Established acetabular radiological reference values can be reliably transferred to reconstructed parallel-beam 2D images from ultra-low-dose pelvic CT

Arnaud Klopfenstein; Thomas Marth; Stefan Sommer; Reto Sutter; Patrick O. Zingg; Dominik Kaiser

doi:10.1016/j.jor.2025.12.063

View/Download PDF

Buy Reprints

PDF

Translate this page into:

74 (); 73-78

doi:

10.1016/j.jor.2025.12.063

Established acetabular radiological reference values can be reliably transferred to reconstructed parallel-beam 2D images from ultra-low-dose pelvic CT

Arnaud Klopfenstein^a,, Thomas Marth^b,c, Stefan Sommer^c,d, Reto Sutter^b, Patrick O. Zingg^a, Dominik Kaiser^a

a Orthopaedic Department, Balgrist University Hospital, University of Zurich, Forchstrasse 340, 8008, Zurich, Switzerland

b Radiology Department, Balgrist University Hospital, University of Zurich, Forchstrasse 340, 8008, Zurich, Switzerland

c Swiss Center for Musculoskeletal Imaging (SCMI), Balgrist Campus AG, Lengghalde 5, 8008, Zurich, Switzerland

d Swiss Innovation Hub, Siemens Healthineers International AG, Zurich, Switzerland

⁎Corresponding author: Arnaud Klopfenstein. arnaud.klopfenstein@balgrist.ch

Received: 2025-12-6, Accepted: 2025-12-27, Published: 2025-12-28

The Authors

Disclaimer:
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.

Abstract

The anteroposterior radiograph of the pelvis is essential for diagnosing hip pathologies. Radiograph-like projections reconstructed as cone-beam images from CT data correlate strongly with conventional radiographs. However, CT inherently uses parallel x-rays rather than a cone-beam geometry. This study aims to determine whether parallel-beam radiograph-like projections from CT provide comparable reference values to cone-beam reconstructions.

63 patients (126 hips) undergoing CT for symptomatic hip pathologies without prior hip surgery were included. From the same CT data, cone-beam and parallel-beam radiograph-like images were reconstructed using a standardized algorithm. Reference values, including lateral center-edge angle (LCEA), medial center-edge angle (MCEA), acetabular index (AI), acetabular arc (AA), extrusion index (EI), crossover sign, and posterior wall sign, were measured on both projection types, eliminating technical bias. Two observers performed all measurements twice to assess inter- and intra-observer reliability, and intraclass correlation coefficients (ICCs) were calculated.

126 hips were analyzed: 52 with LCEA <22° (“acetabular undercoverage”), 49 with LCEA 22°–33° (“normal coverage”), and 25 with LCEA >33° (“acetabular overcoverage”). ICCs between observers and between projection types demonstrated good to excellent reliability for all reference values (0.89–0.99).

Parallel-beam radiograph-like projections demonstrate good to excellent reliability (ICCs: 0.89–0.99) for key reference values of hip pathologies compared to cone-beam radiograph-like projections from the same CT data. These findings suggest parallel beam projections can be reliably used with established reference values for conventional radiographs.

Keywords

Hip joint

Acetabulum

Computed tomography

Radiographic image interpretation

Show Related Articles from PubMed

1 Introduction

Up to date radiographs remain the standard imaging modality in orthopedic hip surgery, routinely performed before and after surgical interventions. Among these, the standardized anteroposterior (AP) radiograph of the pelvis plays an important role in the diagnosis of various hip joint pathologies. Radiologic parameters derived from these images are essential for identifying common conditions such as developmental dysplasia of the hip (DDH) and pincer-type femoroacetabular impingement (FAI). Both conditions are strongly linked to the early onset of osteoarthritis.^1–4 Key parameters used to quantify acetabular coverage and the orientation of the acetabular roof include the lateral center-edge angle (LCEA), medial center-edge angle (MCEA), acetabular index (AI), acetabular arc (AA), extrusion index (EI), crossover sign, and posterior wall sign.¹^,^5–7

