The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence

Anti-Scattering Perovskite Scintillator Arrays for High-Resolution Computed Tomography Imaging

Computed tomography (CT) imaging has emerge as an effective medical diagnostic technique due to its rapid and 3D imaging capabilities, often employing indirect imaging methods through scintillator materials. Arraying scintillators that can confine light scattering to enable high-resolution CT imaging remains an area of ongoing exploration for emerging perovskite scintillators. Here an anti-scattering cesium lead bromide (CsPbBr3) scintillator array embedded within a polyurethane acrylate matrix for CT imaging using a cost-effective solution-processed method is reported. Due to the large refractive index contrast between the scintillator and matrix, photon propagation can be well confined within the CsPbBr3 scintillator array to significantly suppress the light scattering and enhance the light collection efficiency by nearly two times compared to the monolithic film. Furthermore, the scintillator array exhibits low-dosage and high-resolution CT imaging capability by reconstructing a 3D tooth image with a good spatial resolution of 20.1 lp cm-1 at a low effective dose of 0.22 mSv. This work highlights that the CsPbBr3 scintillator array is a highly promising candidate for CT imaging.

Impact of Combined Deep Learning Image Reconstruction and Metal Artifact Reduction Algorithm on CT Image Quality in Different Scanning Conditions for Maxillofacial Region with Metal Implants: A Phantom Study

This study aims to investigate the impact of combining deep learning image reconstruction (DLIR) and metal artifacts reduction (MAR) algorithms on the quality of CT images with metal implants under different scanning conditions. Four images of the maxillofacial region in pigs were taken using different metal implants for evaluation. The scans were conducted at three different dose levels (CTDIvol: 20/10/5 mGy). The images were reconstructed using three different methods: filtered back projection (FBP), adaptive statistical iterative reconstruction with Veo at a 50% level (AV50), and DLIR at three levels (low, medium, and high). Regions of interest (ROIs) were identified in various tissues (near/far/reference fat, muscle, bone) both with and without metal implants and artifacts. Parameters such as standard deviation (SD), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and metal artifact index (MAI) were calculated. Additionally, two experienced radiologists evaluated the subjective image quality (IQ) using a 5-point Likert scale. (1) Both observers rated MAR generated significantly lower artifact scores than non-MAR in all types of tissues (P < 0.01), except for the far shadow and bloom in bone (phantoms 1, 3, 4) and the far bloom in muscle (phantom 3) without significant differences (P = 1.0). (2) Under the same scanning condition, DLIR at three levels produced a smaller SD than those of FBP and AV50 (P < 0.05). (3) Compared to FBP and AV50, DLIR denoted a better reduction of MAI and improvement of SNR and CNR (P < 0.05) for most tissues between the four phantoms. (4) Subjective overall IQ was superior with the increasement of DLIR level (P < 0.05) and both observers agreed that DLIR produced better artifact reductions compared with FBP and AV50. The combination of DLIR and MAR algorithms can enhance image quality, significantly reduce metal artifacts, and offer high clinical value.

Impact of reconstruction techniques on low dose chest CT image quality: comparison of FBP, Clear View at Mzuzu Central Hospital, Malawi

Objective

To investigate the impact of two reconstruction techniques, Filtered Back Projection (FBP) and Clear View (CV) iterative algorithm, on the image quality of low-dose thin-slice chest CT.

Methods

A retrospective study of 42 patients undergoing low-dose chest CT at Mzuzu Central Hospital from Feb-Apr 2024 used automatic tube current modulation at 120 kV Raw data were reconstructed with FBP, 20% CV, 40% CV, 60% CV, and 80% CV, with 1 mm slice thickness and 0.625 mm spacing. Image noise, Signal-to-Noise Ratio (SNR), and Contrast-to-Noise Ratio (CNR) were measured, and image quality was rated on a 5-point scale for lung and mediastinal windows. Qualitative and quantitative parameters of the two different reconstruction algorithms in the five groups were comparatively analyzed.

Results

(1) Objective evaluation showed noise decreased in lung parenchyma, aorta, and erector spinae muscle with increasing CV weight. Mean noise reductions in lung parenchyma were 23.34% and 27.69% in 60% CV and 80% CV (P < 0.05). Aorta noise decreased by 23.43%, 37.16%, and 46.18% in 40% CV, 60% CV, and 80% CV (P < 0.05, P < 0.001, P < 0.001). Erector spinae muscle noise decreased by 35.91% and 44.78% in 60% CV and 80% CV (P < 0.05, P < 0.001). SNR and CNR were higher in CV groups than FBP. Among them, the differences in SNR between the 60% CV and 80% CV groups and the FBP group were statistically significant (P < 0.05). (2) Subjective scores for all groups were > 3, meeting diagnostic standards, with 60% CV yielding the highest lung and mediastinal window image quality (P < 0.05).

Conclusion

Compared to FBP, CV iterative reconstruction reduces noise and improves chest CT image quality under low-dose conditions as the weight increases, with 60% CV showing optimal performance.

Enhancement of abdominal Low-Dose CT image quality utilizing Clear View reconstruction technique at Mzuzu Central Hospital, Malawi

Objective

This study aimed to investigate the impact of Clear View dual-domain iterative reconstruction (IR) technology on the quality of low-dose abdominal CT images and to determine the optimal weight ratio to optimize image quality.

Methods

We studied 40 patients (28 males, 12 females, aged 19-69) undergoing low-dose abdominal CT scans (CTDI = 5.32 ± 0.89 mGy). The scanning parameters were set as follows: tube voltage of 120 kVp, tube current modulation based on Signal to Noise Ratio (SNR) at 0.5 mode (O-Dose automatic tube current modulation technology), pitch of 0.9, rotation time of 0.6 s/r, matrix size of 512 × 512, and collimation width of 16 × 1.25 mm. We applied Clear View IR with four weight ratios (20%, 40%, 60%, 80%) and filtered back projection (FBP). Conventional scanning uses with 120 kVp, 280 mAs, pitch of 0.9, rotation time of 0.6 s/r, matrix size of 512 × 512, and collimation width of 16 × 1.25 mm. Conventional dose abdominal CT scans (CTDI = 11.95 ± 0.00 mGy).CT values, standard deviations (SD), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured for liver, spleen, pancreas, kidneys, and erector spinae muscles. Two deputy chief physicians blindly evaluated image quality on a 1-5 scale. Statistical analysis was done using SPSS 22.0 with P < 0.05 considered significant.

Results

Subjective evaluations revealed the highest diagnostic score with a 40% Clear View reconstruction weight ratio. Higher weight ratios significantly reduced subjective image noise, with the highest noise scores at 80%. Moreover, compared to FBP, especially Clear View reconstruction weight ratios of 20% to 60%, significantly improved the image quality of abdominal solid organs, reducing image artifacts and improving diagnostic acceptability (P < 0.05). Objective evaluation showed that with increasing Clear View reconstruction weight ratios, image noise SD values decreased, while SNR and CNR values increased, and the differences in SD, SNR, and CNR for different reconstruction weight ratios of abdominal solid organs were statistically significant (P < 0.05).

Conclusion

Compared to FBP algorithm, Clear View demonstrates greater potential in low-dose abdominal CT, effectively reducing image noise and artifacts while maintaining image clarity. Based on combined subjective and objective evaluations, a 40% Clear View reconstruction weight ratio provides optimal image quality for abdominal solid organs.

Photon-counting CT systems: A technical review of current clinical possibilities

In recent years, computed tomography (CT) has undergone a number of developments to improve radiological care. The most recent major innovation has been the development of photon-counting detectors. By comparison with the energy-integrating detectors traditionally used in CT, these detectors offer better dose efficiency, eliminate electronic noise, improve spatial resolution and have intrinsic spectral sensitivity. These detectors also allow the energy of each photon to be counted, thus improving the sampling of the X-ray spectrum in multiple energy bins, to better distinguish between photoelectric and Compton attenuation coefficients, resulting in better spectral images and specific color K-edge images. The purpose of this article was to make the reader more familiar with the basic principles and techniques of new photon-counting CT systems equipped with photon-counting detectors and also to describe the currently available devices that could be used in clinical practice.

Challenges and opportunities for high-quality battery production at scale

As the world electrifies, global battery production is expected to surge. However, batteries are both difficult to produce at the gigawatt-hour scale and sensitive to minor manufacturing variation. As a result, the battery industry has already experienced both highly-visible safety incidents and under-the-radar reliability issues-a trend that will only worsen if left unaddressed. Here we highlight both the challenges and opportunities to enable battery quality at scale. We first describe the interplay between various battery failure modes and their numerous root causes. We then discuss how to manage and improve battery quality during production. We hope our perspective brings greater visibility to the battery quality challenge to enable safe global electrification.

Image quality assessment of artificial intelligence iterative reconstruction for low dose unenhanced abdomen: comparison with hybrid iterative reconstruction

OBJECTIVES

To assess the impact of artificial intelligence iterative reconstruction algorithms (AIIR) on image quality with phantom and clinical studies.

METHODS

The phantom images were reconstructed with the hybrid iterative algorithm (HIR: Karl 3D-3, 5, 7, 9) and AIIR (grades 1-5) algorithm. Noise power spectra (NPS), task transfer functions (TTF) were measured, and additionally sharpness was assessed using a "blur metric" procedure. Sixty-two consecutive patients underwent standard-dose and low-dose unenhanced abdominal computed tomography (CT) scans, i.e., SDCT and LDCT groups, respectively. The SDCT images reconstructed using the Karl 3D-5, and the LDCT images reconstructed using the Karl 3D-5 and the AIIR-3 and 5, respectively. CT values, standard deviation (SD), signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were assessed for hepatic parenchyma and paravertebral muscles. Images were independently evaluated by two radiologists for image-quality, noise, sharpness, and lesion diagnostic confidence.

RESULTS

In the phantom study, AIIR algorithm provided higher TTF50% and NPS average spatial frequency compared to HIR. In the clinical study, there was no statistically significant difference in CT values among the four reconstruction images (p > 0.05). The LDCT group AIIR-3 obtained the lowest SD values and the highest mean CNR and SNR values compared to the other three groups (p < 0.05). For qualitative assessment, the image subjective characteristic scores of AIIR-5 in the LDCT group, compared with the SDCT group, were not statistically significant (p > 0.05).

