Iterative metal artefact reduction in CT: can dedicated algorithms improve image quality after spinal instrumentation?

Development and Validation of a Deep-Learning-Based Algorithm for Detecting and Classifying Metallic Implants in Abdominal and Spinal CT Topograms

PURPOSE

To develop and validate a deep-learning-based algorithm (DLA) that is designed to segment and classify metallic objects in topograms of abdominal and spinal CT.

METHODS

DLA training for implant segmentation and classification was based on a U-net-like architecture with 263 annotated hip implant topograms and 2127 annotated spine implant topograms. The trained DLA was validated with internal and external datasets. Two radiologists independently reviewed the external dataset consisting of 2178 abdomen anteroposterior (AP) topograms and 515 spine AP and lateral topograms, all collected in a consecutive manner. Sensitivity and specificity were calculated per pixel row and per patient. Pairwise intersection over union (IoU) was also calculated between the DLA and the two radiologists.

RESULTS

The performance parameters of the DLA were consistently >95% in internal validation per pixel row and per patient. DLA can save 27.4% of reconstruction time on average in patients with metallic implants compared to the existing iMAR. The sensitivity and specificity of the DLA during external validation were greater than 90% for the detection of spine implants on three different topograms and for the detection of hip implants on abdominal AP and spinal AP topograms. The IoU was greater than 0.9 between the DLA and the radiologists. However, the DLA training could not be performed for hip implants on spine lateral topograms.

CONCLUSIONS

A prototype DLA to detect metallic implants of the spine and hip on abdominal and spinal CT topograms improves the scan workflow with good performance for both spine and hip implants.

Validation of a 3D CT imaging method for quantifying implant migration following anatomic total shoulder arthroplasty

Glenoid component loosening remains a common complication following anatomic total shoulder arthroplasty (TSA); however, plain radiographs are unable to accurately detect early implant migration. The purpose of this study was to validate the accuracy of a method of postoperative, three-dimensional (3D) computed tomography (CT) imaging with metal artifact reduction (MAR) to detect glenoid component migration following anatomic TSA. Tantalum bead markers were inserted into polyethylene glenoid components for implant detection on 3D CT. In-vitro validation was performed using a glenoid component placed into a scapula sawbone and incrementally translated and rotated, with MAR 3D CT acquired at each test position. Accuracy was evaluated by root mean square error (RMSE). In-vivo validation was performed on six patients who underwent anatomic TSA, with two postoperative CT scans acquired in each patient and marker-based radiostereometric analysis (RSA) performed on the same days. Glenoid component migration was calculated relative to a scapular coordinate system for both MAR 3D CT and RSA. Accuracy was evaluated by RMSE and paired Student's t-tests. The largest RMSE on in-vitro testing was 0.24 mm in translation and 0.11° in rotation, and on in-vivo testing was 0.47 mm in translation and 1.04° in rotation. There were no significant differences between MAR 3D CT and RSA measurement methods. MAR 3D CT imaging is capable of quantifying glenoid component migration with a high level of accuracy. MAR 3D CT imaging is advantageous over RSA because it is readily available clinically and can also be used to evaluate the implant-bone interface.

Low-dose CT with tin filter combined with iterative metal artefact reduction for guiding lung biopsy

Background

Computed tomography (CT) is currently the imaging modality of choice for guiding pulmonary percutaneous procedures. The use of a tin filter allows low-energy photons to be absorbed which contribute little to image quality but increases the radiation dose that a patient receives. Iterative metal artefact reduction (iMAR) was developed to diminish metal artefacts. This study investigated the impact of using tin filtration combined with an iMAR algorithm on dose reduction and image quality in CT-guided lung biopsy.

Methods

Ninety-nine consecutive patients undergoing CT-guided lung biopsy were randomly assigned to routine-dose CT protocols (groups A and B; without and with iMAR, respectively) or tin filter CT protocols (groups C and D; without or with iMAR, respectively). Subjective image quality was analysed using a 5-point Likert scale. Objective image quality was assessed, and the noise, contrast-to-noise ratio, and figure of merit were compared among the four groups. Metal artefacts were quantified using CT number reduction and metal diameter blurring. The radiation doses, diagnostic performance, and complication rates were also estimated.