Until recently, computed tomography (CT) imaging posed higher radiation risks for young patients; however, advances such as tin prefiltration have significantly reduced radiation doses to levels of conventional radiographs while preserving image quality.⁸^,⁹ In addition to this reduction in radiation exposure, CT provides reliable evaluation of acetabular coverage, roof orientation, 3D deformity analysis, and 3D surgical planning, as well as precise quantification of surgical outcomes.⁴^,¹⁰^,¹¹

A previous study comparing measurements from pelvic AP radiographs and CT scans demonstrated moderate to good reliability, with intermodality intraclass correlation coefficients (ICC) ranging from 0.61 to 0.87.¹² More recently, a study demonstrated that radiograph-like projections reconstructed as cone-beam images from CT data showed no significant differences compared to conventional radiographs, with good to excellent ICCs (0.78–0.99) pre- and postoperatively, highlighting the potential of ultra-low-dose CT scans as a reliable alternative for evaluating clinically relevant pelvic parameters (LCEA, MCEA, AI, AA, EI, crossover sign, and posterior wall sign).¹³

However, as cone beam radiograph-like projections reconstructed from CT data require advanced processing, parallel projection radiograph-like images from CT data can be created independently by clinicians themselves in most image viewers in a few seconds, making them easily accessible. Since the radiographic reference values have been established on cone-beam projections (conventional radiographs), they need to be adapted to parallel projection radiograph-like images from CT data. To achieve this, parallel and cone beam projections are generated from CT data, using an established algorithm, and the clinically relevant pelvic parameters are compared between the two reconstructed projection methods.⁹^,¹⁴

The aim of the study is to validate the reference values for CT-based radiograph-like parallel projections and thus transferring the well-established reference values, such as the LCEA, MCEA, AI, AA, EI as well as the identification of the crossover sign, and posterior wall sign, from conventional cone beam based radiographs to the reconstructed radiograph-like parallel projection CT images.

2 Methods

This is a retrospective study including real-world anonymized data sets from Balgrist University Hospital, Zurich. The study was approved by the local institutional review board (Cantonal Ethics Committee Zurich, BASEC Nr. 2025-00493). All patients or their legal representatives (i.e., parents or guardians) provided written informed consent for the use of their data.

2.1

2.1 Patient population

A series of 63 patients (126 hips) who underwent CT imaging for symptomatic hip pathologies between January 2016 and January 2024 were included in the study. All patients had no prior history of hip surgery. Among the analyzed hips, 52 had an LCEA <22° (“acetabular undercoverage”), 49 had an LCEA between 22° and 33° (“normal coverage”), and 25 had an LCEA >33°(“acetabular overcoverage”). The LCEA is generally classified on AP radiographs of the pelvis as acetabular undercoverage when LCEA is less than 22°, normal coverage when LCEA is between 22° and 33°, and acetabular overcoverage when LCEA is greater than 33°. The classification of LCEA values is shown in Table 1.

Table 1 Anatomically based LCEA reference values on an AP pelvic radiograph.

LCEA (Degrees°)	Classification
< 22	Acetabular undercoverage
22–33	Normal coverage
> 33	Acetabular overcoverage

2.2

2.2 Image acquisition

The pelvic radiograph is acquired in the supine position with the legs internally rotated by 15° and a film-focus distance of 120 cm. The X-ray beam is centered on the midpoint of the symphysis and a line connecting the anterosuperior iliac spines.¹⁵ CT scans are performed similarly in the supine position with legs rotated internally by 15°, using the following parameters: automated tube voltage selection (CARE kV, reference 120 kV), tube current modulation (CARE Dose4D, reference 147 mAs), a pitch of 0.8, a collimation width of 0.6 mm, and a rotation time of 0.5 s. The cost of a standard pelvic radiograph is approximately 100–150 CHF in Switzerland, whereas a non-contrast pelvic CT scan typically costs between 300 and 450 CHF in outpatient settings.¹⁶

Radiograph-like projections are produced from the CT data using two distinct techniques.

The first technique uses a parallel projection algorithm, which is much simpler and significantly faster. Radiograph-like projections can be generated in less than 10 s directly from the CT data. Unlike the cone beam approach, the parallel beam algorithm does not require advanced computational resources or specialized expertise, making it accessible for examiners like orthopedic surgeons to perform independently.

