Value of vendor-agnostic deep learning image denoising in brain computed tomography: A multi-scanner study.

To evaluate the effect of a vendor-agnostic deep learning denoising (DLD) algorithm on diagnostic image quality of non-contrast cranial computed tomography (ncCT) across five CT scanners.This retrospective single-center study included ncCT data of 150 consecutive patients (30 for each of the five scanners) who had undergone routine imaging after minor head trauma. The images were reconstructed using filtered back projection (FBP) and a vendor-agnostic DLD method. Using a 4-point Likert scale, three readers performed a subjective evaluation assessing the following quality criteria: overall diagnostic image quality, image noise, gray matter-white matter differentiation (GM-WM), artifacts, sharpness, and diagnostic confidence. Objective analysis included evaluation of noise, contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR), and an artifact index for the posterior fossa.In subjective image quality assessment, DLD showed constantly superior results compared to FBP in all categories and for all scanners (p<0.05) across all readers. The objective image quality analysis showed significant improvement in noise, SNR, and CNR as well as for the artifact index using DLD for all scanners (p<0.001).The vendor-agnostic deep learning denoising algorithm provided significantly superior results in the subjective as well as in the objective analysis of ncCT images of patients with minor head trauma concerning all parameters compared to the FBP reconstruction. This effect has been observed in all five included scanners. · Significant improvement of image quality for 5 scanners due to the vendor-agnostic DLD. · Subjects were patients with routine imaging after minor head trauma. · Reduction of artifacts in the posterior fossa due to the DLD. · Access to improved image quality even for older scanners from different vendors. · Kapper C, Müller L, Kronfeld A et al. Value of vendor-agnostic deep learning image denoising in brain computed tomography: A multi-scanner study. Fortschr Röntgenstr 2024; DOI 10.1055/a-2290-4781.

Gender differences in self-assessed performance and stress level during training of basic interventional radiology maneuvers.

OBJECTIVES: Gender differences have been reported to influence medical training. We investigated gender differences encountered during training in interventional radiology maneuvers. METHODS: Catheter handling was analyzed under standardized conditions in 64 participants naïve to endovascular procedures (26 women, 38 men). Objective (e.g., catheter pathway, catheter movements, required time) and subjective parameters (stress level) were recorded. The NASA-Task Load Index (NASA-TLX; 1-20 points) was used to assess participants' stress levels and perceived workload. RESULTS: In the easier tasks, no significant differences between male and female participants regarding catheter handling were observed. In the most complex task, female participants took themselves more time (688 ± 363 vs. 501 ± 230 s; p = 0.02), asked for help more frequently (n = 19 vs. n = 8) and earlier than men (203 ± 94 vs. 305 ± 142 s; p = 0.049), whereas men stood out by more agitated catheter handling (6.0 ± 1.8 vs. 4.8 ± 1.6 movements/s; p = 0.005). Overall, female participants perceived tasks to be more difficult (11.5 ± 4.2 vs. 9.6 ± 3.3; p = 0.016), perceived higher stress levels (8.9 ± 4.9 vs. 6.3 ± 4.4; p = 0.037), and rated their own performance lower (9.12 ± 3.3 vs. 11.3 ± 3.3; p = 0.009). However, female participants were able to correlate self-assessed with objective parameters correctly (r between -0.555 and -0.469; p = 0.004-0.018), whereas male participants failed to correctly rate their performance (p between 0.34 and 0.73). Stress levels correlated with objective parameters in males (r between 0.4 and 0.587; p < 0.005), but not in female participants. CONCLUSION: Perceived stress levels, self-evaluation skills, and catheter handling differ greatly between untrained male and female participants trying to solve interventional radiological tasks. These gender-specific differences should be considered in interventional radiology training. CLINICAL RELEVANCE STATEMENT: As psychological aspects may influence individual working strategies gender-specific differences in self-perception while learning interventional radiology maneuvers could be essential regarding success in teaching and treatment outcomes. KEY POINTS: • After performing standardized training, 38 male and 26 female volunteers showed significant differences regarding objective and self-assessed performance, as well as in perceived workload while performing simulated endovascular catheter maneuvers. • After solving simulated endovascular radiological tasks, female participants were able to self-assess their objective performance much more accurately than male participants. • Women took more time to solve simulated endovascular tasks and asked earlier and more frequently for help than men.

