The potential for reduced radiation dose from deep learning-based CT image reconstruction: A comparison with filtered back projection and hybrid iterative reconstruction using a phantom

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.

Enhanced visualization in endoleak detection through iterative and AI-noise optimized spectral reconstructions

To assess the image quality parameters of dual-energy computed tomography angiography (DECTA) 40-, and 60 keV virtual monoenergetic images (VMIs) combined with deep learning-based image reconstruction model (DLM) and iterative reconstructions (IR). CT scans of 28 post EVAR patients were enrolled. The 60 s delayed phase of DECTA was evaluated. Objective [noise, contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR)] and subjective (overall image quality and endoleak conspicuity - 3 blinded readers assessment) image quality analyses were performed. The following reconstructions were evaluated: VMI 40, 60 keV VMI; IR VMI 40, 60 keV; DLM VMI 40, 60 keV. The noise level of the DLM VMI images was approximately 50% lower than that of VMI reconstruction. The highest CNR and SNR values were measured in VMI DLM images. The mean CNR in endoleak in 40 keV was accounted for as 1.83 ± 1.2; 2.07 ± 2.02; 3.6 ± 3.26 in VMI, VMI IR, and VMI DLM, respectively. The DLM algorithm significantly reduced noise and increased lesion conspicuity, resulting in higher objective and subjective image quality compared to other reconstruction techniques. The application of DLM algorithms to low-energy VMIs significantly enhances the diagnostic value of DECTA in evaluating endoleaks. DLM reconstructions surpass traditional VMIs and IR in terms of image quality.

Strategies for Implementing Machine Learning Algorithms in the Clinical Practice of Radiology

Despite recent advancements in machine learning (ML) applications in health care, there have been few benefits and improvements to clinical medicine in the hospital setting. To facilitate clinical adaptation of methods in ML, this review proposes a standardized framework for the step-by-step implementation of artificial intelligence into the clinical practice of radiology that focuses on three key components: problem identification, stakeholder alignment, and pipeline integration. A review of the recent literature and empirical evidence in radiologic imaging applications justifies this approach and offers a discussion on structuring implementation efforts to help other hospital practices leverage ML to improve patient care. Clinical trial registration no. 04242667 © RSNA, 2024 Supplemental material is available for this article.

A Deep Learning Image Reconstruction Algorithm for Improving Image Quality and Hepatic Lesion Detectability in Abdominal Dual-Energy Computed Tomography: Preliminary Results

This study aimed to compare the performance of deep learning image reconstruction (DLIR) and adaptive statistical iterative reconstruction-Veo (ASIR-V) in improving image quality and diagnostic performance using virtual monochromatic spectral images in abdominal dual-energy computed tomography (DECT). Sixty-two patients [mean age ± standard deviation (SD): 56 years ± 13; 30 men] who underwent abdominal DECT were prospectively included in this study. The 70-keV DECT images in the portal phase were reconstructed at 5-mm and 1.25-mm slice thicknesses with 40% ASIR-V (ASIR-V40%) and at 1.25-mm slice with deep learning image reconstruction at medium (DLIR-M) and high (DLIR-H) levels and then compared. Computed tomography (CT) attenuation, SD values, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) were measured in the liver, spleen, erector spinae, and intramuscular fat. The lesions in each reconstruction group at 1.25-mm slice thickness were counted. The image quality and diagnostic confidence were subjectively evaluated by two radiologists using a 5-point scale. For the 1.25-mm images, DLIR-M and DLIR-H had lower SD, higher SNR and CNR, and better subjective image quality compared with ASIR-V40%; DLIR-H performed the best (all P values < 0.001). Furthermore, the 1.25-mm DLIR-H images had similar SD, SNR, and CNR values as the 5-mm ASIR-V40% images (all P > 0.05). Three image groups had similar lesion detection rates, but DLIR groups exhibited higher confidence in diagnosing lesions. Compared with ASIR-V40% at 70 keV, 70-keV DECT with DLIR-H further reduced image noise and improved image quality. Additionally, it improved diagnostic confidence while ensuring a consistent lesion detection rate of liver lesions.