CONCLUSIONS

AIIR reduces radiation dose levels by approximately 78% and still maintains the image quality of unenhanced abdominal CT compared to HIR with SDCT.

THE TRIAL REGISTRATION NUMBER

NCT06142539.

Tin prefiltration in computed tomography does not significantly alter radiation-induced gene expression and DNA double-strand break formation

BACKGROUND

The tin (Sn) prefilter technique is a recently introduced dose-saving technique in computed tomography (CT). This study investigates whether there is an altered molecular biological response in blood cells using the tin prefiltering technique.

METHODS

Blood from 6 donors was X-irradiated ex-vivo with 20 mGy full dose (FD) protocols (Sn 150 kV, 150 kV, and 120 kV) and a tin prefiltered 16.5 mGy low dose (LD) protocol on a CT scanner. Biological changes were determined by quantification of γH2AX DNA double-strand break (DSB) foci, and differential gene expression (DGE) relative to unexposed samples were examined for seven known radiation-induced genes (FDXR, DDB2, BAX, CDKN1A, AEN, EDA2R, APOBEC3H) and 667 microRNAs (miRNA).

RESULTS

EDA2R and DDB2 gene expression (GE) increased 1.7-6-fold (p = 0.0004-0.02) and average DNA DSB foci value (0.31±0.02, p<0.0001) increased significantly relative to unexposed samples, but similarly for the applied radiation protocols. FDXR upregulation (2.2-fold) was significant for FD protocols (p = 0.01-0.02) relative to unexposed samples. miRNA GE changes were not significant (p = 0.15-1.00) and DGE were similar for the examined protocols (p = 0.10-1.00). An increased frequency of lower DGE values was seen in the Sn 150 kV LD protocol compared to the 120 kV FD and Sn 150 kV FD protocols (p = 0.001-0.008).

CONCLUSIONS

The current ex-vivo study indicates no changes regarding transcriptional and post-transcriptional DGE and DNA DSB induction when using the tin prefilter technique and even a significant tendency to lower radiation-induced DGE-changes due to the dose reduction of the tin prefilter with equal image quality compared to classical CT scan protocols was found.

RBP-DIP: Residual back projection with deep image prior for ill-posed CT reconstruction

The success of deep image prior (DIP) in a number of image processing tasks has motivated their application in image reconstruction problems in computed tomography (CT). In this paper, we introduce a residual back projection technique (RBP) that improves the performance of deep image prior framework in iterative CT reconstruction, especially when the reconstruction problem is highly ill-posed. The RBP-DIP framework uses an untrained U-net in conjunction with a novel residual back projection connection to minimize the objective function while improving reconstruction accuracy. In each iteration, the weights of the untrained U-net are optimized, and the output of the U-net in the current iteration is used to update the input of the U-net in the next iteration through the proposed RBP connection. The introduction of the RBP connection strengthens the regularization effects of the DIP framework in the context of iterative CT reconstruction leading to improvements in accuracy. Our experiments demonstrate that the RBP-DIP framework offers improvements over other state-of-the-art conventional IR methods, as well as pre-trained and untrained models with similar network structures under multiple conditions. These improvements are particularly significant in the few-view and limited-angle CT reconstructions, where the corresponding inverse problems are highly ill-posed and the training data is limited. Furthermore, RBP-DIP has the potential for further improvement. Most existing IR algorithms, pre-trained models, and enhancements applicable to the original DIP algorithm can also be integrated into the RBP-DIP framework.

Deep-learning reconstruction enhances image quality of Adamkiewicz Artery in low-keV dual-energy CT

BACKGROUND

Low-keV virtual monoenergetic images (VMIs) of dual-energy computed tomography (CT) enhances iodine contrast for detecting small arteries like the Adamkiewicz artery (AKA), but image noise can be problematic. Deep-learning image reconstruction (DLIR) effectively reduces noise without sacrificing image quality.

PURPOSE

To evaluate whether DLIR on low-keV VMIs of dual-energy CT scans improves the visualization of the AKA.

MATERIAL AND METHODS

We enrolled 29 patients who underwent CT angiography before aortic repair. VMIs obtained at 70 and 40 keV were reconstructed using hybrid iterative reconstruction (HIR), and 40 keV VMIs were reconstructed using DLIR. The image noise of the spinal cord, the maximum CT values of the anterior spinal artery (ASA), and the contrast-to-noise ratio (CNR) of the ASA were compared. The overall image quality and the delineation of the AKA were evaluated on a 4-point score (1 = poor, 4 = excellent).

RESULTS

The mean image noise of the spinal cord was significantly lower on 40-keV DLIR than on 40-keV HIR scans; they were significantly higher than on 70-keV HIR images. The CNR of the ASA was highest on the 40-keV DLIR images among the three reconstruction images. The mean image quality scores for 40-keV DLIR and 70-keV HIR scans were comparable, and higher than of 40-keV HIR images. The mean delineation scores for 40-keV HIR and 40-keV DLIR scans were significantly higher than for 70-keV HIR images.

CONCLUSION

Visualization of the AKA was significantly better on low-keV VMIs subjected to DLIR than conventional HIR images.

Combined influence of quantum iterative reconstruction level and kernel sharpness on image quality in photon counting CT angiography of the upper leg

Aim was to evaluate the influence of different quantum iterative reconstruction (QIR) levels on the image quality of femoral photon-counting CT angiographies (PCD-CTA).Ultra-high resolution PCD-CTA were obtained from both extremities of five extracorporeally-perfused cadavers using constant tube voltage and maximum radiation dose (71.2 ± 11.0 mGy). Images were reconstructed with three kernels (Bv48, Bv60, Bv76) and the four available levels of QIR. Signal attenuation in the arterial lumen, muscle, and fat were measured. Contrast-to-noise ratios (CNR) and blurring scores were calculated for objective assessment. Six radiologists evaluated the subjective image quality using a pairwise comparison tool.Higher QIR level resulted in a decisive image noise reduction, especially with sharper convolution kernels (Bv60: Q1 11.5 ± 6.3 HU vs. Q4 8.4 ± 2.6 HU; p < 0.001). Largest improvement of CNR was recorded with ultra-sharp reconstructions (Bv76: Q1 20.2 ± 4.4 vs. Q4 28.0 ± 3.5; p < 0.001). Blurring decreased with higher QIR levels for soft Bv48, remained constant for medium Bv60, and increased for sharp Bv76 reconstructions. Subjective QIR level preference varied kernel depending, preferred combinations were: Bv48/Q4, Bv60/Q2, Bv76/Q3. Interrater agreement was excellent.Sharp kernels benefited most from noise reduction of higher QIR levels in lower extremity PCD-CTA. In sum, QIR level 3 provided the best objective and subjective image quality results.

Tube current reduction and iterative image reconstruction for computed tomography myelography

This study aimed to systematically evaluate the impact of a low-dose (LD) protocol using tube current reduction on image quality, the confidence for intervention planning and guidance, and diagnostic yield for computed tomography (CT) myelography. We retrospectively analyzed 68 patients who underwent CT myelography, with 34 investigations performed with a standard-dose (SD) and 34 investigations performed with a LD protocol (using tube current reduction). The different scans were matched considering variables such as sex, age, presence of spinal instrumentation, and body diameter. All images were evaluated by two readers (R1 and R2) using Likert scales. Image noise was measured using attenuation values of paraspinal muscle tissue. Images were reconstructed with model-based iterative reconstruction (post-myelography diagnostic scans) or hybrid reconstruction (planning, periprocedural, and diagnostic scans). Image quality, overall artifacts, image contrast, and confidence for planning or intervention guidance were rated good to perfect for both SD and LD scans according to evaluations of both readers. Inter-reader agreement was good to very good for the images from intervention planning (κ ≥ 0.80) as well as for intervention guidance (κ ≥ 0.77), as well as for diagnostic scans (κ ≥ 0.85). Image noise was similar between SD and LD scans performed for planning of the interventional procedures (model-based iterative reconstruction: SD 45.37 ± 7.29 HU vs. LD 45.17 ± 9.12 HU; hybrid reconstruction: SD 46.05 ± 7.43 HU vs. LD 45.05 ± 8.69 HU; p > 0.05). The volume-weighted CT dose index (CTDIvol) and size-specific dose estimate (SSDE) were significantly lower for the planning scans as well as the periprocedural scans when using the LD protocol as compared to the SD protocol (p < 0.05). In conclusion, implementation of a LD protocol with tube current reduction for CT myelography is a feasible option to reduce radiation exposure, especially when combined with iterative image reconstruction. In our study, LD imaging did not have a relevant negative impact on image quality, confidence for intervention planning or guidance, or diagnostic certainty for CT myelography.

Artificial intelligence in endodontics: Fundamental principles, workflow, and tasks

The integration of artificial intelligence (AI) in healthcare has seen significant advancements, particularly in areas requiring image interpretation. Endodontics, a specialty within dentistry, stands to benefit immensely from AI applications, especially in interpreting radiographic images. However, there is a knowledge gap among endodontists regarding the fundamentals of machine learning and deep learning, hindering the full utilization of AI in this field. This narrative review aims to: (A) elaborate on the basic principles of machine learning and deep learning and present the basics of neural network architectures; (B) explain the workflow for developing AI solutions, from data collection through clinical integration; (C) discuss specific AI tasks and applications relevant to endodontic diagnosis and treatment. The article shows that AI offers diverse practical applications in endodontics. Computer vision methods help analyse images while natural language processing extracts insights from text. With robust validation, these techniques can enhance diagnosis, treatment planning, education, and patient care. In conclusion, AI holds significant potential to benefit endodontic research, practice, and education. Successful integration requires an evolving partnership between clinicians, computer scientists, and industry.

Effect of Model-Based Iterative Reconstruction on Image Quality of Chest Computed Tomography for COVID-19 Pneumonia

PURPOSE

This study aimed to compare the image quality of chest computed tomography (CT) scans for COVID-19 pneumonia using forward-projected model-based iterative reconstruction solution-LUNG (FIRST-LUNG) with filtered back projection (FBP) and hybrid iterative reconstruction (HIR).

