (geometric phantom, anthropomorphic phantom, post-mortem subject, or evaluation of clinically acquired images), and type of DDR detector used and the results for image quality. Any disagreements that arise between the reviewers will be resolved through discussion. Authors of papers will be contacted to request missing or additional data where required.

Data synthesis

Papers will, where possible, be pooled in statistical meta-analysis using OpenMeta [Analyst]. For head-to-head comparisons, effect sizes will be expressed as either odds ratios (for dichotomous data) or weighted (or standardized) mean differences (for continuous data), and their 95% confidence intervals will be calculated for analysis. Heterogeneity will be assessed statistically using the standard chi-squared and I-squared tests. The choice of model (random or fixed effects) and method for meta-analysis will be based on the guidance by Tufanaru et al.27 OpenMeta [Analyst] will be used for single group analyses of continuous variables using means and for single group analyses of dichotomous variables using a Freeman-Tukey transformation.

Subgroup analyses will be conducted where there are sufficient data to investigate specific ranges of radiographic technique parameters for adult and paediatric populations, for specific examinations of discrete body regions. Sensitivity analyses will be conducted to test decisions made regarding our analytical approach and our assumptions regarding the grouping of similar data where uncertainty arises.

A funnel plot will be generated to assess publication bias if there are 10 or more studies included in a meta-analysis. Statistical tests for funnel plot asymmetry (Egger test, Begg test, Harbord test) will be performed where appropriate.

Assessing certainty in the findings

A Summary of Findings will be created using GRADEpro software (McMaster University, ON, Canada). As no formal Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach exists currently for these types of reviews, we will try to align our judgment of certainty in the results to the GRADE principles while tailoring to this type of review. The Summary of Findings will present a ranking of the quality of the evidence based on study limitations (risk of bias), indirectness, inconsistency, imprecision and publication bias. The following outcomes will be included in the Summary of Findings: subjective image quality, objective image quality and patient dose.

Acknowledgments

This review protocol contributes to the partial fulfilment of the requirements of the Master of Clinical Science degree program for author CS awarded by The University of Adelaide in conjunction with JBI. This program was funded by the Australian Government through an Australian Government Research Training Program (RTP) Scholarship.

The authors would also like to acknowledge Mr Michael Neep for his guidance and comments.

Appendix I: Search strategy for PubMed

Appendix II: Critical appraisal form

Appendix III: Data extraction tool

References