Implementation of AI image reconstruction in CT-how is it validated and what dose reductions can be achieved

CT reconstruction has undergone a substantial change over the last decade with the introduction of iterative reconstruction (IR) and now with deep learning reconstruction (DLR). In this review, DLR will be compared to IR and filtered back-projection (FBP) reconstructions. Comparisons will be made using image quality metrics such as noise power spectrum, contrast-dependent task-based transfer function, and non-prewhitening filter detectability index (dNPW'). Discussion on how DLR has impacted CT image quality, low-contrast detectability, and diagnostic confidence will be provided. DLR has shown the ability to improve in areas that IR is lacking, namely: noise magnitude reduction does not alter noise texture to the degree that IR did, and the noise texture found in DLR is more aligned with noise texture of an FBP reconstruction. Additionally, the dose reduction potential for DLR is shown to be greater than IR. For IR, the consensus was dose reduction should be limited to no more than 15-30% to preserve low-contrast detectability. For DLR, initial phantom and patient observer studies have shown acceptable dose reduction between 44 and 83% for both low- and high-contrast object detectability tasks. Ultimately, DLR is able to be used for CT reconstruction in place of IR, making it an easy "turnkey" upgrade for CT reconstruction. DLR for CT is actively being improved as more vendor options are being developed and current DLR options are being enhanced with second generation algorithms being released. DLR is still in its developmental early stages, but is shown to be a promising future for CT reconstruction.

Investigating the Feasibility of Using DenseNet to Improve Coronary Calcification Detection in CT

OBJECTIVES

Interscan reproducibility of coronary artery calcium (CAC) scoring can be improved by using a smaller slice thickness but at the cost of higher image noise. This study aimed to investigate the feasibility of using densely connected convolutional network (DenseNet) to reduce the image noise in CAC scans reconstructed with slice thickness < 3 mm for improving coronary calcification detection in CT.

METHODS

Phantom data acquired with QRM and CIRS phantoms were used for model training and testing, where the DenseNet model adopted in this work was a convolutional neural network (CNN) designed for super resolution recovery. After phantom study, the proposed method was evaluated in terms of its ability to improve calcification detection using patient data. The CNN input images (IMGinput) were CAC scans reconstructed with 0.5-, 1.0- and 1.5-mm slice thickness, while CNN label images were CAC scans reconstructed with 3-mm slice thickness (IMG3mm). Region of interest (ROI) analysis was carried out on IMG3mm, IMGinput and CNN output images (IMGoutput). Two-sample t test was used to compare the difference in Hounsfield Unit (HU) values within ROI between IMG3mm and IMGoutput.

RESULTS

For the calcifications in QRM phantoms, no statistically significant difference was found when comparing the HU values of 400- and 800-HA calcifications identified on IMG3mm to those on IMGoutput with slice thickness of 0.5, 1.0 or 1.5 mm. On the other hand, statistically significant difference was found when comparing the HU values of 200-HA calcifications identified on IMG3mm to those on IMGoutput with a slice thickness of 0.5 and 1.0 mm. Meanwhile, no statistically significant difference was found when comparing the HU values of 200-HA calcifications identified on IMG3mm to those on IMGoutput with a slice thickness of 1.5 mm. As for the rod inserts in CIRS phantoms simulating 9 different tissue types in human body, there was no statistically significant difference between IMG3mm and IMGoutput with slice thickness of 1.5 mm, and all the p values were larger than 0.10. With regards to patient study, more calcification pixels were detected on IMGoutput with a slice thickness of 1.5 mm than on IMG3mm, so calcifications were more clear on the denoised images.

CONCLUSION

According to our results, the CNN-based denoising method could reduce statistical noise in IMGinput with a slice thickness of 1.5 mm without causing significant texture change or variation in HU values. The proposed method could improve cardiovascular risk prediction by detecting small and soft calcifications that are barely identified on 3-mm slice images used in conventional CAC scans.