METHOD

The CT images of 44 inpatients diagnosed with COVID-19 pneumonia between December 2022 and June 2023 were retrospectively analyzed. The CT images were reconstructed using FBP, HIR, and FIRST-LUNG-MILD/STANDARD/STRONG. The CT values and noise of the lumen of the main trachea and erector spine muscle were measured for each group. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. Subjective evaluations included overall image quality, noise, streak artifact, visualization of normal lung structures, and abnormal CT features. One-way analysis of variance was used to compare the objective and subjective indicators among the five groups. The task-based transfer function was derived for three distinct contrasts representing anatomical structures, lower-contrast lesion, and higher-contrast lesion.

RESULTS

The results of the study demonstrated significant differences in image noise, SNR, and CNR among the five groups ( P < 0.001). The FBP images exhibited the highest levels of noise and the lowest SNR and CNR among the five groups ( P < 0.001). When compared to the FBP and HIR groups, the noise was lower in the FIRST-LUNG-MILD/STANDARD/STRONG group, while the SNR and CNR were higher ( P < 0.001). The subjective overall image quality score of FIRST-LUNG-MILD/STANDARD was significantly better than FBP and FIRST-LUNG-STRONG ( P < 0.001). FIRST-LUNG-MILD was superior to FBP, HIR, FIRST-LUNG-STANDARD, and FIRST-LUNG-STRONG in visualizing proximal and peripheral bronchovascular and subpleural vessels ( P < 0.05). Additionally, FIRST-LUNG-MILD achieved the best scores in evaluating abnormal lung structure ( P < 0.001). The overall interobserver agreement was substantial (intraclass correlation coefficient = 0.891). The task-based transfer function 50% values of FIRST reconstructions are consistently higher compared to FBP and HIR.

CONCLUSIONS

The FIRST-LUNG-MILD/STANDARD algorithm can enhance the image quality of chest CT in patients with COVID-19 pneumonia, while preserving important details of the lesions, better than the FBP and HIR algorithms. After evaluating various COVID-19 pneumonia lesions and considering the improvement in image quality, we recommend using the FIRST-LUNG-MILD reconstruction for diagnosing COVID-19 pneumonia.

Multi-Channel Optimization Generative Model for Stable Ultra-Sparse-View CT Reconstruction

Score-based generative model (SGM) has risen to prominence in sparse-view CT reconstruction due to its impressive generation capability. The consistency of data is crucial in guiding the reconstruction process in SGM-based reconstruction methods. However, the existing data consistency policy exhibits certain limitations. Firstly, it employs partial data from the reconstructed image of the iteration process for image updates, which leads to secondary artifacts with compromising image quality. Moreover, the updates to the SGM and data consistency are considered as distinct stages, disregarding their interdependent relationship. Additionally, the reference image used to compute gradients in the reconstruction process is derived from the intermediate result rather than ground truth. Motivated by the fact that a typical SGM yields distinct outcomes with different random noise inputs, we propose a Multi-channel Optimization Generative Model (MOGM) for stable ultra-sparse-view CT reconstruction by integrating a novel data consistency term into the stochastic differential equation model. Notably, the unique aspect of this data consistency component is its exclusive reliance on original data for effectively confining generation outcomes. Furthermore, we pioneer an inference strategy that traces back from the current iteration result to ground truth, enhancing reconstruction stability through foundational theoretical support. We also establish a multi-channel optimization reconstruction framework, where conventional iterative techniques are employed to seek the reconstruction solution. Quantitative and qualitative assessments on 23 views datasets from numerical simulation, clinical cardiac and sheep's lung underscore the superiority of MOGM over alternative methods. Reconstructing from just 10 and 7 views, our method consistently demonstrates exceptional performance.

[Healthcare 4.0-Medicine in transition]

Healthcare 4.0 describes the future transformation of the healthcare sector driven by the combination of digital technologies, such as artificial intelligence (AI), big data and the Internet of Medical Things, enabling the advancement of precision medicine. This overview article addresses various areas such as large language models (LLM), diagnostics and robotics, shedding light on the positive aspects of Healthcare 4.0 and showcasing exciting methods and application examples in cardiology. It delves into the broad knowledge base and enormous potential of LLMs, highlighting their immediate benefits as digital assistants or for administrative tasks. In diagnostics, the increasing usefulness of wearables is emphasized and an AI for predicting heart filling pressures based on cardiac magnetic resonance imaging (MRI) is introduced. Additionally, it discusses the revolutionary methodology of a digital simulation of the physical heart (digital twin). Finally, it addresses both regulatory frameworks and a brief vision of data-driven healthcare delivery, explaining the need for investments in technical personnel and infrastructure to achieve a more effective medicine.

[Artificial intelligence in diagnostic radiology for dose management : Advances and perspectives using the example of computed tomography]

CLINICAL-METHODOLOGICAL PROBLEM

Imaging procedures employing ionizing radiation require compliance with European directives and national regulations in order to protect patients. Each exposure must be indicated, individually adapted, and documented. Unacceptable dose exceedances must be detected and reported. These tasks are time-consuming and require meticulous diligence.

STANDARD RADIOLOGICAL METHODS

Computed tomography (CT) is the most important contributor to medical radiation exposure. Optimizing the patient's dose is therefore mandatory. Use of modern technology and reconstruction algorithms already reduces exposure. Checking the indication, planning, and performing the examination are further important process steps with regard to radiation protection. Patient exposure is usually monitored by dose management systems (DMS). In special cases, a risk assessment is required by calculating the organ doses.

METHODOLOGICAL INNOVATIONS

Artificial intelligence (AI)-assisted techniques are increasingly used in various steps of the process: they support examination planning, improve patient positioning, and enable automated scan length adjustments. They also provide real-time estimates of individual organ doses.

EVALUATION

The integration of AI into medical imaging is proving successful in terms of dose optimization in various areas of the radiological workflow, from reconstruction to examination planning and performing exams. However, the use of AI in conjunction with DMS has not yet been considered on a large scale.

PRACTICAL RECOMMENDATION

AI processes offer promising tools to support dose management. However, their implementation in the clinical setting requires further research, extensive validation, and continuous monitoring.

Pulmonary nodule visualization and evaluation of AI-based detection at various ultra-low-dose levels using photon-counting detector CT

BACKGROUND

Radiation dose should be as low as reasonably achievable. With the invention of photon-counting detector computed tomography (PCD-CT), the radiation dose may be considerably reduced.

PURPOSE

To evaluate the potential of PCD-CT for dose reduction in pulmonary nodule visualization for human readers as well as for computer-aided detection (CAD) studies.

MATERIAL AND METHODS

A chest phantom containing pulmonary nodules of different sizes/densities (range 3-12 mm and -800-100 HU) was scanned on a PCD-CT with standard low-dose protocol as well as with half, quarter, and 1/40 dose (CTDIvol 0.4-0.03 mGy). Dose-matched scans were performed on a third-generation energy-integrating detector CT (EID-CT). Evaluation of nodule visualization and detectability was performed by two blinded radiologists. Subjective image quality was rated on a 5-point Likert scale. Artificial intelligence (AI)-based nodule detection was performed using commercially available software.

RESULTS

Highest image noise was found at the lowest dose setting of 1/40 radiation dose (eff. dose = 0.01mSv) with 166.1 ± 18.5 HU for PCD-CT and 351.8 ± 53.0 HU for EID-CT. Overall sensitivity was 100% versus 93% at standard low-dose protocol (eff. dose = 0.2 mSv) for PCD-CT and EID-CT, respectively. At the half radiation dose, sensitivity remained 100% for human reader and CAD studies in PCD-CT. At the quarter radiation dose, PCD-CT achieved the same results as EID-CT at the standard radiation dose setting (93%, P = 1.00) in human reading studies. The AI-CAD system delivered a sensitivity of 93% at the lowest radiation dose level in PCD-CT.

CONCLUSION

At half dose, PCD CT showed pulmonary nodules similar to full-dose PCD, and at quarter dose, PCD CT performed comparably to standard low-dose EID CT. The CAD algorithm is effective even at ultra-low doses.

Prognostic performance of gray-white matter ratio in adult out-of-hospital cardiac arrest patients after receiving extracorporeal cardiopulmonary resuscitation

BACKGROUND

Gray-to-white matter ratio (GWR), measured by computed tomography (CT), is commonly used to predict poor neurological outcomes after out-of-hospital cardiac arrest (OHCA). The prognostic performance of GWR in OHCA patients receiving extracorporeal cardiopulmonary resuscitation (ECPR) is not known.

METHODS

This study is a secondary analysis of data from the SAVE-J II registry, a retrospective, multicenter study. Participants were divided into four groups according to average GWR (aGWR) values ranging from 1.00 to 1.39, separated by 0.1 intervals. The aGWR values were calculated for bilateral basal ganglia, centrum semiovale, and high convexity obtained by head CT within 24 h after ECPR. Primary outcome was poor neurological outcomes at 30-day.

RESULTS

In total, 1,146 OHCA patients treated with ECPR were included in our analysis. Overall, participants with lower aGWR more likely had poor neurological outcomes, aGWR 1.00-1.09 (94.6%), aGWR 1.10-1-19 (87.8%), aGWR 1.20-1.29 (78.5%), and aGWR 1.30-1.39 (70.3%). Multivariable logistic regression showed that lower aGWR was associated with poor neurological outcome at 30-day, aGWR 1.30-1.39: reference, aGWR 1.00-1.09: adjusted odds ratio (aOR) 10.01 (95% confidence interval (CI) [3.58-27.99]), aGWR 1.10-1.19: aOR 4.83 (95% CI [2.31-10.12]), aGWR 1.20-1.29: aOR 2.16 (95% CI [1.02-4.55]). Receiver operating characteristic curve analysis revealed that the prognostic performance of aGWR had an area under the curve of 0.628, 95% CI [0.59-0.66]). The aGWR threshold of 1.005 for predicting poor neurological outcome reached 100% specificity with 0.1% sensitivity.