Ultrafast Brain MRI Protocol at 1.5 T Using Deep Learning and Multi-shot EPI.

RATIONALE AND OBJECTIVES: To evaluate clinical feasibility and image quality of a comprehensive ultrafast brain MRI protocol with multi-shot echo planar imaging and deep learning-enhanced reconstruction at 1.5T. MATERIALS AND METHODS: Thirty consecutive patients who underwent clinically indicated MRI at a 1.5 T scanner were prospectively included. A conventional MRI (c-MRI) protocol, including T1-, T2-, T2*-, T2-FLAIR, and diffusion-weighted images (DWI)-weighted sequences were acquired. In addition, ultrafast brain imaging with deep learning-enhanced reconstruction and multi-shot EPI (DLe-MRI) was performed. Subjective image quality was evaluated by three readers using a 4-point Likert scale. To assess interrater agreement, Fleiss' kappa (Ï°) was determined. For objective image analysis, relative signal intensity levels for grey matter, white matter, and cerebrospinal fluid were calculated. RESULTS: Time of acquisition (TA) of c-MRI protocols added up to 13:55 minutes, whereas the TA of DLe-MRI-based protocol added up to 3:04 minutes, resulting in a time reduction of 78%. All DLe-MRI acquisitions yielded diagnostic image quality with good absolute values for subjective image quality. C-MRI demonstrated slight advantages for DWI in overall subjective image quality (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.87 [+/- 0.37], P = .04) and diagnostic confidence (c-MRI: 3.93 [+/- 0.25] vs DLe-MRI: 3.83 [+/- 3.83], P = .01). For most evaluated quality scores, moderate interobserver agreement was found. Objective image evaluation revealed comparable results for both techniques. CONCLUSION: DLe-MRI is feasible and allows for highly accelerated comprehensive brain MRI within 3minutes at 1.5 T with good image quality. This technique may potentially strengthen the role of MRI in neurological emergencies.

Clinical feasibility and validation of the accelerated T2 mapping sequence GRAPPATINI in brain imaging.

Rationale and objectives: To prospectively evaluate feasibility and robustness of an accelerated T2 mapping sequence (GRAPPATINI) in brain imaging and to assess its synthetic T2-weighted images (sT2w) in comparison with a standard T2-weighted sequence (T2 TSE). Material and methods: Volunteers were included to evaluate the robustness and consecutive patients for morphological evaluation. They were scanned on a 3 T MR-scanner. Healthy volunteers underwent GRAPPATINI of the brain three times (day 1: scan/rescan; day 2: follow-up). Patients between the ages of 18 and 85 years who were able to provide written informed consent and who had no MRI contraindications were included. For morphological comparison two radiologists with 5 and 7 years of experience in brain MRI evaluated image quality using a Likert scale (1 being poor, 4 being excellent) in a blinded and randomized fashion. Results: Images were successfully acquired in ten volunteers with a mean age of 25 years (ranging from 22 to 31 years) and 52 patients (23 men/29 women) with a mean age of 55 years (range of 22-83 years). Most brain regions showed repeatable and reproducible T2 values (rescan: CoV 0.75%-2.06%, ICC 69%-92.3%; follow-up: CoV 0.41%-1.59%, ICC 79.4%-95.8%), except for the caudate nucleus (rescan: CoV 7.25%, ICC 66.3%; follow-up: CoV 4.78%, ICC 80.9%). Image quality of sT2w was rated inferior to T2 TSE (median for T2 TSE: 3; sT2w: 1-2), but measurements revealed good interrater reliability of sT2w (lesion counting: ICC 0.85; diameter measure: ICC 0.68 and 0.67). Conclusion: GRAPPATINI is a feasible and robust T2 mapping sequence of the brain on intra- and intersubject level. The resulting sT2w depict brain lesions comparable to T2 TSE despite its inferior image quality.