Latest Developments in Adapting Deep Learning for Assessing TAVR Procedures and Outcomes

Aortic valve defects are among the most prevalent clinical conditions. A severely damaged or non-functioning aortic valve is commonly replaced with a bioprosthetic heart valve (BHV) via the transcatheter aortic valve replacement (TAVR) procedure. Accurate pre-operative planning is crucial for a successful TAVR outcome. Assessment of computational fluid dynamics (CFD), finite element analysis (FEA), and fluid-solid interaction (FSI) analysis offer a solution that has been increasingly utilized to evaluate BHV mechanics and dynamics. However, the high computational costs and the complex operation of computational modeling hinder its application. Recent advancements in the deep learning (DL) domain can offer a real-time surrogate that can render hemodynamic parameters in a few seconds, thus guiding clinicians to select the optimal treatment option. Herein, we provide a comprehensive review of classical computational modeling approaches, medical imaging, and DL approaches for planning and outcome assessment of TAVR. Particularly, we focus on DL approaches in previous studies, highlighting the utilized datasets, deployed DL models, and achieved results. We emphasize the critical challenges and recommend several future directions for innovative researchers to tackle. Finally, an end-to-end smart DL framework is outlined for real-time assessment and recommendation of the best BHV design for TAVR. Ultimately, deploying such a framework in future studies will support clinicians in minimizing risks during TAVR therapy planning and will help in improving patient care.

Deep Learning Image Reconstruction for CT: Technical Principles and Clinical Prospects

Filtered back projection (FBP) has been the standard CT image reconstruction method for 4 decades. A simple, fast, and reliable technique, FBP has delivered high-quality images in several clinical applications. However, with faster and more advanced CT scanners, FBP has become increasingly obsolete. Higher image noise and more artifacts are especially noticeable in lower-dose CT imaging using FBP. This performance gap was partly addressed by model-based iterative reconstruction (MBIR). Yet, its "plastic" image appearance and long reconstruction times have limited widespread application. Hybrid iterative reconstruction partially addressed these limitations by blending FBP with MBIR and is currently the state-of-the-art reconstruction technique. In the past 5 years, deep learning reconstruction (DLR) techniques have become increasingly popular. DLR uses artificial intelligence to reconstruct high-quality images from lower-dose CT faster than MBIR. However, the performance of DLR algorithms relies on the quality of data used for model training. Higher-quality training data will become available with photon-counting CT scanners. At the same time, spectral data would greatly benefit from the computational abilities of DLR. This review presents an overview of the principles, technical approaches, and clinical applications of DLR, including metal artifact reduction algorithms. In addition, emerging applications and prospects are discussed.

Evaluation of thin-slice abdominal DECT using deep-learning image reconstruction in 74 keV virtual monoenergetic images: an image quality comparison

PURPOSE

To compare noise, contrast-to-noise ratio (CNR), signal-to-noise ratio (SNR) and image quality using deep-learning image reconstruction (DLIR) vs. adaptive statistical iterative reconstruction (ASIR-V) in 0.625 and 2.5 mm slice thickness gray scale 74 keV virtual monoenergetic (VM) abdominal dual-energy CT (DECT).

METHODS

This retrospective study was approved by the institutional review board and regional ethics committee. We analysed 30 portal-venous phase abdominal fast kV-switching DECT (80/140kVp) scans. Data were reconstructed to ASIR-V 60% and DLIR-High at 74 keV in 0.625 and 2.5 mm slice thickness. Quantitative HU and noise assessment were measured within liver, aorta, adipose tissue and muscle. Two board-certified radiologists evaluated image noise, sharpness, texture and overall quality based on a five-point Likert scale.

RESULTS

DLIR significantly reduced image noise and increased CNR as well as SNR compared to ASIR-V, when slice thickness was maintained (p < 0.001). Slightly higher noise of 5.5-16.2% was measured (p < 0.01) in liver, aorta and muscle tissue at 0.625 mm DLIR compared to 2.5 mm ASIR-V, while noise in adipose tissue was 4.3% lower with 0.625 mm DLIR compared to 2.5 mm ASIR-V (p = 0.08). Qualitative assessments demonstrated significantly improved image quality for DLIR particularly in 0.625 mm images.

CONCLUSIONS

DLIR significantly reduced image noise, increased CNR and SNR and improved image quality in 0.625 mm slice images, when compared to ASIR-V. DLIR may facilitate thinner image slice reconstructions for routine contrast-enhanced abdominal DECT.

Improved image quality in transcatheter aortic valve implantation planning CT using deep learning-based image reconstruction

Background

This study aims to evaluate the impact of a novel deep learning-based image reconstruction (DLIR) algorithm on the image quality in computed tomographic angiography (CTA) for pre-interventional planning of transcatheter aortic valve implantation (TAVI).