CONCLUSION

Early neuro-prognostication depending on GWR may not be sufficient after ECPR and requires a multimodal approach.

Abdominal image quality and dose reduction with energy-integrating or photon-counting detectors dual-source CT: A phantom study

PURPOSE

The purpose of this study was to assess image-quality and dose reduction potential using a photon-counting computed tomography (PCCT) system by comparison with two different dual-source CT (DSCT) systems using two phantoms.

MATERIALS AND METHODS

Acquisitions on phantoms were performed using two DSCT systems (DSCT1 [Somatom Force] and DSCT2 [Somatom Pro.Pulse]) and one PCCT system (Naeotom Alpha) at four dose levels (13/6/3.4/1.8 mGy). Noise power spectrum (NPS) and task-based transfer function (TTF) were computed to assess noise magnitude and noise texture and spatial resolution (f50), respectively. Detectability indexes (d') were computed to model the detection of abdominal lesions: one unenhanced high-contrast task, one contrast-enhanced high-contrast task and one unenhanced low-contrast task. Image quality was subjectively assessed on an anthropomorphic phantom by two radiologists.

RESULTS

For all dose levels, noise magnitude values were lower with PCCT than with DSCTs. For all CT systems, similar noise texture values were found at 13 and 6 mGy, but the greatest noise texture values were found for DSCT2 and the lowest for PCCT at 3.4 and 1.8 mGy. For high-contrast inserts, similar or lower f50 values were found with PCCT than with DSCT1 and the opposite pattern was found for the low-contrast insert. For the three simulated lesions, d' values were greater with PCCT than with DSCTs. Abdominal images were rated satisfactory for clinical use by the radiologists for all dose levels with PCCT and for 13 and 6 mGy with DSCTs.

CONCLUSION

By comparison with DSCTs, PCCT reduces image-noise and improves detectability of simulated abdominal lesions without altering the spatial resolution and image texture. Image-quality obtained with PCCT seem to indicate greater potential for dose optimization than those obtained with DSCTs.

Radiation Detectors and Sensors in Medical Imaging

Medical imaging instrumentation design and construction is based on radiation sources and radiation detectors/sensors. This review focuses on the detectors and sensors of medical imaging systems. These systems are subdivided into various categories depending on their structure, the type of radiation they capture, how the radiation is measured, how the images are formed, and the medical goals they serve. Related to medical goals, detectors fall into two major areas: (i) anatomical imaging, which mainly concerns the techniques of diagnostic radiology, and (ii) functional-molecular imaging, which mainly concerns nuclear medicine. An important parameter in the evaluation of the detectors is the combination of the quality of the diagnostic result they offer and the burden of the patient with radiation dose. The latter has to be minimized; thus, the input signal (radiation photon flux) must be kept at low levels. For this reason, the detective quantum efficiency (DQE), expressing signal-to-noise ratio transfer through an imaging system, is of primary importance. In diagnostic radiology, image quality is better than in nuclear medicine; however, in most cases, the dose is higher. On the other hand, nuclear medicine focuses on the detection of functional findings and not on the accurate spatial determination of anatomical data. Detectors are integrated into projection or tomographic imaging systems and are based on the use of scintillators with optical sensors, photoconductors, or semiconductors. Analysis and modeling of such systems can be performed employing theoretical models developed in the framework of cascaded linear systems analysis (LCSA), as well as within the signal detection theory (SDT) and information theory.

Fabrication of ZnO Nanowire Cold Cathode Flat-Panel X-ray Source with a Reflective Anode

A novel reflective anode flat-panel X-ray source using ZnO nanowire cold cathode and a metal anode has been developed. Simulation analysis indicated that the reflective anode structure reduces electric field concentration compared to a transmission anode structure. The current-voltage characteristics, X-ray radiation dose rate, and stability of the fabricated device were thoroughly characterized. The device demonstrated a maximum emission current of 481.1 μA and a maximum radiation dose rate of 303 μGy/s at an anode voltage of 40 kV. The X-ray imaging of various objects was also conducted. Our findings are of significance for developing high-performance, robust flat-panel X-ray sources for diverse applications.

Evolving Radiological Approaches in the Diagnosis and Monitoring of Arachnoiditis Ossificans

Arachnoiditis ossificans (AO) is a rare and complex neurological condition characterized by pathological calcification or ossification of the arachnoid membrane. Arachnoiditis ranks as the third most frequent cause of failed back surgery syndrome (FBSS). This narrative review explores the evolving radiological approaches in its diagnosis and monitoring. The historical perspective traces the progression from plain radiographs to advanced imaging techniques. Current radiological modalities, including X-ray, computed tomography (CT), and magnetic resonance imaging (MRI), are discussed, highlighting their respective roles, advantages, and limitations. Emerging and advanced imaging modalities, such as high-resolution CT, 3T and 7T MRI, and PET/CT or PET/MRI, are examined for their potential to enhance diagnostic accuracy and monitoring capabilities. A comparative analysis of these imaging modalities considers their sensitivity, specificity, cost-effectiveness, and radiation exposure implications. The review also explores the crucial role of imaging in disease monitoring and treatment planning, including follow-up protocols, evaluation of disease progression, and guidance for interventional procedures. Future directions in the field are discussed, focusing on promising research areas, the potential of artificial intelligence and machine learning in image analysis, and identified gaps in current knowledge. The review emphasizes the importance of a multimodal imaging approach and the need for standardized protocols. It concludes that while significant advancements have been made, further research is necessary to fully understand the correlation between imaging findings and clinical outcomes. The continued evolution of radiological approaches is expected to significantly improve patient care and outcomes in AO.

Applying Super-Resolution and Tomography Concepts to Identify Receptive Field Subunits in the Retina

Spatially nonlinear stimulus integration by retinal ganglion cells lies at the heart of various computations performed by the retina. It arises from the nonlinear transmission of signals that ganglion cells receive from bipolar cells, which thereby constitute functional subunits within a ganglion cell's receptive field. Inferring these subunits from recorded ganglion cell activity promises a new avenue for studying the functional architecture of the retina. This calls for efficient methods, which leave sufficient experimental time to leverage the acquired knowledge for further investigating identified subunits. Here, we combine concepts from super-resolution microscopy and computed tomography and introduce super-resolved tomographic reconstruction (STR) as a technique to efficiently stimulate and locate receptive field subunits. Simulations demonstrate that this approach can reliably identify subunits across a wide range of model variations, and application in recordings of primate parasol ganglion cells validates the experimental feasibility. STR can potentially reveal comprehensive subunit layouts within only a few tens of minutes of recording time, making it ideal for online analysis and closed-loop investigations of receptive field substructure in retina recordings.

Image quality and radiation doses in abdominal CT: A multicenter study

PURPOSE

To benchmark image quality and corresponding radiation doses for acute abdominal CT examination across different laboratories and CT manufacturers.

METHOD

An anthropomorphic phantom was scanned once with local abdominal CT protocols at 40 CT scanners, from four vendors, in thirty-three sites. Quantitative image quality was evaluated by CNR and SNR in the liver and kidney parenchyma. Qualitative image quality was assessed by visual grading analysis performed by three experienced radiologists using a five-point Likert scale to score thirteen image quality criteria. The CTDIvol was recorded for each scan. Pearson's correlation coefficient was calculated for the continuous variables, and the intraclass correlation coefficient was used to investigate interrater reliability between the radiologists.

RESULTS

CTDIvol ranged from 3.5 to 12 mGy (median 5.3 mGy, third quartile 6.7 mGy). SNR in liver parenchyma ranged from 4.4 to 14.4 (median 8.5), and CNR ranged from 2.7 to 11.2 (median 6.1). A weak correlation was found between CTDIvol and CNR (r = 0.270, p = 0.092). Variations in CNR across scanners at the same dose level CTDIvol were observed. No significant difference in CTDIvol or CNR was found based on scanner installation year. The oldest scanners had a 15 % higher median CTDIvol and a 12 % lower median CNR. The ICC showed acceptable agreement for all dose groups: low (ICC=0.889), medium (ICC=0.767), high (ICC=0.847), and in low (ICC=0.803) and medium (ICC=0.811) CNR groups.

CONCLUSION

There was large variation in radiation dose and image quality across the different CT scanners. Interestingly, the weak correlation between CTDIvol and CNR indicates that higher doses do not consistently improve CNR, indicating a need for systematic assessment and optimization of image quality and radiation doses for the abdominal CT examination.

Critical but commonly neglected factors that affect contrast medium administration in CT

OBJECTIVE

Past decades of research into contrast media injections and optimization thereof in radiology clinics have focused on scan acquisition parameters, patient-related factors, and contrast injection protocol variables. In this review, evidence is provided that a fourth bucket of crucial variables has been missed which account for previously unexplained phenomena and higher-than-expected variability in data. We propose how these critical factors should be considered and implemented in the contrast-medium administration protocols to optimize contrast enhancement.

METHODS

This article leverages a combination of methodologies for uncovering and quantifying confounding variables associated with or affecting the contrast-medium injection. Engineering benchtop equipment such as Coriolis flow meters, pressure transducers, and volumetric measurement devices are combined with small, targeted systematic evaluations querying operators, equipment, and the physics and fluid dynamics that make a seemingly simple task of injecting fluid into a patient a complex and non-linear endeavor.

RESULTS

Evidence is presented around seven key factors affecting the contrast-medium injection including a new way of selecting optimal IV catheters, degraded performance from longer tubing sets, variability associated with the mechanical injection system technology, common operator errors, fluids exchanging places stealthily based on gravity and density, wasted contrast media and inefficient saline flushes, as well as variability in the injected flow rate vs. theoretical expectations.

CONCLUSION

There remain several critical, but not commonly known, sources of error associated with contrast-medium injections. Elimination of these hidden sources of error where possible can bring immediate benefits and help to drive standardized and optimized contrast-media injections.

CRITICAL RELEVANCE STATEMENT

This review brings to light the commonly neglected/unknown factors negatively impacting contrast-medium injections and provides recommendations that can result in patient benefits, quality improvements, sustainability increases, and financial benefits by enabling otherwise unachievable optimization.