Results

The subjective image quality of the two scan types was compared. Images with iMAR reconstruction were superior to those without iMAR reconstruction (group A: 3.49±0.65 vs. group B: 4.63±0.57; P<0.001, and group C: 3.88±0.66 vs. group D: 4.82±0.39; P<0.001). Images taken with a tin filter were found to have a significantly higher figure-of-merit than those taken without a tin filter (group A: 14,041±7,230 vs. group C: 21,866±10,656; P=0.001, and group B: 13,836±6,849 vs. group D: 21,639±9,964; P=0.001). In terms of metal artefact reduction, tin filtration combined with iMAR showed the lowest CT number reduction (116.62±103.48 HU) and metal diameter blurring (0.85±0.30) among the protocols. The effective radiation dose in the tin filter groups was 73.2% lower than that in the routine-dose groups. The complication rate and diagnostic performance (sensitivity, specificity, and overall accuracy) did not differ significantly between the tin filter and routine-dose groups (all P>0.05).

Conclusions

Tin filtration combined with an iMAR algorithm may reduce the radiation dose compared to the routine-dose CT protocol, while maintaining comparable diagnostic accuracy and image quality and producing fewer metal artefacts.

Reduction of CT artifacts from cardiac implantable electronic devices using a combination of virtual monoenergetic images and post-processing algorithms

OBJECTIVES

To evaluate the reduction of artifacts from cardiac implantable electronic devices (CIEDs) by virtual monoenergetic images (VMI), metal artifact reduction (MAR) algorithms, and their combination (VMI_MAR) derived from spectral detector CT (SDCT) of the chest compared to conventional CT images (CI).

METHODS

In this retrospective study, we included 34 patients (mean age 74.6 ± 8.6 years), who underwent a SDCT of the chest and had a CIED in place. CI, MAR, VMI, and VMI_MAR (10 keV increment, range: 100-200 keV) were reconstructed. Mean and standard deviation of attenuation (HU) among hypo- and hyperdense artifacts adjacent to CIED generator and leads were determined using ROIs. Two radiologists qualitatively evaluated artifact reduction and diagnostic assessment of adjacent tissue.

RESULTS

Compared to CI, MAR and VMI_MAR ≥ 100 keV significantly increased attenuation in hypodense and significantly decreased attenuation in hyperdense artifacts at CIED generator and leads (p < 0.05). VMI ≥ 100 keV alone only significantly decreased hyperdense artifacts at the generator (p < 0.05). Qualitatively, VMI ≥ 100 keV, MAR, and VMI_MAR ≥ 100 keV provided significant reduction of hyper- and hypodense artifacts resulting from the generator and improved diagnostic assessment of surrounding structures (p < 0.05). Diagnostic assessment of structures adjoining to the leads was only improved by MAR and VMI_MAR 100 keV (p < 0.05), whereas keV values ≥ 140 with and without MAR significantly worsened diagnostic assessment (p < 0.05).

CONCLUSIONS

The combination of VMI and MAR as well as MAR as a standalone approach provides effective reduction of artifacts from CIEDs. Still, higher keV values should be applied with caution due to a loss of soft tissue and vessel contrast along the leads.

KEY POINTS

• The combination of VMI and MAR as well as MAR as a standalone approach enables effective reduction of artifacts from CIEDs. • Higher keV values of both VMI and VMI_MAR at CIED leads should be applied with caution since diagnostic assessment can be hampered by a loss of soft tissue and vessel contrast. • Recommended keV values for CIED generators are between 140 and 200 keV and for leads around 100 keV.

Role of spectral-detector CT in reduction of artifacts from contrast media in axillary and subclavian veins: single institution study in 50 patients

Background

In CT imaging, a high concentration of iodinated contrast media in axillary and subclavian veins after brachial application can cause perivenous artifacts impairing diagnostic assessment of local vascular structures and soft tissue.

Purpose

To investigate reduction of perivenous hypo- and hyperattenuating artifacts of the axillary and subclavian veins using virtual monoenergetic images (VMI) in comparison to conventional CT images (CI), acquired on spectral-detector CT.

Material and Methods

50 spectral-detector CT datasets of patients with perivenous artifacts from contrast media were included in this retrospective, institutional review board-approved study. CT images and virtual monoenergetic images (range 40–200 keV, 10-keV increments) were reconstructed from the same scans. Quantitative analysis was performed by region of interest-based assessment of mean attenuation (HU) and standard deviation in most pronounced hypo- and hyperdense artifacts and artifact-impaired arteries as well as muscle. Visually, artifact reduction, assessment of vessels, and surrounding soft tissue were rated on 5-point Likert-scales by two radiologists.

