Contribution of an artificial intelligence deep-learning reconstruction algorithm for dose optimization in lumbar spine CT examination: A phantom study

Ultra-high-resolution CT of the temporal bone: Comparison between deep learning reconstruction and hybrid and model-based iterative reconstruction

PURPOSE

The purpose of this study was to evaluate the ability of ultra-high-resolution computed tomography (UHR-CT) to assess stapes and chorda tympani nerve anatomy using a deep learning (DLR), a model-based, and a hybrid iterative reconstruction algorithm compared to simulated conventional CT.

MATERIALS AND METHODS

CT acquisitions were performed with a Mercury 4.0 phantom. Images were acquired with a 1024 × 1024 matrix and a 0.25 mm slice thickness and reconstructed using DLR, model-based, and hybrid iterative reconstruction algorithms. To simulate conventional CT, images were also reconstructed with a 512 × 512 matrix and a 0.5 mm slice thickness. Spatial resolution, noise power spectrum, and objective high-contrast detectability were compared. Three radiologists evaluated the clinical acceptability of these algorithms by assessing the thickness and image quality of the stapes footplate and superstructure elements, as well as the image quality of the chorda tympani nerve bony and tympanic segments using a 5-point confidence scale on 13 temporal bone CT examinations reconstructed with the four algorithms.

RESULTS

UHR-CT provided higher spatial resolution than simulated conventional CT at the penalty of higher noise. DLR and model-based iterative reconstruction provided better noise reduction than hybrid iterative reconstruction, and DLR had the highest detectability index, regardless of the dose level. All stapedial structure thicknesses were thinner using UHR-CT by comparison with conventional simulated CT (P < 0.009). DLR showed the best visualization scores compared to the other reconstruction algorithms (P < 0.032).

CONCLUSION

UHR-CT with DLR results in less noise than UHR-CT with hybrid iterative reconstruction and significantly improves stapes and tympanic chorda tympani nerve depiction compared to simulated conventional CT and UHR-CT with iterative reconstruction.

Ultra-high-resolution CT of the temporal bone: The end of stapes prosthesis dimensional error and correlation with patient symptoms

PURPOSE

To describe the reliability of ultra-high-resolution computed tomography (UHR-CT) in the measurement of titanium stapes prostheses using manufacturer data as a reference.

MATERIALS AND METHODS

This retrospective study included patients treated by stapedectomy with titanium prostheses who underwent UHR-CT between January 2020 and October 2023. Images were acquired using an ultra-high-resolution mode (slice thickness: 0.25 mm; matrix, 1024 × 1024). Two radiologists independently evaluated the length, diameter, and intra-vestibular protrusion of the prosthesis. Post-operative air-bone gaps (ABGs) were recorded.

RESULTS

Fourteen patients were enrolled (mean age, 44.3 ± 13.8 [SD] years, 9 females), resulting in 16 temporal bone UHR-CTs. The exact length was obtained in 81.3 % (n = 13/16) and underestimated by 0.1 to 0.3 mm in the remaining 18.7 % (n = 3/16) CT scans for both readers (mean misestimation: -0.02 ± 0.06 [SD] mm, overall underestimation of 0.43 %). The exact diameter was reported in 75 % (n = 12/16) and 87.5 % (n = 14/16) of the CT scans for readers 1 and 2, respectively, and was off by 0.1 mm in all discrepancies (mean misestimation: 0.01 ± 0.04 [SD] mm, overall overestimation of 2.43 %). Intravestibular prosthesis protrusion was of 0.5 ± 0.43 [SD] mm (range: 0-1) and 0.49 ± 0.44 [SD] mm (range: 0-1.1) for readers 1 and 2, respectively, and did not correlate with ABGs (r = 0.25 and 0.22; P = 0.39 and 0.47 for readers 1 and 2, respectively). Intra and interobserver agreements were excellent.

CONCLUSION

UHR-CT provides 99.6 % and 97.6 % accuracy for prosthesis length and diameter measurements, respectively.