KEY POINTS

How IV contrast media is administered is a rarely considered source of CT imaging variability. IV catheter selection, tubing length, injection systems, and insufficient flushing can result in unintended variability. These findings can be immediately addressed to improve standardization in contrast-enhanced CT imaging.

Optimizing computed tomography image reconstruction for focal hepatic lesions: Deep learning image reconstruction vs iterative reconstruction

Background

Deep learning image reconstruction (DLIR) is a novel computed tomography (CT) reconstruction technique that minimizes image noise, enhances image quality, and enables radiation dose reduction. This study aims to compare the diagnostic performance of DLIR and iterative reconstruction (IR) in the evaluation of focal hepatic lesions.

Methods

We conducted a retrospective study of 216 focal hepatic lesions in 109 adult participants who underwent abdominal CT scanning at our institution. We used DLIR (low, medium, and high strength) and IR (0 %, 10 %, 20 %, and 30 %) techniques for image reconstruction. Four experienced abdominal radiologists independently evaluated focal hepatic lesions based on five qualitative aspects (lesion detectability, lesion border, diagnostic confidence level, image artifact, and overall image quality). Quantitatively, we measured and compared the level of image noise for each technique at the liver and aorta.

Results

There were significant differences (p < 0.001) among the seven reconstruction techniques in terms of lesion borders, image artifacts, and overall image quality. Low-strength DLIR (DLIR-L) exhibited the best overall image quality. Although high-strength DLIR (DLIR-H) had the least image noise and fewest artifacts, it also had the lowest scores for lesion borders and overall image quality. Image noise showed a weak to moderate positive correlation with participants' body mass index and waist circumference.

Conclusions

The optimal-strength DLIR significantly improved overall image quality for evaluating focal hepatic lesions compared to the IR technique. DLIR-L achieved the best overall image quality while maintaining acceptable levels of image noise and quality of lesion borders.

Comparison of the spectral performance between two dual-source CT systems on low-energy virtual monoenergetic images: A phantom study

PURPOSE

To compare the spectral performance of two different DSCT (DSCT-Pulse and DSCT-Force) on virtual monoenergetic images (VMIs) at low energy levels.

METHODS

An image quality phantom was scanned on the two DSCTs at three dose levels: 11/6/1.8 mGy. Level 3 of an advanced modeled iterative reconstruction algorithm was used. Noise power spectrum and task-based transfer function were computed on VMIs from 40 to 70 keV to assess noise magnitude and noise texture (fav) and spatial resolution (f50). A detectability index (d') was computed to assess the detection of one contrast-enhanced abdominal lesion as a function of the keV level used.

RESULTS

For all dose levels and all energy levels, noise magnitude was significantly higher (p < 0.05) with DSCT-Pulse than with DSCT-Force (12.6 ± 2.7 % at 1.8 mGy, 9.1 ± 2.9 % at 6 mGy and 4.0 ± 2.7 % at 11 mGy). For all energy levels, fav values were significantly higher (p < 0.05) with DSCT-Pulse than with DSCT-Force at 1.8 mGy (4.8 ± 3.9 %) and at 6 mGy (5.5 ± 2.5 %) but similar at 11 mGy (0.2 ± 3.6 %; p = 0.518). For all energy levels, f50 values were significantly higher with DSCT-Pulse than with DSCT-Force (12.7 ± 5.6 % at 1.8 mGy, 17.9 ± 4.5 % at 6 mGy and 13.1 ± 2.6 % at 11 mGy). For all keV, similar d' values were found with both DSCT-Force and DSCT-Pulse at 11 mGy (-1.0 ± 3.1 %; p = 0.084). For other dose levels, d' values were significantly lower with DSCT-Pulse than with DSCT-Force (9.1 ± 3.2 % at 1.8 mGy and -6.3 ± 3.9 % at 6 mGy).

CONCLUSION

Compared with the DSCT-Force, the DSCT-Pulse improved noise texture and spatial resolution, but noise magnitude was slightly higher and detectability slightly lower, particularly when the dose level was reduced.

Validating the safety of low-dose CTPA in pregnancy: results from the OPTICA (Optimised CT Pulmonary Angiography in Pregnancy) Study

BACKGROUND

Pulmonary embolism (PE) is a leading cause of pregnancy-related mortality. CT pulmonary angiogram (CTPA) is the first-line advanced imaging modality for suspected PE in pregnancy at institutes offering low-dose techniques; however, a protocol balancing safety with low dose remains undefined. The wide range of CTPA doses reported in pregnancy suggests a lack of confidence in implementing low-dose techniques in this group.

PURPOSE

To define and validate the safety, radiation dose and image quality of a low-dose CTPA protocol optimised for pregnancy.

MATERIALS AND METHODS

The OPTICA study is a prospective observational study. Pregnant study participants with suspected PE underwent the same CTPA protocol between May 2018 and February 2022. The primary outcome, CTPA safety, was judged by the reference standard; the 3-month incidence of venous thromboembolism (VTE) in study participants with a negative index CTPA. Secondary outcomes defined radiation dose and image quality. Absorbed breast, maternal effective and fetal doses were estimated by Monte-Carlo simulation on gestation-matched phantoms. Image quality was assessed by signal-to-noise and contrast-to-noise ratios and a Likert score for pulmonary arterial enhancement.

RESULTS

A total of 116 CTPAs were performed in 113 pregnant women of which 16 CTPAs were excluded. PE was diagnosed on 1 CTPA and out-ruled in 99. The incidence of recurrent symptomatic VTE was 0.0% (one-sided 95% CI, 2.66%) at follow-up. The mean absorbed breast dose was 2.9 ± 2.1mGy, uterine/fetal dose was 0.1 ± 0.2mGy and maternal effective dose was 1.4 ± 0.9mSv. Signal-to-noise ratio (SNR) was 11.9 ± 3.7. Contrast-to-noise ratio (CNR) was 10.4 ± 3.5.

CONCLUSION

The OPTICA CTPA protocol safely excluded PE in pregnant women across all trimesters, with low fetal and maternal radiation.

CLINICAL RELEVANCE

OPTICA (Optimised CT Pulmonary Angiography in Pregnancy) is the first prospective study to define the achievable radiation dose, image-quality and safety of a low-dose CT pulmonary angiogram protocol optimised for pregnancy (NCT04179487). It provides the current benchmark for safe and achievable CT pulmonary angiogram doses in the pregnant population.

KEY POINTS

• Despite the increased use of CT pulmonary angiogram in pregnancy, an optimised low-dose protocol has not been defined and reported doses in pregnancy continue to vary widely. • The OPTICA (Optimised CT Pulmonary Angiography in Pregnancy) study prospectively defines the achievable dose, image quality and safety of a low-dose CT pulmonary angiogram protocol using widely available technology. • OPTICA provides a benchmark for safe and achievable CT pulmonary angiogram doses in the pregnant population.

CAR and CSTR Cardiac Computed Tomography (CT) Practice Guidelines: Part 1 Coronary CT Angiography (CCTA)

Imaging the heart is one of the most technically challenging applications of Computed Tomography (CT) due to the presence of cardiac motion limiting optimal visualization of small structures such as the coronary arteries. Electrocardiographic gating during CT data acquisition facilitates motion free imaging of the coronary arteries. Since publishing the first version of the Canadian Association of Radiologists (CAR) cardiac CT guidelines, many technological advances in CT hardware and software have emerged necessitating an update. The goal of these cardiac CT practice guidelines is to present an overview of the current evidence supporting the use of cardiac CT in various clinical scenarios and to outline standards of practice for patient safety and quality of care when establishing a cardiac CT program in Canada.

Evolving and Novel Applications of Artificial Intelligence in Thoracic Imaging

The advent of artificial intelligence (AI) is revolutionizing medicine, particularly radiology. With the development of newer models, AI applications are demonstrating improved performance and versatile utility in the clinical setting. Thoracic imaging is an area of profound interest, given the prevalence of chest imaging and the significant health implications of thoracic diseases. This review aims to highlight the promising applications of AI within thoracic imaging. It examines the role of AI, including its contributions to improving diagnostic evaluation and interpretation, enhancing workflow, and aiding in invasive procedures. Next, it further highlights the current challenges and limitations faced by AI, such as the necessity of 'big data', ethical and legal considerations, and bias in representation. Lastly, it explores the potential directions for the application of AI in thoracic radiology.

3D reconstruction of an inertial-confinement fusion implosion with neural networks using multiple heterogeneous data sources

3D asymmetries are major degradation mechanisms in inertial-confinement fusion implosions at the National Ignition Facility (NIF). These asymmetries can be diagnosed and reconstructed with the neutron imaging system (NIS) on three lines of sight around the NIF target chamber. Conventional tomographic reconstructions are used to reconstruct the 3D morphology of the implosion using NIS [Volegov et al., J. Appl. Phys. 127, 083301 (2020)], but the problem is ill-posed with only three imaging lines of sight. Asymmetries can also be diagnosed with the real-time neutron activation diagnostics (RTNAD) and the neutron time-of-flight (nToF) suite. Since the NIS, RTNAD, and nToF each sample a different part of the implosion using different physical principles, we propose that it is possible to overcome the limitations of too few imaging lines of sight by performing 3D reconstructions that combine information from all three heterogeneous data sources. This work presents a new machine learning-based reconstruction technique to do just this. By using a simple physics model and group of neural networks to map 3D morphologies to data, this technique can easily account for data of multiple different types. A simple proof-of-principle is presented, demonstrating that this technique can accurately reconstruct a hot-spot shape using synthetic primary neutron images and a hot-spot velocity vector. In particular, the hot-spot's asymmetry, quantified as spherical harmonic coefficients, is reconstructed to within ±4% of the radius in 90% of test cases. In the future, this technique will be applied to actual NIS, RTNAD, and nToF data to better understand 3D asymmetries at the NIF.