Single-Energy Metal Artifact Reduction (SEMAR) in Ultra-High-Resolution CT Angiography of Patients with Intracranial Implants.

PURPOSE: To evaluate the effects of single-energy metal artifact reduction (SEMAR) on image quality of ultra-high-resolution CT-angiography (UHR-CTA) with intracranial implants after aneurysm treatment. METHODS: Image quality of standard and SEMAR-reconstructed UHR-CT-angiography images of 54 patients who underwent coiling or clipping was retrospectively evaluated. Image noise (i.e., index for metal-artifact strength) was analyzed in close proximity to and more distally from the metal implant. Frequencies and intensities of metal artifacts were additionally measured and intensity-differences between both reconstructions were compared in different frequencies and distances. Qualitative analysis was performed by two radiologists using a four-point Likert-scale. All measured results from both quantitative and qualitative analysis were then compared between coils and clips. RESULTS: Metal artifact index (MAI) and the intensity of coil-artifacts were significantly lower in SEMAR than in standard CTA in close vicinity to and more distally from the coil-package (p < 0.001, each). MAI and the intensity of clip-artifacts were significantly lower in close vicinity (p = 0.036; p < 0.001, respectively) and more distally from the clip (p = 0.007; p < 0.001, respectively). In patients with coils, SEMAR was significantly superior in all qualitative categories to standard images (p < 0.001), whereas in patients with clips, only artifacts were significantly less (p < 0.05) for SEMAR. CONCLUSION: SEMAR significantly reduces metal artifacts in UHR-CT-angiography images with intracranial implants and improves image quality and diagnostic confidence. SEMAR effects were strongest in patients with coils, whereas the effects were minor in patients with titanium-clips due to the absent of or minimal artifacts.

Feasibility of Ultra-High Resolution Supra-Aortic CT Angiography: An Assessment of Diagnostic Image Quality and Radiation Dose.

(1) Background: To evaluate diagnostic image quality and radiation exposure of ultra-high resolution cerebral Computed-Tomography (CT) angiography (CTA) obtained on an ultra-high resolution computed tomography scanner (UHR-CT). (2) Methods: Fifty consecutive patients with UHR-CTA were enrolled. Image reconstruction was processed with a 1024 × 1024 matrix and a slice thickness of 0.25 mm. Quantitative analyses comprising CT values, contrast-noise ratio (CNR) and signal-to-noise ratio (SNR) were performed. Subjective assessment of image quality, vessel contrast, noise, artefacts and delineation of different sized vessels were assessed by two readers on a 4-point scale. Radiation exposure was determined. (3) Results: Hounsfield values (ACI: 461.8 ± 16.8 HU; MCA: 406.1 ± 24.2 HU; BA: 412.2 ± 22.3 HU), SNR (ACI: 35.4 ± 13.1; MCA: 20.8 ± 12.4; BA: 23.7 ± 12.9) and CNR (ACI: 48.7 ± 21; MCA: 63.9 ± 26.9; BA: 48.1 ± 21.4) were remarkably high in all segments. Subjective analysis by two raters (fair agreement, k = 0.26) indicated excellent image qualities (image quality = 4; contrast = 4; noise = 3; artefacts = 4).Our analysis revealed a notably high traceability of the cerebral perforators (3 Points). Radiation exposure was at moderate dose levels (effective dose = 2.5 ± 0.6mSv). (4) Conclusions: UHR-CTA generates highly valuable image qualities that allow the depiction of vessels including cerebral perforators at acceptable dose levels. The UHR-CTA may therefore enhance the detection of small cerebral pathologies and may improve interpretability, especially in settings where high image qualities are crucial for the diagnostic accuracy.