Artificial Intelligence Iterative Reconstruction in Computed Tomography Angiography: An Evaluation on Pulmonary Arteries and Aorta With Routine Dose Settings

OBJECTIVE

The objective of this study is to investigate whether a newly introduced deep learning-based iterative reconstruction algorithm, namely, the artificial intelligence iterative reconstruction (AIIR), has a clinical value in computed tomography angiography (CTA), especially for visualizing vascular structures and related lesions, with routine dose settings.

METHODS

A total of 63 patients were retrospectively collected from the triple rule-out CTA examinations, where both pulmonary and aortic data were available for each patient and were taken as the example for investigation. The images were reconstructed using the filtered back projection (FBP), hybrid iterative reconstruction (HIR), and the AIIR. The visibility of vasculature and pulmonary emboli and the general image quality were assessed.

RESULTS

Artificial intelligence iterative reconstruction resulted in significantly ( P < 0.001) lower noise as well as higher signal-to-noise ratio and contrast-to-noise ratio compared with FBP and HIR. Besides, AIIR achieved the highest subjective scores on general image quality ( P < 0.05). For the vasculature visibility, AIIR offered the best vessel conspicuity, especially for the small vessels ( P < 0.05). Also, >90% of emboli on the AIIR images were graded as sharp (score 5), whereas <15% of emboli on FBP and HIR images were scored 5.

CONCLUSION

As demonstrated for pulmonary and aortic CTAs, AIIR improves the image quality and offers a better depiction for vascular structures compared with FBP and HIR. The visibility of the pulmonary emboli was also increased by AIIR.

Comparison of two deep-learning image reconstruction algorithms on cardiac CT images: A phantom study

PURPOSE

The purpose of this study was to compare the performance of Precise IQ Engine (PIQE) and Advanced intelligent Clear-IQ Engine (AiCE) algorithms on image-quality according to the dose level in a cardiac computed tomography (CT) protocol.

MATERIALS AND METHODS

Acquisitions were performed using the CT ACR 464 phantom at three dose levels (volume CT dose indexes: 7.1/5.2/3.1 mGy) using a prospective cardiac CT protocol. Raw data were reconstructed using the three levels of AiCE and PIQE (Mild, Standard and Strong). The noise power spectrum (NPS) and task-based transfer function (TTF) for bone and acrylic inserts were computed. The detectability index (d') was computed to model the detectability of the coronary lumen (350 Hounsfield units and 4-mm diameter) and non-calcified plaque (40 Hounsfield units and 2-mm diameter).

RESULTS

Noise magnitude values were lower with PIQE than with AiCE (-13.4 ± 6.0 [standard deviation (SD)] % for Mild, -20.4 ± 4.0 [SD] % for Standard and -32.6 ± 2.6 [SD] % for Strong levels). The average NPS spatial frequencies shifted towards higher frequencies with PIQE than with AiCE (21.9 ± 3.5 [SD] % for Mild, 20.1 ± 3.0 [SD] % for Standard and 12.5 ± 3.5 [SD] % for Strong levels). The TTF values at fifty percent (f50) values shifted towards higher frequencies with PIQE than with AiCE for acrylic inserts but, for bone inserts, f50 values were found to be close. Whatever the dose and DLR level, d' values of both simulated cardiac lesions were higher with PIQE than with AiCE. For the simulated coronary lumen, d' values were better by 35.1 ± 9.3 (SD) % on average for all dose levels for Mild, 43.2 ± 5.0 (SD) % for Standard, and 62.6 ± 1.2 (SD) % for Strong levels.

CONCLUSION

Compared to AiCE, PIQE reduced noise, improved spatial resolution, noise texture and detectability of simulated cardiac lesions. PIQE seems to have a greater potential for dose reduction in cardiac CT acquisition.

Practical Applications of Artificial Intelligence in Spine Imaging: A Review

Artificial intelligence (AI), a transformative technology with unprecedented potential in medical imaging, can be applied to various spinal pathologies. AI-based approaches may improve imaging efficiency, diagnostic accuracy, and interpretation, which is essential for positive patient outcomes. This review explores AI algorithms, techniques, and applications in spine imaging, highlighting diagnostic impact and challenges with future directions for integrating AI into spine imaging workflow.