A comparative analysis of deep learning and hybrid iterative reconstruction algorithms with contrast-enhancement-boost post-processing on the image quality of indirect computed tomography venography of the lower extremities

PURPOSE

To examine whether there is a significant difference in image quality between the deep learning reconstruction (DLR [AiCE, Advanced Intelligent Clear-IQ Engine]) and hybrid iterative reconstruction (HIR [AIDR 3D, adaptive iterative dose reduction three dimensional]) algorithms on the conventional enhanced and CE-boost (contrast-enhancement-boost) images of indirect computed tomography venography (CTV) of lower extremities.

MATERIALS AND METHODS

In this retrospective study, seventy patients who underwent CTV from June 2021 to October 2022 to assess deep vein thrombosis and varicose veins were included. Unenhanced and enhanced images were reconstructed for AIDR 3D and AiCE, AIDR 3D-boost and AiCE-boost images were obtained using subtraction software. Objective and subjective image qualities were assessed, and radiation doses were recorded.

RESULTS

The CT values of the inferior vena cava (IVC), femoral vein ( FV), and popliteal vein (PV) in the CE-boost images were approximately 1.3 (1.31-1.36) times higher than in those of the enhanced images. There were no significant differences in mean CT values of IVC, FV, and PV between AIDR 3D and AiCE, AIDR 3D-boost and AiCE-boost images. Noise in AiCE, AiCE-boost images was significantly lower than in AIDR 3D and AIDR 3D-boost images ( P < 0.05). The SNR (signal-to-noise ratio), CNR (contrast-to-noise ratio), and subjective scores of AiCE-boost images were the highest among 4 groups, surpassing AiCE, AIDR 3D, and AIDR 3D-boost images (all P < 0.05).

CONCLUSION

In indirect CTV of the lower extremities images, DLR with the CE-boost technique could decrease the image noise and improve the CT values, SNR, CNR, and subjective image scores. AiCE-boost images received the highest subjective image quality score and were more readily accepted by radiologists.

Ultra-high resolution coronary CT angiography on photon-counting detector CT: bi-centre study on the impact of quantum iterative reconstruction on image quality and accuracy of stenosis measurements

PURPOSE

To assess the impact of different quantum iterative reconstruction (QIR) levels on objective and subjective image quality of ultra-high resolution (UHR) coronary CT angiography (CCTA) images and to determine the effect of strength levels on stenosis quantification using photon-counting detector (PCD)-CT.

METHOD

A dynamic vessel phantom containing two calcified lesions (25 % and 50 % stenosis) was scanned at heart rates of 60, 80 and 100 beats per minute with a PCD-CT system. In vivo CCTA examinations were performed in 102 patients. All scans were acquired in UHR mode (slice thickness0.2 mm) and reconstructed with four different QIR levels (1-4) using a sharp vascular kernel (Bv64). Image noise, signal-to-noise ratio (SNR), sharpness, and percent diameter stenosis (PDS) were quantified in the phantom, while noise, SNR, contrast-to-noise ratio (CNR), sharpness, and subjective quality metrics (noise, sharpness, overall image quality) were assessed in patient scans.

RESULTS

Increasing QIR levels resulted in significantly lower objective image noise (in vitro and in vivo: both p < 0.001), higher SNR (both p < 0.001) and CNR (both p < 0.001). Sharpness and PDS values did not differ significantly among QIRs (all pairwise p > 0.008). Subjective noise of in vivo images significantly decreased with increasing QIR levels, resulting in significantly higher image quality scores at increasing QIR levels (all pairwise p < 0.001). Qualitative sharpness, on the other hand, did not differ across different levels of QIR (p = 0.15).

CONCLUSIONS

The QIR algorithm may enhance the image quality of CCTA datasets without compromising image sharpness or accurate stenosis measurements, with the most prominent benefits at the highest strength level.

Impact of reconstruction parameters on the accuracy of myocardial extracellular volume quantification on a first-generation, photon-counting detector CT

BACKGROUND

The potential role of cardiac computed tomography (CT) has increasingly been demonstrated for the assessment of diffuse myocardial fibrosis through the quantification of extracellular volume (ECV). Photon-counting detector (PCD)-CT technology may deliver more accurate ECV quantification compared to energy-integrating detector CT. We evaluated the impact of reconstruction settings on the accuracy of ECV quantification using PCD-CT, with magnetic resonance imaging (MRI)-based ECV as reference.

METHODS

In this post hoc analysis, 27 patients (aged 53.1 ± 17.2 years (mean ± standard deviation); 14 women) underwent same-day cardiac PCD-CT and MRI. Late iodine CT scans were reconstructed with different quantum iterative reconstruction levels (QIR 1-4), slice thicknesses (0.4-8 mm), and virtual monoenergetic imaging levels (VMI, 40-90 keV); ECV was quantified for each reconstruction setting. Repeated measures ANOVA and t-test for pairwise comparisons, Bland-Altman plots, and Lin's concordance correlation coefficient (CCC) were used.

RESULTS

ECV values did not differ significantly among QIR levels (p = 1.000). A significant difference was observed throughout different slice thicknesses, with 0.4 mm yielding the highest agreement with MRI-based ECV (CCC = 0.944); 45-keV VMI reconstructions showed the lowest mean bias (0.6, 95% confidence interval 0.1-1.4) compared to MRI. Using the most optimal reconstruction settings (QIR4. slice thickness 0.4 mm, VMI 45 keV), a 63% reduction in mean bias and a 6% increase in concordance with MRI-based ECV were achieved compared to standard settings (QIR3, slice thickness 1.5 mm; VMI 65 keV).

CONCLUSIONS

The selection of appropriate reconstruction parameters improved the agreement between PCD-CT and MRI-based ECV.

RELEVANCE STATEMENT

Tailoring PCD-CT reconstruction parameters optimizes ECV quantification compared to MRI, potentially improving its clinical utility.

KEY POINTS

• CT is increasingly promising for myocardial tissue characterization, assessing focal and diffuse fibrosis via late iodine enhancement and ECV quantification, respectively. • PCD-CT offers superior performance over conventional CT, potentially improving ECV quantification and its agreement with MRI-based ECV. • Tailoring PCD-CT reconstruction parameters optimizes ECV quantification compared to MRI, potentially improving its clinical utility.

DDT-Net: Dose-Agnostic Dual-Task Transfer Network for Simultaneous Low-Dose CT Denoising and Simulation

Deep learning (DL) algorithms have achieved unprecedented success in low-dose CT (LDCT) imaging and are expected to be a new generation of CT reconstruction technology. However, most DL-based denoising models often lack the ability to generalize to unseen dose data. Moreover, most simulation tools for LDCT typically operate on proprietary projection data, which is generally not accessible without an established collaboration with CT manufacturers. To alleviate these issues, in this work, we propose a dose-agnostic dual-task transfer network, termed DDT-Net, for simultaneous LDCT denoising and simulation. Concretely, the dual-task learning module is constructed to integrate the LDCT denoising and simulation tasks into a unified optimization framework by learning the joint distribution of LDCT and NDCT data. We approximate the joint distribution of continuous dose level data by training DDT-Net with discrete dose data, which can be generalized to denoising and simulation of unseen dose data. In particular, the mixed-dose training strategy adopted by DDT-Net can promote the denoising performance of lower-dose data. The paired dataset simulated by DDT-Net can be used for data augmentation to further restore the tissue texture of LDCT images. Experimental results on synthetic data and clinical data show that the proposed DDT-Net outperforms competing methods in terms of denoising and generalization performance at unseen dose data, and it also provides a simulation tool that can quickly simulate realistic LDCT images at arbitrary dose levels.

Evaluation of image quality on low contrast media with deep learning image reconstruction algorithm in prospective ECG-triggering coronary CT angiography

To assess the impact of low-dose contrast media (CM) injection protocol with deep learning image reconstruction (DLIR) algorithm on image quality in coronary CT angiography (CCTA). In this prospective study, patients underwent CCTA were prospectively and randomly assigned to three groups with different contrast volume protocols (at 320mgI/mL concentration and constant flow rate of 5ml/s). After pairing basic information, 210 patients were enrolled in this study: Group A, 0.7mL/kg (n = 70); Group B, 0.6mL/kg (n = 70); Group C, 0.5mL/kg (n = 70). All patients were examined via a prospective ECG-triggered scan protocol within one heartbeat. A high level DLIR (DLIR-H) algorithm was used for image reconstruction with a thickness and interval of 0.625mm. The CT values of ascending aorta (AA), descending aorta (DA), three main coronary arteries, pulmonary artery (PA), and superior vena cava (SVC) were measured and analyzed for objective assessment. Two radiologists assessed the image quality and diagnostic confidence using a 5-point Likert scale. The CM doses were 46.81 ± 6.41mL, 41.96 ± 7.51mL and 34.65 ± 5.38mL for Group A, B and C, respectively. The objective assessments on AA, DA and the three main coronary arteries and the overall subjective scoring showed no significant difference among the three groups (all p > 0.05). The subjective assessment proved that excellent CCTA images can be obtained from the three different contrast media protocols. There were no significant differences in intracoronary attenuation values between the higher HR subgroup and the lower HR subgroup among three groups. CCTA reconstructed with DLIR could be realized with adequate enhancement in coronary arteries, excellent image quality and diagnostic confidence at low contrast dose of a 0.5mL/kg. The use of lower tube voltages may further reduce the contrast dose requirement.

Recanalization status and temporal evolution of early ischemic changes following stroke thrombectomy

INTRODUCTION

Present-day computer tomography (CT) scanners have excellent spatial resolution and signal-to-noise ratio and are instrumental detecting early ischemic changes (EIC) in brain. We assessed the temporal changes of EIC based on the recanalization status after thrombectomy.

PATIENTS AND METHODS

The cohort comprises consecutive patients with acute ischemic stroke in anterior circulation treated with thrombectomy in tertiary referral hospital. All baseline and follow-up scans were screened for any ischemic changes and further classified using Alberta Stroke Program Early CT Score (ASPECTS). Generalized linear mixed models were used to analyze the impact of recanalization status using modified Thrombolysis in Cerebral Infarction (mTICI) on temporal evolution of ischemic changes.