Methods

We analyzed 50 consecutive patients (median age 80 years, 25 men) who underwent TAVI planning CT on a 256-dectector-row CT. Images were reconstructed with adaptive statistical iterative reconstruction V (ASIR-V) and DLIR. Intravascular image noise, edge sharpness, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were quantified for ascending aorta, descending aorta, abdominal aorta and iliac arteries. Two readers (one radiologist and one interventional cardiologist) scored task-specific subjective image quality on a five-point scale.

Results

DLIR significantly reduced median image noise by 29-57% at all anatomical locations (all P<0.001). Accordingly, median SNR improved by 44-133% (all P<0.001) and median CNR improved by 44-125% (all P<0.001). DLIR significantly improved subjective image quality for all four pre-specified TAVI-specific tasks (measuring the annulus, assessing valve morphology and calcifications, the coronary ostia, and the suitability of the aorto-iliac access route) for both the radiologist and the interventional cardiologist (P≤0.001). Measurements of the aortic annulus circumference, area and diameter did not differ between ASIR-V and DLIR reconstructions (all P>0.05).

Conclusions

DLIR significantly improves objective and subjective image quality in TAVI planning CT compared to a state-of-the-art iterative reconstruction without affecting measurements of the aortic annulus. This may provide an opportunity for further reductions in contrast medium volume in this population.

Complex Relationship Between Artificial Intelligence and CT Radiation Dose

Concerns over need for CT radiation dose optimization and reduction led to improved scanner efficiency and introduction of several reconstruction techniques and image processing-based software. The latest technologies use artificial intelligence (AI) for CT dose optimization and image quality improvement. While CT dose optimization has and can benefit from AI, variations in scanner technologies, reconstruction methods, and scan protocols can lead to substantial variations in radiation doses and image quality across and within different scanners. These variations in turn can influence performance of AI algorithms being deployed for tasks such as detection, segmentation, characterization, and quantification. We review the complex relationship between AI and CT radiation dose.

Radiation Dose Reduction Opportunities in Vascular Imaging

Computed tomography angiography (CTA) has been the gold standard imaging modality for vascular imaging due to a variety of factors, including the widespread availability of computed tomography (CT) scanners, the ease and speed of image acquisition, and the high sensitivity of CTA for vascular pathology. However, the radiation dose experienced by the patient during imaging has long been a concern of this image acquisition method. Advancements in CT image acquisition techniques in combination with advancements in non-ionizing radiation imaging techniques including magnetic resonance angiography (MRA) and contrast-enhanced ultrasound (CEUS) present growing opportunities to reduce total radiation dose to patients. This review provides an overview of advancements in imaging technology and acquisition techniques that are helping to minimize radiation dose associated with vascular imaging.

Current challenges of implementing artificial intelligence in medical imaging

The idea of using artificial intelligence (AI) in medical practice has gained vast interest due to its potential to revolutionise healthcare systems. However, only some AI algorithms are utilised due to systems' uncertainties, besides the never-ending list of ethical and legal concerns. This paper intends to provide an overview of current AI challenges in medical imaging with an ultimate aim to foster better and effective communication among various stakeholders to encourage AI technology development. We identify four main challenges in implementing AI in medical imaging, supported with consequences and past events when these problems fail to mitigate. Among them is the creation of a robust AI algorithm that is fair, trustable and transparent. Another issue is on data governance, in which best practices in data sharing must be established to promote trust and protect the patients' privacy. Next, stakeholders, such as the government, technology companies and hospital management, should come to a consensus in creating trustworthy AI policies and regulatory frameworks, which is the fourth challenge, to support, encourage and spur innovation in digital AI healthcare technology. Lastly, we discussed the efforts of various organizations such as the World Health Organisation (WHO), American College of Radiology (ACR), European Society of Radiology (ESR) and Radiological Society of North America (RSNA), who are already actively pursuing ethical developments in AI. The efforts by various stakeholders will eventually overcome hurdles and the deployment of AI-driven healthcare applications in clinical practice will become a reality and hence lead to better healthcare services and outcomes.