The second technique employs a 3D cone beam projection algorithm making use of and implemented in MATLAB (The MathWorks, Inc., Version R2018b).⁹^,¹⁴ These cone beam radiograph-like projections replicate the parameters of a standard pelvic radiograph, including identical central beam alignment and a virtual film-focus distance of 120 cm. However, generating images with the cone beam method involves offline processing and requires up to 5 min per image.

2.3

2.3 Measurement

The radiographic parameters on the radiograph-like projections from both the parallel beam and cone beam methods were independently measured twice by an orthopedic surgeon (Examiner 1, AK) and a board-certified musculoskeletal radiologist (Examiner 2, TM). All measurements were performed using the institutional picture archiving and measurement system (Phönix PACS GmbH, Freiburg im Breisgau, Germany). The angles were determined according to the definitions provided in Table 2.⁷

Table 2 Definitions of the investigated radiographic hip parameters.

Lateral center edge angle (LCEA)	Angle formed by a line parallel to the longitudinal pelvic axis and a line connecting the center of the femoral head with the lateral edge of the acetabular sourcil
Medial center edge angle (MCEA)	Angle formed by a line parallel to the longitudinal pelvic axis and a line connecting the center of the femoral head with the medial edge of the acetabular sourcil
Acetabular arc (AA)	Angle formed by two lines connecting the center of the femoral head with the medial and the lateral edge of the acetabular sourcil (sum of the LCEA angle and the MCEA angle)
Extrusion index (EI)	Percentage of uncovered femoral head in comparison to the total horizontal head diameter
Acetabular index (AI)	Angle formed by a horizontal line drawn in the horizontal plane of the pelvis and a line through the most medial point of the sclerotic zone of the acetabular roof and the lateral edge of the acetabulum
Crossover sign	Positive if the projected anterior wall crosses the posterior wall
Posterior wall sign	Positive if the posterior acetabular rim is projected medial of the center of the hip
Retroversion index (RI)	Quantifies the overlap in percentage of the anterior and posterior walls in cases with a positive crossover sign

2.4

2.4 Statistical analysis

The Intraclass Correlation Coefficient (ICC) for the various parameters was calculated using SPSS 27.0 (IBM, Armonk, NY, USA). First, ICC values were determined for each examiner (each examiner performed the measurements twice for each parameter on radiograph-like cone beam projections and radiograph-like parallel beam projections). Subsequently, ICC values were calculated between the two examiners (orthopedic surgeon and musculoskeletal radiologist) and between the two imaging modalities (radiograph-like cone beam projections and radiograph-like parallel beam projections). ICC values of <0.5 indicate poor reliability, 0.5–0.75 indicate moderate reliability, 0.75–0.9 indicate good reliability, and values > 0.9 indicate excellent reliability.

3 Results

A total of 63 patients (126 hips) were included in the study. For every patient, a radiograph-like projections based on CT examinations (63 parallel projection and 63 cone beam projection images) were analyzed (Fig. 1). The mean differences in the measured parameters are summarized in Table 3, and the ICCs are presented in Table 4 and Fig. 2.

Comparison of cone beam radiograph-like projections on the left and parallel beam radiograph-like projections on the right.

Table 3 Overview of measured values.

	Examiner 1		Examiner 2
Mean values radiograph-like projections (parallel beam)
	Mean	SD	Mean	SD
LCEA [°]	24.59	9.44	24.63	9.28
MCEA [°]	37.47	8.21	37.22	8.08
Acetabular arc [°]	62.05	7.73	61.85	7.44
Acetabular index [°]	8.53	7.17	8.40	7.24
Extrusion index	0.23	0.09	0.22	0.09
Number of patients with positive and negative cross over sign/posterior wall sign (parallel beam)
	Positive	Negative	Positive	Negative
Crossover sign (RI ^a ± SD)	67 (23.60±9.20)	59	63 (22.00±9.26)	63
Posterior wall sign	68	58	67	59
Mean values radiograph-like projections (cone beam)
	Mean	SD	Mean	SD
LCEA [°]	25.13	9.47	24.81	9.28
MCEA [°]	37.86	7.89	36.77	8.12
Acetabular arc [°]	63.00	7.46	61.58	7.32
Acetabular index [°]	8.29	7.05	8.15	7.17
Extrusion index	0.23	0.09	0.22	0.09
Number of patients with positive and negative cross over sign/posterior wall sign (cone beam)
	Positive	Negative	Positive	Negative
Crossover sign (RI ^a with SD)	64 (25.50±9.90)	62	66 (25.48±10.89)	60
Posterior wall sign	71	55	73	53
Δ RI with SD ^b	1.41±3.95		4.84±4.04

a Retroversion index (RI) quantifies the overlap in % of the anterior and posterior walls in cases with a positive crossover sign.

b Mean Δ RI in % between imaging methods.