RESULTS

We included 614 patients with ICA, M1, or M2 occlusions. Median ASPECTS score was 9 (IQR 7-10) at baseline and 7 (5-8) at approximately 24 h. mTICI 3 was achieved in 207 (33.8%), 2B 241 (39.3%), 2A in 77 (12.6%), and 0-1 in 88 (14.3%) patients. Compared to patients with mTICI 3, those with mTICI 0-1 and 2A had less favorable temporal changes of ASPECTS (p < 0.001). Effect of recanalization was noted in the cortical regions of ICA/M1 patients, but not in their deep structures or patients with M2 occlusions. All ischemic changes detected at baseline were also present at all follow-up images, regardless of the recanalization status.

CONCLUSIONS

Temporal evolution of the ischemic changes and ASPECTS are related to the success of the recanalization therapy in cortical regions of ICA/M1 patients, but not in their deep brain structures or M2 patients. In none of the patients did EIC revert in any brain region after successful recanalization.

Insights about cervical lymph nodes: Evaluating deep learning-based reconstruction for head and neck computed tomography scan

Purpose

This study aimed to investigate differences in cervical lymph node image quality on dual-energy computed tomography (CT) scan with datasets reconstructed using filter back projection (FBP), hybrid iterative reconstruction (IR), and deep learning-based image reconstruction (DLIR) in patients with head and neck cancer.

Method

Seventy patients with head and neck cancer underwent follow-up contrast-enhanced dual-energy CT examinations. All datasets were reconstructed using FBP, hybrid IR with 30 % adaptive statistical IR (ASiR-V), and DLIR with three selectable levels (low, medium, and high) at 2.5- and 0.625-mm slice thicknesses. Herein, signal, image noise, signal-to-noise ratio, and contrast-to-noise ratio of lymph nodes and overall image quality, artifact, and noise of selected regions of interest were evaluated by two radiologists. Next, cervical lymph node sharpness was evaluated using full width at half maximum.

Results

DLIR exhibited significantly reduced noise, ranging from 3.8 % to 35.9 % with improved signal-to-noise ratio (11.5-105.6 %) and contrast-to-noise ratio (10.5-107.5 %) compared with FBP and ASiR-V, for cervical lymph nodes (p < 0.001). Further, 0.625-mm-thick images reconstructed using DLIR-medium and DLIR-high had a lower noise than 2.5-mm-thick images reconstructed using FBP and ASiR-V. The lymph node margins and vessels on DLIR-medium and DLIR-high were sharper than those on FBP and ASiR-V (p < 0.05). Both readers agreed that DLIR had a better image quality than the conventional reconstruction algorithms.

Conclusion

DLIR-medium and -high provided superior cervical lymph node image quality in head and neck CT. Improved image quality affords thin-slice DLIR images for dose-reduction protocols in the future.

Probabilistic U-Net model observer for the DDC method in CT scan protocol optimization

Optimizing complex imaging procedures within Computed Tomography, considering both dose and image quality, presents significant challenges amidst rapid technological advancements and the adoption of machine learning (ML) methods. A crucial metric in this context is the Difference-Detailed Curve, which relies on human observer studies. However, these studies are labor-intensive and prone to both inter- and intra-observer variability. To tackle these issues, a ML-based model observer utilizing the U-Net architecture and a Bayesian methodology is proposed. In order to train a model observer unaffected by the spatial arrangement of low-contrast objects, the image preprocessing incorporates a Gaussian Process-based noise model. Additionally, gradient-weighted class activation mapping is utilized to gain insights into the model observer's decision-making process. By training on data from a diverse group of observers, well-calibrated probabilistic predictions that quantify observer variability are achieved. Leveraging the principles of Beta regression, the Bayesian methodology is used to derive a model observer performance metric, effectively gauging the model observer's strength in terms of an 'effective number of observers'. Ultimately, this framework enables to predict the DDC distribution by applying thresholds to the inferred probabilities (Part of this work has been presented at: Stocker D, Sommer C, Gueng S, Stäuble J, Özden I, Griessinger J, Weyland M S, Lutters G, Scheidegger S (2023). Probabilistic U-Net Model Observer for the DDC Method in CT Scan Protocol Optimization. The 56th SSRMP Annual Meeting 2023, November 30. - December 1., 2023, Luzern, Switzerland).

Patient-derived PixelPrint phantoms for evaluating clinical imaging performance of a deep learning CT reconstruction algorithm

Objective. Deep learning reconstruction (DLR) algorithms exhibit object-dependent resolution and noise performance. Thus, traditional geometric CT phantoms cannot fully capture the clinical imaging performance of DLR. This study uses a patient-derived 3D-printed PixelPrint lung phantom to evaluate a commercial DLR algorithm across a wide range of radiation dose levels.Method. The lung phantom used in this study is based on a patient chest CT scan containing ground glass opacities and was fabricated using PixelPrint 3D-printing technology. The phantom was placed inside two different size extension rings to mimic a small- and medium-sized patient and was scanned on a conventional CT scanner at exposures between 0.5 and 20 mGy. Each scan was reconstructed using filtered back projection (FBP), iterative reconstruction, and DLR at five levels of denoising. Image noise, contrast to noise ratio (CNR), root mean squared error, structural similarity index (SSIM), and multi-scale SSIM (MS SSIM) were calculated for each image.Results.DLR demonstrated superior performance compared to FBP and iterative reconstruction for all measured metrics in both phantom sizes, with better performance for more aggressive denoising levels. DLR was estimated to reduce dose by 25%-83% in the small phantom and by 50%-83% in the medium phantom without decreasing image quality for any of the metrics measured in this study. These dose reduction estimates are more conservative compared to the estimates obtained when only considering noise and CNR.Conclusion. DLR has the capability of producing diagnostic image quality at up to 83% lower radiation dose, which can improve the clinical utility and viability of lower dose CT scans. Furthermore, the PixelPrint phantom used in this study offers an improved testing environment with more realistic tissue structures compared to traditional CT phantoms, allowing for structure-based image quality evaluation beyond noise and contrast-based assessments.

Impact of deep learning image reconstruction on volumetric accuracy and image quality of pulmonary nodules with different morphologies in low-dose CT

BACKGROUND

This study systematically compares the impact of innovative deep learning image reconstruction (DLIR, TrueFidelity) to conventionally used iterative reconstruction (IR) on nodule volumetry and subjective image quality (IQ) at highly reduced radiation doses. This is essential in the context of low-dose CT lung cancer screening where accurate volumetry and characterization of pulmonary nodules in repeated CT scanning are indispensable.

MATERIALS AND METHODS

A standardized CT dataset was established using an anthropomorphic chest phantom (Lungman, Kyoto Kaguku Inc., Kyoto, Japan) containing a set of 3D-printed lung nodules including six diameters (4 to 9 mm) and three morphology classes (lobular, spiculated, smooth), with an established ground truth. Images were acquired at varying radiation doses (6.04, 3.03, 1.54, 0.77, 0.41 and 0.20 mGy) and reconstructed with combinations of reconstruction kernels (soft and hard kernel) and reconstruction algorithms (ASIR-V and DLIR at low, medium and high strength). Semi-automatic volumetry measurements and subjective image quality scores recorded by five radiologists were analyzed with multiple linear regression and mixed-effect ordinal logistic regression models.

RESULTS

Volumetric errors of nodules imaged with DLIR are up to 50% lower compared to ASIR-V, especially at radiation doses below 1 mGy and when reconstructed with a hard kernel. Also, across all nodule diameters and morphologies, volumetric errors are commonly lower with DLIR. Furthermore, DLIR renders higher subjective IQ, especially at the sub-mGy doses. Radiologists were up to nine times more likely to score the highest IQ-score to these images compared to those reconstructed with ASIR-V. Lung nodules with irregular margins and small diameters also had an increased likelihood (up to five times more likely) to be ascribed the best IQ scores when reconstructed with DLIR.

CONCLUSION

We observed that DLIR performs as good as or even outperforms conventionally used reconstruction algorithms in terms of volumetric accuracy and subjective IQ of nodules in an anthropomorphic chest phantom. As such, DLIR potentially allows to lower the radiation dose to participants of lung cancer screening without compromising accurate measurement and characterization of lung nodules.

Image Quality and Lesion Detection of Multiplanar Reconstruction Images Using Deep Learning: Comparison with Hybrid Iterative Reconstruction

Background

We assessed and compared the image quality of normal and pathologic structures as well as the image noise in chest computed tomography images using "adaptive statistical iterative reconstruction-V" (ASiR-V) or deep learning reconstruction "TrueFidelity".

Methods

Forty consecutive patients with suspected lung disease were evaluated. The 1.25-mm axial images and 2.0-mm coronal multiplanar images were reconstructed under the following three conditions: (i) ASiR-V, lung kernel with 60% of ASiR-V; (ii) TF-M, standard kernel, image filter (Lung) with TrueFidelity at medium strength; and (iii) TF-H, standard kernel, image filter (Lung) with TrueFidelity at high strength. Two radiologists (readers) independently evaluated the image quality of anatomic structures using a scale ranging from 1 (best) to 5 (worst). In addition, readers ranked their image preference. Objective image noise was measured using a circular region of interest in the lung parenchyma. Subjective image quality scores, total scores for normal and abnormal structures, and lesion detection were compared using Wilcoxon's signed-rank test. Objective image quality was compared using Student's paired t-test and Wilcoxon's signed-rank test. The Bonferroni correction was applied to the P value, and significance was assumed only for values of P < 0.016.

Results

Both readers rated TF-M and TF-H images significantly better than ASiR-V images in terms of visualization of the centrilobular region in axial images. The preference score of TF-M and TF-H images for reader 1 were better than that of ASiR-V images, and the preference score of TF-H images for reader 2 were significantly better than that of ASiR-V and TF-M images. TF-M images showed significantly lower objective image noise than ASiR-V or TF-H images.

Conclusion

TrueFidelity showed better image quality, especially in the centrilobular region, than ASiR-V in subjective and objective evaluations. In addition, the image texture preference for TrueFidelity was better than that for ASiR-V.