A CT deep learning reconstruction algorithm: Image quality evaluation for brain protocol at decreasing dose indexes in comparison with FBP and statistical iterative reconstruction algorithms

PURPOSE

To characterise the impact of Precise Image (PI) deep learning reconstruction algorithm on image quality, compared to filtered back-projection (FBP) and iDose4 iterative reconstruction for brain computed tomography (CT) phantom images.

METHODS

Catphan-600 phantom was acquired with an Incisive CT scanner using a dedicated brain protocol, at six different dose levels (volume computed tomography dose index (CTDIvol): 7/14/29/49/56/67 mGy). Images were reconstructed using FBP, levels 2/5 of iDose4, and PI algorithm (Sharper/Sharp/Standard/Smooth/Smoother). Image quality was assessed by evaluating CT numbers, image histograms, noise, image non-uniformity (NU), noise power spectrum, target transfer function, and detectability index.

RESULTS

The five PI levels did not significantly affect the mean CT number. For a given CTDIvol using Sharper-to-Smoother levels, the spatial resolution for all the investigated materials and the detectability index increased while the noise magnitude decreased, slightly affecting noise texture. For a fixed PI level increasing the CTDIvol the detectability index increased, the noise magnitude decreased. From 29 mGy, NU values converged within 1 Hounsfield Unit from each other without a substantial improvement at higher CTDIvol values.

CONCLUSIONS

The improved performances of intermediate PI levels in brain protocols compared to conventional algorithms seem to suggest a potential reduction of CTDIvol.

In vivo depiction of cortical bone vascularization with ultra-high resolution-CT and deep learning algorithm reconstruction using osteoid osteoma as a model

PURPOSE

The purpose of this study was to evaluate the ability to depict in vivo bone vascularization using ultra-high-resolution (UHR) computed tomography (CT) with deep learning reconstruction (DLR) and hybrid iterative reconstruction algorithm, compared to simulated conventional CT, using osteoid osteoma as a model.

MATERIALS AND METHODS

Patients with histopathologically proven cortical osteoid osteoma who underwent UHR-CT between October 2019 and October 2022 were retrospectively included. Images were acquired with a 1024 × 1024 matrix and reconstructed with DLR and hybrid iterative reconstruction algorithm. To simulate conventional CT, images with a 512 × 512 matrix were also reconstructed. Two radiologists (R1, R2) independently evaluated the number of blood vessels entering the nidus and crossing the bone cortex, as well as vessel identification and image quality with a 5-point scale. Standard deviation (SD) of attenuation in the adjacent muscle and that of air were used as image noise and recorded.

RESULTS

Thirteen patients with 13 osteoid osteomas were included. There were 11 men and two women with a mean age of 21.8 ± 9.1 (SD) years. For both readers, UHR-CT with DLR depicted more nidus vessels (11.5 ± 4.3 [SD] (R1) and 11.9 ± 4.6 [SD] (R2)) and cortical vessels (4 ± 3.8 [SD] and 4.3 ± 4.1 [SD], respectively) than UHR-CT with hybrid iterative reconstruction (10.5 ± 4.3 [SD] and 10.4 ± 4.6 [SD], and 4.1 ± 3.8 [SD] and 4.3 ± 3.8 [SD], respectively) and simulated conventional CT (5.3 ± 2.2 [SD] and 6.4 ± 2.5 [SD], 2 ± 1.2 [SD] and 2.4 ± 1.6 [SD], respectively) (P < 0.05). UHR-CT with DLR provided less image noise than simulated conventional CT and UHR-CT with hybrid iterative reconstruction (P < 0.05). UHR-CT with DLR received the greatest score and simulated conventional CT the lowest score for vessel identification and image quality.

CONCLUSION

UHR-CT with DLR shows less noise than UHR-CT with hybrid iterative reconstruction and significantly improves cortical bone vascularization depiction compared to simulated conventional CT.