Table 4 Summary of ICC.

N = 126	LCEA	MCEA	Acetabular Arc	Acetabular index	Extrusion index	Crossover sign^b	Posterior wall sign^b
ICC ^a between radiograph-like projections (parallel beam) measure I and II (Examiner 1)
	0.99	0.99	0.99	0.99	0.99	0 (0 %)	0 (0 %)
ICC ^a between radiograph-like projections (parallel beam) measure I and II (Examiner 2)
	0.99	0.99	0.98	0.99	0.98	0 (0 %)	0 (0 %)
ICC ^a between radiograph-like projections (cone beam) measure I and II (Examiner 1)
	0.99	0.99	0.98	0.98	0.94	0 (0 %)	0 (0 %)
ICC ^a between radiograph-like projections (cone beam) measure I and II (Examiner 2)
	0.99	0.99	0.98	0.99	0.98	0 (0 %)	0 (0 %)
ICC ^a between Examiner I and II radiograph-like projections (parallel beam)
	0.99	0.98	0.97	0.99	0.95	6 (4.76 %)	5 (3.96 %)
ICC ^a between Examiner I and II radiograph-like projections (cone beam)
	0.98	0.96	0.93	0.98	0.94	4 (3.17 %)	6 (4.76 %)
ICC ^a between radiograph-like projections (parallel beam) and radiograph-like projections (cone beam) (Examiner 1)
	0.99	0.98	0.96	0.99	0.97	3 (2.38 %)	3 (2.38 %)
ICC ^a between radiograph-like projections (parallel beam) and radiograph-like projections (cone beam) (Examiner 2)
	0.99	0.99	0.97	0.99	0.89	5 (3.96 %)	6 (4.76 %)

a ICC:<0.5: poor reliability, 0.5–0.75: moderate reliability, 0.75–0.9: good, >0.9: excellent reliability.

b Number of different interpretations.

All measured parameters demonstrated excellent reliability (ICC >0.9) when comparing the LCEA, MCEA, AA, EI, and AI between the two examiners. Reliability between the two imaging modalities ranged from good (ICC 0.75–0.9) to excellent (ICC >0.9) for both Examiner 1 and Examiner 2.

The evaluation of the crossover sign and posterior wall sign showed minimal variation between the two imaging methods (2.38 %–4.76 %). Similarly, the variation between the two examiners was minimal (3.17 %–4.76 %). The RI showed minimal mean Δ differences between imaging methods, with values of 1.41 % for examiner 1 and 4.84 % for examiner 2, both higher in cone-beam radiograph-like projections compared to parallel-beam radiograph-like projections.

4 Discussion

The aim of the study is to validate reference values for CT-based radiograph-like parallel projections and thus transferring the well-established reference values for the LCEA, MCEA, AI, AA and EI as well as the identification of crossover sign, and posterior wall sign, from conventional radiographs to parallel projection radiograph-like CT images.

To achieve this, we compared a parallel radiograph-like projection to the corresponding cone beam radiograph-like projection reconstructed from the same CT image data. This approach avoids any potential bias (e.g. variability in patient positioning) as opposed to comparing two separate acquisitions of conventional radiographs and CT scan.¹²^,¹³

Our findings demonstrated good to excellent reliability (ICCs: 0.89–0.99) when comparing radiograph-like projections obtained using cone beam and parallel projection methods. Excellent inter-examiner reliability was also observed across all parameters (ICCs: 0.93–0.99), highlighting the reproducibility of the measurements. Minimal variation was noted between the two methods for the crossover sign and posterior wall sign, with differences ranging from 2.38 % to 4.76 %. Similarly, examiner-dependent variation remained low, ranging from 3.17 % to 4.76 %. These results confirm the robustness of both methods in evaluating clinically relevant pelvic parameters.

