Added value of double reading in diagnostic radiology,a systematic review

Abstract

ObjectivesDouble reading in diagnostic radiology can find discrepancies in the original report, but a systematic program of double reading is resource consuming. There are conflicting opinions on the value of double reading. The purpose of the current study was to perform a systematic review on the value of double reading.MethodsA systematic review was performed to find studies calculating the rate of misses and overcalls with the aim of establishing the added value of double reading by human observers.ResultsThe literature search resulted in 1610 hits. After abstract and full-text reading, 46 articles were selected for analysis. The rate of discrepancy varied from 0.4 to 22% depending on study setting. Double reading by a sub-specialist, in general, led to high rates of changed reports.ConclusionsThe systematic review found rather low discrepancy rates. The benefit of double reading must be balanced by the considerable number of working hours a systematic double-reading scheme requires. A more profitable scheme might be to use systematic double reading for selected, high-risk examination types. A second conclusion is that there seems to be a value of sub-specialisation for increased report quality. A consequent implementation of this would have far-reaching organisational effects.Key Points • In double reading, two or more radiologists read the same images. • A systematic literature review was performed. • The discrepancy rates varied from 0.4 to 22% in various studies. • Double reading by sub-specialists found high discrepancy rates. Electronic supplementary materialThe online version of this article (10.1007/s13244-018-0599-0) contains supplementary material, which is available to authorised users.

Computed tomographic comparison of esophageal hiatal size in brachycephalic and non‐brachycephalic breed dogs

“…This modality proved an effective error-reducing technique, including at subspecialization level for individuals familiar with the measurements. 33 However, the circumstances of CT acquisition could have produced a possible underestimation of the EH dimension. The acquisition of the images was performed in a state of temporary diaphragmatic relaxation, allowing an accurate scan of the thorax.…”

Time Requirement and Feasibility of a Systematic Quality Peer Review of Reporting in Radiology

“…Quality measures can be adapted to the local requirements that can be very specific in relation to the referrers [26]. From Scandinavia, where in different countries (e. g., Sweden and Norway) double reading for different radiological examination procedures is widespread, extensive data is available which show high discrepancy rates of up to over 20 % between the compared radiological reports [14,15]. As an added value for patients, an increased detection rate of breast cancer was demonstrated when using a systematic double reading of mammograms in breast cancer screening [27].…”

“…As various specialist disciplines have developed in medicine over the last decades and the degree of subspecialization has massively increased among referring physicians, radiologists are expected not only to have ever-increasing specialized diagnostic knowledge but also to take into account very complex treatment pathways [10][11][12]. In this context, new approaches aim at improving not only the processes and workflows in radiology, but also the content of radio-logical reports either by using double reading like in different Scandinavian countries or using systematic retrospective peer reviews [13][14][15]. In an extensive review, Geijer et al [14] found discrepancy rates of up to 22 % when comparing reports of double readings.…”

“…In this context, new approaches aim at improving not only the processes and workflows in radiology, but also the content of radio-logical reports either by using double reading like in different Scandinavian countries or using systematic retrospective peer reviews [13][14][15]. In an extensive review, Geijer et al [14] found discrepancy rates of up to 22 % when comparing reports of double readings.…”

Artificial Intelligence in Radiology—Ethical Considerations

“…It is difficult to program such ill-defined concepts into an AI algorithm; mathematics likes certainty and absolutes, and may need artificial lumping of findings into "definitely normal" or "definitely abnormal" categories, thereby reducing the power and accuracy of the AI tool. Furthermore, not all radiologists will come to the same conclusion about an imaging study; ground truth labelling accuracy can be enhanced by combining interpretations by more than one radiologist [17].…”