Results

In comparison to CT images, virtual monoenergetic images of ≥90 keV showed a significant reduction of hypo- and hyperattenuating artifacts (hypodense: CI -220.0±171.2 HU; VMI130keV -13.4±49.1 HU; hyperdense: CI 274.6±184.4 HU; VMI130keV 24.2±84.9 HU; P<0.001). Subjective analysis confirmed that virtual-monoenergetic images of ≥100 keV significantly reduced artifacts (hypodense: CI 2[1–3]; VMI130keV 5[4–5], hyperdense: CI 2[1–4]; VMI130keV 5[5–5], P<0.001) and improved diagnostic assessment. Best results for diagnostic assessment were noted for virtual monoenergetic images at 130 keV. Overcorrection of artifacts was observed at higher keV values. Interrater agreement was excellent for each evaluation and keV value (intraclass correlation coefficient 0.89).

Conclusion

Higher keV virtual monoenergetic images yielded significant reduction of contrast media artifacts and led to improved assessment of vessels and surrounding soft tissue. Recommended keV values for best diagnostic assessment are in the range of 100–160 keV.

Artifact Reduction in the Diagnosis of Vasospasm in Computed Tomographic Perfusion: Potential of Iterative Metal Artifact Reduction

OBJECTIVE

This study aimed to analyze the possibility of artifact reduction using a new iterative metal artifact reduction algorithm (iMAR) in the diagnosis of perfusion deficits due to vasospasms and to evaluate its clinical relevance.

METHODS

Sixty-one volume perfusion computed tomographies of 24 patients after coiling or aneurysm clipping were reconstructed using standard-filtered back-projection and iMAR retrospectively. The degree of artifacts was evaluated as well as the size of the nonevaluable area. Diagnostic performance was evaluated compared with digital subtraction angiography.

RESULTS

Artifacts were present in 39 of 61 volume perfusion computed tomography examinations. Image quality (score, 1.0 vs 1.6; P < 0.01) was higher and the size of the signal loss was reduced significantly by iMAR (intracranial metal artifacts, 887 mm vs 359 mm [P < 0.01]; cranial bolt, 3008 mm vs 837 mm [P < 0.01]). Digital subtraction angiography confirmed vasospasms in 11 (92%) of 12 patients.

CONCLUSION

The iMAR yields higher image quality by reducing artifacts compared with filtered back-projection.

Metal artifacts in patients with large dental implants and bridges: combination of metal artifact reduction algorithms and virtual monoenergetic images provides an approach to handle even strongest artifacts

OBJECTIVES

This study compares reduction of strong metal artifacts from large dental implants/bridges using spectral detector CT-derived virtual monoenergetic images (VMI), metal artifact reduction algorithms/reconstructions (MAR), and a combination of both methods (VMI_MAR) to conventional CT images (CI).

METHODS

Forty-one spectral detector CT (SDCT) datasets of patients that obtained additional MAR reconstructions due to strongest artifacts from large oral implants were included. CI, VMI, MAR, and VMI_MAR ranging from 70 to 200 keV (10 keV increment) were reconstructed. Objective image analyses were performed ROI-based by measurement of attenuation (HU) and standard deviation in most pronounced hypo-/hyperdense artifacts as well as artifact impaired soft tissue (mouth floor/soft palate). Extent of artifact reduction, diagnostic assessment of soft tissue, and appearance of new artifacts were rated visually by two radiologists.

RESULTS

The hypo-/hyperattenuating artifacts showed an increase and decrease of HU values in MAR and VMI_MAR (CI/MAR/VMI_MAR-200keV: - 369.8 ± 239.6/- 37.3 ± 109.6/- 46.2 ± 71.0 HU, p < 0.001 and 274.8 ± 170.2/51.3 ± 150.8/36.6 ± 56.0, p < 0.001, respectively). Higher keV values in hyperdense artifacts allowed for additional artifact reduction; however, this trend was not significant. Artifacts in soft tissue were reduced significantly by MAR and VMI_MAR. Visually, high-keV VMI, MAR, and VMI_MAR reduced artifacts and improved diagnostic assessment of soft tissue. Overcorrection/new artifacts were reported that mostly did not hamper diagnostic assessment. Overall interrater agreement was excellent (ICC = 0.85).