In a previous study we have shown that conventional radiographs lead to results with good to excellent reliability (ICCs: 0.78–0.99) compared to cone-beam radiograph-like projections reconstructed from CT data.¹³

Given the comparable outcomes observed between cone beam and parallel beam radiograph-like projections in our study and based on the previous study, we can affirm that CT-based radiograph-like projections using parallel beam represent an easily accessible, reliable and accurate alternative to traditional radiographs.¹³ Our data confirms, that the well-established reference values for conventional radiographs can be applied to CT images as well.

CT-based imaging offers significant advantages over conventional radiographs without increased radiation dose. Stern et al. demonstrated in another study, that tin-filtered ultra-low-dose pelvic CT scans can achieve high-quality imaging with a median effective dose of 0.38 mSv, which is 84 % lower than standard CT scans.⁹ Additionally, CT imaging facilitates 3D deformity analysis, 3D surgical planning, and precise quantification of surgical outcomes. It also allows for the reconstruction of various views, such as pelvic AP, cross-table radiographs, and false-profile images from the same imaging data,¹⁷ enabling additional measurements like the anterior center-edge angle of the acetabular roof.¹⁸ These enhanced capabilities surpass the limitations of conventional radiography, which is prone to rotational and centering errors that may necessitate repeat examinations and thereby potentially increase the radiation dose. These limitations underscore the clinical and operational advantages of adopting CT-based methods, particularly in reducing unnecessary repeat examinations and improving diagnostic efficiency.

Unlike the virtual cone-beam projection approach, which requires more advanced processing algorithm, the parallel projection radiograph-like projections can be generated in less than 10 s directly in most medical imaging viewers, e.g., directly in the PACS by creating a mean intensity projection over a volume of interest (MIP). This reduction in processing improves clinical workflows and no specialized expertise is necessary, enabling examiners (i.e. orthopedic surgeons) to perform the measurements independently.

In summary, we believe that reconstructing radiograph-like images based on parallel projections from tin-filtered ultra-low-dose CT scans can maintain diagnostic quality in cases of DDH and FAI as the established reference values may be applied. This method potentially reduces radiation exposure while offering an efficient imaging workflow. In the near future ultra-low-dose CT imaging has the potential to obliviate the need for conventional radiographs of the pelvis.

There are limitations to this study. One limitation is that all patients included had no prior history of hip surgery, which may limit the generalizability of the findings. However, we do not believe this would alter the clear outcomes and the good to excellent accuracy observed. Additionally, we focused exclusively on parameters derived from pelvic AP radiographs, as these are the most clinically relevant. Future research could include reconstructed cross-table axial and Lequesne's false profile view to validate additional radiographic parameters.

5 Conclusion

In conclusion, parallel projection radiograph-like images demonstrate good to excellent reliability (ICCs: 0.89–0.99) for key reference values of hip pathologies compared to cone-beam radiograph-like projections based on the same CT data. These findings suggest, that CT-based parallel projections can be reliably used to measure the established reference values ((LCEA, MCEA, AI, AA, EI, crossover sign, and posterior wall sign) based on conventional radiographs. Ultra-low dose CT scans have the potential to replace conventional radiographs as conventional radiograph-based measured can be extracted and additional insights are provided by 3-dimensional nature of CT scans without increased radiation dose.

Authors' contributions

Arnaud Klopfenstein acquired and analyzed the data, and drafted the manuscript.

Thomas Marth performed radiological measurements and revised the manuscript.

Stefan Sommer processed the data and critically revised the manuscript.

Reto Sutter supervised radiological methodology and critically revised the manuscript.

Patrick O. Zingg contributed to study design and critically revised the manuscript.

Dominik Kaiser supervised the project and critically revised the manuscript.

All authors read and approved the final manuscript."

Consent for publication

All patients or their legal representative (i.e., parent or guardian) provided written informed consent to use their data.

Ethics approval

The study was approved by the local institutional review board (KEK ZH: BASEC Nr. 2025-00493).

Funding

No financial contributions were made.