Coronary artery assessment on pre transcatheter aortic valve implantation computed tomography may avoid the need for additional coronary angiography

PURPOSE

The purpose of this study was to evaluate the percentage of coronary angiography that can be securely avoided by the interpretation of coronary arteries on pre transcatheter aortic valve implantation CT (TAVI-CT), using CT images obtained with deep-learning reconstruction and motion correction algorithms.

MATERIAL AND METHOD

All consecutive patients who underwent TAVI-CT and coronary angiography, from December 2021 to July 2022 were screened for inclusion in the study. Patients who had previous coronary artery revascularization or who did not undergo TAVI were excluded. All TAVI-CT examinations were obtained using deep-learning reconstruction and motion correction algorithms. On TAVI-CT examinations, quality and stenosis of coronary artery were analyzed retrospectively. When insufficient image quality and/or when diagnosis or doubt of one significant coronary artery stenosis, patients were considered as having possible coronary artery stenosis. The results of coronary angiography were used as the standard of reference for significant CAS.

RESULTS

A total of 206 patients (92 men; mean age, 80.6 years) were included; of these 27/206 (13%) had significant coronary artery stenosis on coronary angiography and were referred for potential revascularization. Sensitivity, specificity, negative predictive value, positive predictive value, and accuracy of TAVI-CT to identify patients requiring coronary artery revascularization was 100% (95% confidence interval [CI]: 87.2-100%), 100% (95% CI: 96.3-100%), 54% (95% CI: 46.6-61.6), 25% (95% CI: 17.0-34.0%) and 60% (95% CI: 53.1-66.9%) respectively. Intra- and inter observer variability was substantial agreement for quality and decision to recommend coronary angiography. Mean reading time was 2 ± 1.2 (standard deviation) min (range: 1-5 min). Overall, TAVI-CT could potentially rule out indication for revascularization for 97 patients (47%).

CONCLUSION

Analysis of coronary artery on TAVI-CT using deep-learning reconstruction and motion correction algorithms can potentially safely avoid coronary angiography in 47% of patients.

Impact of tin filter on the image quality of ultra-low dose chest CT: A phantom study on three CT systems

PURPOSE

The purpose of this study was to assess the impact of a tin filter on the image quality of ultra-low dose (ULD) chest computed tomography (CT) on three different CT systems.

MATERIALS AND METHODS

An image quality phantom was scanned on three CT systems including two split-filter dual-energy CT (SFCT-1 and SFCT-2) scanners and one dual-source CT scanner (DSCT). Acquisitions were performed with a volume CT dose index (CTDIvol) of 0.4 mGy, first at 100 kVp without tin filter (Sn), and second, at Sn100/Sn140 kVp, Sn100/Sn110/Sn120/Sn130/Sn140/Sn150 kVp and Sn100/Sn150 kVp for SFCT-1, SFCT-2 and DSCT respectively. Noise-power-spectrum and task-based transfer function were computed. The detectability index (d') was computed to model the detection of two chest lesions.

RESULTS

For DSCT and SFCT-1, noise magnitude values were higher with 100kVp than with Sn100 kVp and with Sn140 kVp or Sn150 kVp than with Sn100 kVp. For SFCT-2, noise magnitude increased from Sn110 kVp to Sn150 kVp and was higher at Sn100 kVp than at Sn110 kVp. For most kVp with the tin filter, the noise amplitude values were lower than those obtained at 100 kVp. For each CT system, noise texture and spatial resolution values were similar with 100 kVp and with all kVp used with a tin filter. For all simulated chest lesions, the highest d' values were obtained at Sn100 kVp for SFCT-1 and DSCT and at Sn110 kVp for SFCT-2.

CONCLUSION

For ULD chest CT protocols, the lowest noise magnitude and highest detectability values for simulated chest lesions are obtained with Sn100 kVp for the SFCT-1 and DSCT CT systems and at Sn110 kVp for SFCT-2.