Efficacy Evaluation of 64-Slice Spiral Computed Tomography Images in Laparoscopic-Assisted Distal Gastrectomy for Gastric Cancer under the Reconstruction Algorithm

This study was aimed to analyze the application value of the filtered back-projection (FBP) reconstruction algorithm of computed tomography (CT) images in laparoscopic-assisted distal gastrectomy. In this study, 56 patients with gastric cancer were selected as research subjects and randomly divided into the control group (CT-guided laparoscopic radical gastrectomy) and the observation group (CT-guided laparoscopic radical gastrectomy with the FBP reconstruction algorithm), with 28 patients in each group. Fourier transform and iterative reconstruction were introduced for comparison, and finally, the postoperative curative effect and adverse events were compared between the two groups. The results showed that the CT image quality score processed by the FBP reconstruction algorithm (4.31 ± 0.31) was significantly higher than that of the iterative reconstruction method (3.5 ± 0.29) and the Fourier transform method (3.97 ± 0.38) (P < 0.05). The incidences of postoperative wound infection and gastric motility disorder (5.88% and 8.16%, respectively) in the observation group were significantly lower than those in the control group (8.21% and 10.82%, respectively) (P < 0.05). The levels of serum interleukin-6 (IL-6) (280.35 ± 15.08 ng/L) and tumor necrosis factor-α (TNF-α) (144.32 ± 10.32 ng/L) in the observation group after the treatment were significantly lower than those in the control group, which were 399.71 ± 14.19 ng/L and 165.33 ± 10.08 ng/L, respectively (P < 0.05). In conclusion, the FBP reconstruction algorithm was better than other algorithms in the processing of gastric cancer CT images. The FBP reconstruction algorithm showed a good reconstruction effect on CT images of gastric cancer; CT images based on this algorithm helped to formulate targeted surgical treatment plans for gastric cancer, showing a high clinical application value.

First Performance Evaluation of an Artificial Intelligence-Based Computer-Aided Detection System for Pulmonary Nodule Evaluation in Dual-Source Photon-Counting Detector CT at Different Low-Dose Levels

OBJECTIVE

The aim of this study was to evaluate the image quality (IQ) and performance of an artificial intelligence (AI)-based computer-aided detection (CAD) system in photon-counting detector computed tomography (PCD-CT) for pulmonary nodule evaluation at different low-dose levels.

MATERIALS AND METHODS

An anthropomorphic chest-phantom containing 14 pulmonary nodules of different sizes (range, 3-12 mm) was imaged on a PCD-CT and on a conventional energy-integrating detector CT (EID-CT). Scans were performed with each of the 3 vendor-specific scanning modes (QuantumPlus [Q+], Quantum [Q], and High Resolution [HR]) at decreasing matched radiation dose levels (volume computed tomography dose index ranging from 1.79 to 0.31 mGy) by adapting IQ levels from 30 to 5. Image noise was measured manually in the chest wall at 8 different locations. Subjective IQ was evaluated by 2 readers in consensus. Nodule detection and volumetry were performed using a commercially available AI-CAD system.

RESULTS

Subjective IQ was superior in PCD-CT compared with EID-CT (P < 0.001), and objective image noise was similar in the Q+ and Q-mode (P > 0.05) and superior in the HR-mode (PCD 55.8 ± 11.7 HU vs EID 74.8 ± 5.4 HU; P = 0.01). High resolution showed the lowest image noise values among PCD modes (P = 0.01). Overall, the AI-CAD system delivered comparable results for lung nodule detection and volumetry between PCD- and dose-matched EID-CT (P = 0.08-1.00), with a mean sensitivity of 95% for PCD-CT and of 86% for dose-matched EID-CT in the lowest evaluated dose level (IQ5). Q+ and Q-mode showed higher false-positive rates than EID-CT at lower-dose levels (IQ10 and IQ5). The HR-mode showed a sensitivity of 100% with a false-positive rate of 1 even at the lowest evaluated dose level (IQ5; CDTIvol, 0.41 mGy).

CONCLUSIONS

Photon-counting detector CT was superior to dose-matched EID-CT in subjective IQ while showing comparable to lower objective image noise. Fully automatized AI-aided nodule detection and volumetry are feasible in PCD-CT, but attention has to be paid to false-positive findings.