References

Ganz R., Parvizi J., Beck M., Leunig M., Nötzli H., Siebenrock K.A., . Femoroacetabular impingement: a cause for osteoarthritis of the hip. Clin Orthop Relat Res. 2003;417:112-120.
[Google Scholar]
Hipp J.A., Sugano N., Millis M.B., Murphy S.B., . Planning acetabular redirection osteotomies based on joint contact pressures. Clin Orthop Relat Res. 1999;364:134-143.
[Google Scholar]
Jacobsen S., Sonne-Holm S., Søballe K., Gebuhr P., Lund B., . Hip dysplasia and osteoarthrosis: a survey of 4151 subjects from the osteoarthrosis substudy of the Copenhagen City heart study. Acta Orthop. 2005;76(2):149-158.
[Google Scholar]
Murphy S.B., Ganz R., Müller M.E., . The prognosis in untreated dysplasia of the hip. A study of radiographic factors that predict the outcome. J Bone Joint Surg Am. 1995;77(7):985-989.
[Google Scholar]
Ganz R., Klaue K., Vinh T.S., Mast J.W., . A new periacetabular osteotomy for the treatment of hip dysplasias. Technique and preliminary results. Clin Orthop Relat Res. 1988;232:26-36.
[Google Scholar]
Mesure des angles FONDAMENTAUX de la hanche RADIOGRAPHIQUE de L'ADULTE par UN rapporteur COMBIN'E [COXOMETRY. Measurement of the basic angles of the adult radiographic HIP BY a combined protractor] Rev Rhum Mal Osteoartic. 1963;30:479-485.
[Google Scholar]
Tannast M., Hanke M.S., Zheng G., Steppacher S.D., Siebenrock K.A., . What are the radiographic reference values for acetabular under- and overcoverage? Clin Orthop Relat Res. 2015;473(4):1234-1246.
[Google Scholar]
Wylie J.D., Jenkins P.A., Beckmann J.T., Peters C.L., Aoki S.K., Maak T.G., . Computed tomography scans in patients with young adult hip pain carry a lifetime risk of malignancy. Arthroscopy. 2018;34(1):155-163.e3.
[Google Scholar]
Stern C., Sommer S., Germann C., et al . Pelvic bone CT: can tin-filtered ultra-low-dose CT and virtual radiographs be used as alternative for standard CT and digital radiographs? Eur Radiol. 2021;31(9):6793-6801.
[Google Scholar]
Dandachli W., Islam S.U., Liu M., Richards R., Hall-Craggs M., Witt J., . Three-dimensional CT analysis to determine acetabular retroversion and the implications for the management of femoro-acetabular impingement. J Bone Joint Surg Br. 2009;91(8):1031-1036.
[Google Scholar]
Tönnis D., Heinecke A., . Acetabular and femoral anteversion: relationship with osteoarthritis of the hip. J Bone Joint Surg Am. 1999;81(12):1747-1770.
[Google Scholar]
Monazzam S., Bomar J.D., Cidambi K., Kruk P., Hosalkar H., . Lateral center-edge angle on conventional radiography and computed tomography. Clin Orthop Relat Res. 2013;471(7):2233-2237.
[Google Scholar]
Kaiser D., Hoch A., Stern C., Sommer S., Sutter R., Zingg P.O., . Accuracy of pelvic measurements on virtual radiographic projections based on computed tomography scans compared to conventional radiographs pre- and postoperatively. Arch Orthop Trauma Surg. 2023;143(6):2965-2971.
[Google Scholar]
Kim K., . 3D cone beam CT (CBCT) projection backprojec- tion FDK, iterative reconstruction matlab examples. Mathworks March. 2012;10
[Google Scholar]
Tannast M., Siebenrock K.A., Anderson S.E., . Femoroacetabular impingement: radiographic diagnosis--what the radiologist should know. AJR Am J Roentgenol. 2007;188(6):1540-1552.
[Google Scholar]
[Google Scholar]
Needell S.D., Borzykowski R.M., Carreira D.S., Kozy J., . CT false-profile view of the hip: a reproducible method of measuring anterior acetabular coverage using volume CT data. Skelet Radiol. 2014;43(11):1605-1611.
[Google Scholar]
Lequesne M., de S.E.Z.E., . [False profile of the pelvis. A new radiographic incidence for the study of the hip. Its use in dysplasias and different coxopathies] Rev Rhum Mal Osteoartic. 1961;28:643-652.
[Google Scholar]