First Results of a New Deep Learning Reconstruction Algorithm on Image Quality and Liver Metastasis Conspicuity for Abdominal Low-Dose CT

The study's aim was to assess the impact of a deep learning image reconstruction algorithm (Precise Image; DLR) on image quality and liver metastasis conspicuity compared with an iterative reconstruction algorithm (IR). This retrospective study included all consecutive patients with at least one liver metastasis having been diagnosed between December 2021 and February 2022. Images were reconstructed using level 4 of the IR algorithm (i4) and the Standard/Smooth/Smoother levels of the DLR algorithm. Mean attenuation and standard deviation were measured by placing the ROIs in the fat, muscle, healthy liver, and liver tumor. Two radiologists assessed the image noise and image smoothing, overall image quality, and lesion conspicuity using Likert scales. The study included 30 patients (mean age 70.4 ± 9.8 years, 17 men). The mean CTDI_vol was 6.3 ± 2.1 mGy, and the mean dose-length product 314.7 ± 105.7 mGy.cm. Compared with i4, the HU values were similar in the DLR algorithm at all levels for all tissues studied. For each tissue, the image noise significantly decreased with DLR compared with i4 (p < 0.01) and significantly decreased from Standard to Smooth (-26 ± 10%; p < 0.01) and from Smooth to Smoother (-37 ± 8%; p < 0.01). The subjective image assessment confirmed that the image noise significantly decreased between i4 and DLR (p < 0.01) and from the Standard to Smoother levels (p < 0.01), but the opposite occurred for the image smoothing. The highest scores for overall image quality and conspicuity were found for the Smooth and Smoother levels.

Task-Based Image Quality Assessment Comparing Classical and Iterative Cone Beam CT Images on Halcyon®

BACKGROUND

Despite the development of iterative reconstruction (IR) in diagnostic imaging, CBCT are generally reconstructed with filtered back projection (FBP) in radiotherapy. Varian medical systems, recently released with their latest Halcyon^® V2.0 accelerator, a new IR algorithm for CBCT reconstruction.

PURPOSE

To assess the image quality of radiotherapy CBCT images reconstructed with FBP and an IR algorithm.

METHODS

Three CBCT acquisition modes (head, thorax and pelvis large) available on a Halcyon^® were assessed. Five acquisitions were performed for all modes on an image quality phantom and reconstructed with FBP and IR. Task-based image quality assessment was performed with noise power spectrum (NPS), task-based transfer function (TTF) and detectability index (d'). To illustrate the image quality obtained with both reconstruction types, CBCT acquisitions were made on 6 patients.

RESULTS

The noise magnitude and the spatial frequency of the NPS peak was lower with IR than with FBP for all modes. For all low and high-contrast inserts, the values for TTF at 50% were higher with IR than with FBP. For all inserts and all modes, the contrast values were similar with FBP and IR. For all low and high-contrast simulated lesions, d' values were higher with IR than with FBP for all modes. These results were also found on the 6 patients where the images were less noisy but smoother with IR-CBCT.

CONCLUSIONS

Using the IR algorithm for CBCT images in radiotherapy improve image quality and thus could increase the accuracy of online registration and limit positioning errors during processing.

Spectral CT imaging: Technical principles of dual-energy CT and multi-energy photon-counting CT

Spectral computed tomography (CT) imaging encompasses a unique generation of CT systems based on a simple principle that makes use of the energy-dependent information present in CT images. Over the past two decades this principle has been expanded with the introduction of dual-energy CT systems. The first generation of spectral CT systems, represented either by dual-source or dual-layer technology, opened up a new imaging approach in the radiology community with their ability to overcome the limitations of tissue characterization encountered with conventional CT. Its expansion worldwide can also be considered as an important leverage for the recent groundbreaking technology based on a new chain of detection available on photon counting CT systems, which holds great promise for extending CT towards multi-energy CT imaging. The purpose of this article was to detail the basic principles and techniques of spectral CT with a particular emphasis on the newest technical developments of dual-energy and multi-energy CT systems.