CONCLUSIONS

In the presence of strong artifacts due to large oral implants, MAR is a powerful mean for artifact reduction. For hyperdense artifacts, MAR should be supplemented by VMI ranging from 140 to 200 keV. This combination yields optimal artifact reduction and improves the diagnostic image assessment in imaging of the head and neck.

KEY POINTS

• Large oral implants can cause strong artifacts. • MAR is a powerful tool for artifact reduction considering such strong artifacts. • Hyperdense artifact reduction is supplemented by VMI of 140-200 keV from SDCT.

CT metal artifacts in patients with total hip replacements: for artifact reduction monoenergetic reconstructions and post-processing algorithms are both efficient but not similar

OBJECTIVES

This study compares metal artifact (MA) reduction in imaging of total hip replacements (THR) using virtual monoenergetic images (VMI), for MA-reduction-specialized reconstructions (MAR) and conventional CT images (CI) from detector-based dual-energy computed tomography (SDCT).

METHODS

Twenty-seven SDCT-datasets of patients carrying THR were included. CI, MAR and VMI with different energy-levels (60-200 keV) were reconstructed from the same scans. MA width was measured. Attenuation (HU), noise (SD) and contrast-to-noise ratio (CNR) were determined in: extinction artifact, adjacent bone, muscle and bladder. Two radiologists assessed MA-reduction and image quality visually.

RESULTS

In comparison to CI, VMI (200 keV) and MAR showed a strong artifact reduction (MA width: CI 29.9±6.8 mm, VMI 17.6±13.6 mm, p<0.001; MAR 16.5±14.9 mm, p<0.001; MA density: CI -412.1±204.5 HU, VMI -279.7±283.7 HU; p<0.01; MAR -116.74±105.6 HU, p<0.001). In strong artifacts reduction was superior by MAR. In moderate artifacts VMI was more effective. MAR showed best noise reduction and CNR in bladder and muscle (p<0.05), whereas VMI were superior for depiction of bone (p<0.05). Visual assessment confirmed that VMI and MAR improve artifact reduction and image quality (p<0.001).

CONCLUSIONS

MAR and VMI (200 keV) yielded significant MA reduction. Each showed distinct advantages both regarding effectiveness of artifact reduction, MAR regarding assessment of soft tissue and VMI regarding assessment of bone.

KEY POINTS

• Spectral-detector computed tomography improves assessment of total hip replacements and surrounding tissue. • Virtual monoenergetic images and MAR reduce metal artifacts and enhance image quality. • Evaluation of bone, muscle and pelvic organs can be improved by SDCT.

Reduction of artifacts caused by orthopedic hardware in the spine in spectral detector CT examinations using virtual monoenergetic image reconstructions and metal-artifact-reduction algorithms

OBJECTIVE

Aim of this study was to assess the artifact reduction in patients with orthopedic hardware in the spine as provided by (1) metal-artifact-reduction algorithms (O-MAR) and (2) virtual monoenergetic images (MonoE) as provided by spectral detector CT (SDCT) compared to conventional iterative reconstruction (CI).

METHODS

In all, 28 consecutive patients with orthopedic hardware in the spine who underwent SDCT-examinations were included. CI, O-MAR and MonoE (40-200 keV) images were reconstructed. Attenuation (HU) and noise (SD) were measured in order to calculate signal-to-noise ratio (SNR) of paravertebral muscle and spinal canal. Subjective image quality was assessed by two radiologists in terms of image quality and extent of artifact reduction.

RESULTS

O-MAR and high-keV MonoE showed significant decrease of hypodense artifacts in terms of higher attenuation as compared to CI (CI vs O-MAR, 200 keV MonoE: -396.5HU vs. -115.2HU, -48.1HU; both p ≤ 0.001). Further, artifacts as depicted by noise were reduced in O-MAR and high-keV MonoE as compared to CI in (1) paravertebral muscle and (2) spinal canal-CI vs. O-MAR/200 keV: (1) 34.7 ± 19.0 HU vs. 26.4 ± 14.4 HU, p ≤ 0.05/27.4 ± 16.1, n.s.; (2) 103.4 ± 61.3 HU vs. 72.6 ± 62.6 HU/60.9 ± 40.1 HU, both p ≤ 0.001. Subjectively both O-MAR and high-keV images yielded an artifact reduction in up to 24/28 patients.

CONCLUSION

Both, O-MAR and high-keV MonoE reconstructions as provided by SDCT lead to objective and subjective artifact reduction, thus the combination of O-MAR and MonoE seems promising for further reduction.