Improving image quality of triple-low-protocol renal artery CT angiography with deep-learning image reconstruction: a comparative study with standard-dose single-energy and dual-energy CT with adaptive statistical iterative reconstruction

AIM

To investigate the improvement in image quality of triple-low-protocol (low radiation, low contrast medium dose, low injection speed) renal artery computed tomography (CT) angiography (RACTA) using deep-learning image reconstruction (DLIR), in comparison with standard-dose single- and dual-energy CT (DECT) using adaptive statistical iterative reconstruction-Veo (ASIR-V) algorithm.

MATERIALS AND METHODS

Ninety patients for RACTA were divided into different groups: standard-dose single-energy CT (S group) using ASIR-V at 60% strength (60%ASIR-V), DECT (DE group) with 60%ASIR-V including virtual monochromatic images at 40 keV (DE40 group) and 70 keV (DE70 group), and the triple-low protocol single-energy CT (L group) with DLIR at high level (DLIR-H). The effective dose (ED), contrast medium dose, injection speed, standard deviation (SD), signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of abdominal aorta (AA), and left/right renal artery (LRA, RRA), and subjective scores were compared among the different groups.

RESULTS

The L group significantly reduced ED by 37.6% and 31.2%, contrast medium dose by 33.9% and 30.5%, and injection speed by 30% and 30%, respectively, compared to the S and DE groups. The L group had the lowest SD values for all arteries compared to the other groups (p<0.001). The SNR of RRA and LRA in the L group, and the CNR of all arteries in the DE40 group had highest value compared to others (p<0.05). The L group had the best comprehensive score with good consistency (p<0.05).

CONCLUSIONS

The triple-low protocol RACTA with DLIR-H significantly reduces the ED, contrast medium doses, and injection speed, while providing good comprehensive image quality.

Artificial intelligence in coronary artery calcium score: rationale, different approaches, and outcomes

Almost 35 years after its introduction, coronary artery calcium score (CACS) not only survived technological advances but became one of the cornerstones of contemporary cardiovascular imaging. Its simplicity and quantitative nature established it as one of the most robust approaches for atherosclerotic cardiovascular disease risk stratification in primary prevention and a powerful tool to guide therapeutic choices. Groundbreaking advances in computational models and computer power translated into a surge of artificial intelligence (AI)-based approaches directly or indirectly linked to CACS analysis. This review aims to provide essential knowledge on the AI-based techniques currently applied to CACS, setting the stage for a holistic analysis of the use of these techniques in coronary artery calcium imaging. While the focus of the review will be detailing the evidence, strengths, and limitations of end-to-end CACS algorithms in electrocardiography-gated and non-gated scans, the current role of deep-learning image reconstructions, segmentation techniques, and combined applications such as simultaneous coronary artery calcium and pulmonary nodule segmentation, will also be discussed.

Influence of deep learning image reconstruction algorithm for reducing radiation dose and image noise compared to iterative reconstruction and filtered back projection for head and chest computed tomography examinations: a systematic review

Background

The most recent advances in Computed Tomography (CT) image reconstruction technology are Deep learning image reconstruction (DLIR) algorithms. Due to drawbacks in Iterative reconstruction (IR) techniques such as negative image texture and nonlinear spatial resolutions, DLIRs are gradually replacing them. However, the potential use of DLIR in Head and Chest CT has to be examined further. Hence, the purpose of the study is to review the influence of DLIR on Radiation dose (RD), Image noise (IN), and outcomes of the studies compared with IR and FBP in Head and Chest CT examinations.

Methods

We performed a detailed search in PubMed, Scopus, Web of Science, Cochrane Library, and Embase to find the articles reported using DLIR for Head and Chest CT examinations between 2017 to 2023. Data were retrieved from the short-listed studies using Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines.

Results

Out of 196 articles searched, 15 articles were included. A total of 1292 sample size was included. 14 articles were rated as high and 1 article as moderate quality. All studies compared DLIR to IR techniques. 5 studies compared DLIR with IR and FBP. The review showed that DLIR improved IQ, and reduced RD and IN for CT Head and Chest examinations.

Conclusions

DLIR algorithm have demonstrated a noted enhancement in IQ with reduced IN for CT Head and Chest examinations at lower dose compared with IR and FBP. DLIR showed potential for enhancing patient care by reducing radiation risks and increasing diagnostic accuracy.

Low-dose liver CT: image quality and diagnostic accuracy of deep learning image reconstruction algorithm

OBJECTIVES

To perform a comprehensive within-subject image quality analysis of abdominal CT examinations reconstructed with DLIR and to evaluate diagnostic accuracy compared to the routinely applied adaptive statistical iterative reconstruction (ASiR-V) algorithm.

MATERIALS AND METHODS

Oncologic patients were prospectively enrolled and underwent contrast-enhanced CT. Images were reconstructed with DLIR with three intensity levels of reconstruction (high, medium, and low) and ASiR-V at strength levels from 10 to 100% with a 10% interval. Three radiologists characterized the lesions and two readers assessed diagnostic accuracy and calculated signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), figure of merit (FOM), and subjective image quality, the latter with a 5-point Likert scale.

RESULTS

Fifty patients (mean age: 70 ± 10 years, 23 men) were enrolled and 130 liver lesions (105 benign lesions, 25 metastases) were identified. DLIR_H achieved the highest SNR and CNR, comparable to ASiR-V 100% (p ≥ .051). DLIR_M returned the highest subjective image quality (score: 5; IQR: 4-5; p ≤ .001) and significant median increase (29%) in FOM (p < .001). Differences in detection were identified only for lesions ≤ 0.5 cm: 32/33 lesions were detected with DLIR_M and 26 lesions were detected with ASiR-V 50% (p = .031). Lesion accuracy of was 93.8% (95% CI: 88.1, 97.3; 122 of 130 lesions) for DLIR and 87.7% (95% CI: 80.8, 92.8; 114 of 130 lesions) for ASiR-V 50%.

CONCLUSIONS

DLIR yields superior image quality and provides higher diagnostic accuracy compared to ASiR-V in the assessment of hypovascular liver lesions, in particular for lesions ≤ 0.5 cm.

CLINICAL RELEVANCE STATEMENT

Deep learning image reconstruction algorithm demonstrates higher diagnostic accuracy compared to iterative reconstruction in the identification of hypovascular liver lesions, especially for lesions ≤ 0.5 cm.

KEY POINTS

• Iterative reconstruction algorithm impacts image texture, with negative effects on diagnostic capabilities. • Medium-strength deep learning image reconstruction algorithm outperforms iterative reconstruction in the diagnostic accuracy of ≤ 0.5 cm hypovascular liver lesions (93.9% vs 78.8%), also granting higher objective and subjective image quality. • Deep learning image reconstruction algorithm can be safely implemented in routine abdominal CT protocols in place of iterative reconstruction.

Deep-learning reconstruction with low-contrast media and low-kilovoltage peak for CT of the liver

AIM

To compare images using reduced CM, low-kVp scanning and DLR reconstruction with conventional images (no CM reduction, normal tube voltage, reconstructed with HBIR. To compare images using reduced contrast media (CM), low kilovoltage peak (kVp) scanning and deep-learning reconstruction (DLR) with conventional image quality (no CM reduction, normal tube voltage, reconstructed with hybrid-type iterative reconstruction method [HBIR protocol]).

MATERIALS AND METHODS

A retrospective analysis was performed on 70 patients with liver disease and three-phase dynamic imaging using computed tomography (CT) from April 2020 to March 2022 at Oita University Hospital. Of these cases, 39 were reconstructed using the DLR protocol at a tube voltage of 80 kVp and CM of 300 mg iodine/kg while 31 were imaged at a tube voltage of 120 kVp with CM of 600 mg iodine/kg and were reconstructed by the usual HBIR protocol. Images from the DLR and HBIR protocols were analysed and compared based on the contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR), figure-of-merit (FOM), and visual assessment. The CT dose index (CTDI)vol and size-specific dose estimates (SSDE) were compared with respect to radiation dose.

RESULTS

The DLR protocol was superior, with significant differences in CNR, SNR, and FOM except hepatic parenchyma in the arterial phase. For visual assessment, the DLR protocol had better values for vascular visualisation for the portal vein, image noise, and contrast enhancement of the hepatic parenchyma. Regarding comparison of the radiation dose, the DLR protocol was superior for all values of CTDIvol and SSDE, with significant differences (p<0.01; max. 52%).

CONCLUSION

Protocols using DLR with reduced CM and low kVp have better image quality and lower radiation dose compared to protocols using conventional HBIR.

Investigating the Small Pixel Effect in Ultra-High Resolution Photon-Counting CT of the Lung

OBJECTIVES

The aim of this study was to investigate potential benefits of ultra-high resolution (UHR) over standard resolution scan mode in ultra-low dose photon-counting detector CT (PCD-CT) of the lung.

MATERIALS AND METHODS

Six cadaveric specimens were examined with 5 dose settings using tin prefiltration, each in UHR (120 × 0.2 mm) and standard mode (144 × 0.4 mm), on a first-generation PCD-CT scanner. Image quality was evaluated quantitatively by noise comparisons in the trachea and both main bronchi. In addition, 16 readers (14 radiologists and 2 internal medicine physicians) independently completed a browser-based pairwise forced-choice comparison task for assessment of subjective image quality. The Kendall rank coefficient ( W ) was calculated to assess interrater agreement, and Pearson's correlation coefficient ( r ) was used to analyze the relationship between noise measurements and image quality rankings.

RESULTS

Across all dose levels, image noise in UHR mode was lower than in standard mode for scan protocols matched by CTDI vol ( P < 0.001). UHR examinations exhibited noise levels comparable to the next higher dose setting in standard mode ( P ≥ 0.275). Subjective ranking of protocols based on 5760 pairwise tests showed high interrater agreement ( W = 0.99; P ≤ 0.001) with UHR images being preferred by readers in the majority of comparisons. Irrespective of scan mode, a substantial indirect correlation was observed between image noise and subjective image quality ranking ( r = -0.97; P ≤ 0.001).

CONCLUSIONS

In PCD-CT of the lung, UHR scan mode reduces image noise considerably over standard resolution acquisition. Originating from the smaller detector element size in fan direction, the small pixel effect allows for superior image quality in ultra-low dose examinations with considerable potential for radiation dose reduction.