Image quality of ultralow-dose chest CT using deep learning techniques: potential superiority of vendor-agnostic post-processing over vendor-specific techniques

Artificial Intelligence Iterative Reconstruction for Dose Reduction in Pediatric Chest CT: A Clinical Assessment via Below 3 Years Patients With Congenital Heart Disease

PURPOSE

To assess the performance of a newly introduced deep learning-based reconstruction algorithm, namely the artificial intelligence iterative reconstruction (AIIR), in reducing the dose of pediatric chest CT by using the image data of below 3-year-old patients with congenital heart disease (CHD).

MATERIALS AND METHODS

The lung image available from routine-dose cardiac CT angiography (CTA) on below 3 years patients with CHD was employed as a reference for evaluating the paired low-dose chest CT. A total of 191 subjects were prospectively enrolled, where the dose for chest CT was reduced to ~0.1 mSv while the cardiac CTA protocol was kept unchanged. The low-dose chest CT images, obtained with the AIIR and the hybrid iterative reconstruction (HIR), were compared in image quality, ie, overall image quality and lung structure depiction, and in diagnostic performance, ie, severity assessment of pneumonia and airway stenosis.

RESULTS

Compared with the reference, lung image quality was not found significantly different on low-dose AIIR images (all P>0.05) but obviously inferior with the HIR (all P<0.05). Compared with the HIR, low-dose AIIR images also achieved a closer pneumonia severity index (AIIR 4.32±3.82 vs. Ref 4.37±3.84, P>0.05; HIR 5.12±4.06 vs. Ref 4.37±3.84, P<0.05) and airway stenosis grading (consistently graded: AIIR 88.5% vs. HIR 56.5% ) to the reference.

CONCLUSIONS

AIIR has the potential for large dose reduction in chest CT of patients below 3 years of age while preserving image quality and achieving diagnostic results nearly equivalent to routine dose scans.

Radiation dose reduction in pediatric computed tomography (CT) using deep convolutional neural network denoising

AIM

We evaluated the quality of noncontrast chest computed tomography (CT) for pediatric patients at two dose levels with and without denoising using a deep convolutional neural network (CNN).

MATERIALS AND METHODS

Forty children underwent noncontrast chest CTs for "chronic cough" using a routine dose (RD) protocol. Images were reconstructed using iterative reconstruction (IR). A validated noise insertion method was used to simulate 20% dose (TD) data for each case. A deep CNN model was trained and validated on 10 cases and then applied to the remaining 30 cases. Three certificate of qualification (CAQ)-certified pediatric radiologists evaluated 30 cases under 4 conditions: (1) RD + IR; (2) RD + CNN; (3) TD + IR; and (4) TD + CNN. Likert scales were used to score subjective image quality (1-5, 5 = excellent) and subjective noise artifact (1-4, 4 = no noise). Images were reviewed for specific findings.

RESULTS

For the 30 patients evaluated (14 female, mean age: 10.8 years, range: 0.17-17), the mean effective dose was 0.46 ± 0.21 mSv for the original RD exam, with an effective dose of 0.09 mSv for the TD exam. Both RD + CNN (3.6 ± 1.1, p < 0.001) and TD + CNN (3.4 ± 0.9, p = 0.023) had higher image quality than RD + IR (3.1 ± 0.9). Both RD + CNN (3.2 ± 0.9, p-value = <0.001) and TD + CNN (2.9 ± 0.6, p-value = 0.001) showed significantly lower subjective noise artifact scores than RD + IR (2.7 ± 0.7). There was excellent intrareader (RD + IR-RD + CNN: mean κ = 0.96, RD + IR-TD + CNN = 0.96, RD + IR-TD + IR = 0.98) and moderate inter-reader reliability (RD + IR: mean κ = 0.55, RD + CNN = 0.50, TD + CNN = 0.54, TD + IR = 0.57) on all 4 image reconstructions.

CONCLUSION

CNN denoising outperforms IR as a means of radiation dose reduction in pediatric CT.

Potential dose reduction and image quality improvement in chest CT with a photon-counting CT compared to a new dual-source CT

PURPOSE

To compare potential dose reduction and quality improvement in chest CT images with Photon-Counting CT (PCCT) versus a Dual-Source CT (DSCT).

METHODS

Acquisitions on phantoms were performed on a DSCT and a PCCT at 5 dose levels (9.5/7.5/6.0/2.5/0.4 mGy). Noise power spectrum (NPS) and task-based transfer function (TTF) were calculated to assess noise magnitude and noise texture (fav) and spatial resolution (f50), respectively. Computed detectability indexes (d') modelled the detection of 2 chest lesions: subsolid pulmonary nodules (SPN) and high-contrast pulmonary nodules (HCN). Two radiologists subjectively assessed the quality of chest images on an anthropomorphic phantom.

RESULTS

For all dose levels, noise magnitude was significantly lower with PCCT than with DSCT (-44.7 ± 3.0 %; p < 0.05). Identical outcomes were found for noise texture (fav; -6.2 ± 0.5 %; p < 0.05). f50 values were significantly higher with DSCT than with PCCT from 9.5 to 6 mGy for iodine insert (p < 0.05) and from 7.5 to 2.5 mGy for air insert (p < 0.05), but similar for both inserts at other dose levels. For all dose levels, d' values were significantly higher with PCCT than DSCT (71.9 ± 5.4 % for HCN and 65.6 ± 13.5 % for SPN). From 9.5 to 2.5 mGy, the potential dose reduction was -59.0 ± 3.9 % for both lesions with PCCT compared to DSCT. Chest images were rated satisfactory for clinical use by the radiologists with both CTs for all dose levels, except at 0.4 mGy.

CONCLUSION

Noise magnitude and detectability of chest lesions were better with PCCT than with the DSCT. PCCT may offer great potential for dose reduction in patients undergoing chest CT examinations.

Evaluation of a Vendor-Agnostic Deep Learning Model for Noise Reduction and Image Quality Improvement in Dental CBCT

BACKGROUND/OBJECTIVES

To assess the impact of a vendor-agnostic deep learning model (DLM) on image quality parameters and noise reduction in dental cone-beam computed tomography (CBCT) reconstructions.

METHODS

This retrospective study was conducted on CBCT scans of 93 patients (41 males and 52 females, mean age 41.2 years, SD 15.8 years) from a single center using the inclusion criteria of standard radiation dose protocol images. Objective and subjective image quality was assessed in three predefined landmarks through contrast-to-noise ratio (CNR) measurements and visual assessment using a 5-point scale by three experienced readers. The inter-reader reliability and repeatability were calculated.

RESULTS

Eighty patients (30 males and 50 females; mean age 41.5 years, SD 15.94 years) were included in this study. The CNR in DLM reconstructions was significantly greater than in native reconstructions, and the mean CNR in regions of interest 1-3 (ROI1-3) in DLM images was 11.12 ± 9.29, while in the case of native reconstructions, it was 7.64 ± 4.33 (p < 0.001). The noise level in native reconstructions was significantly higher than in the DLM reconstructions, and the mean noise level in ROI1-3 in native images was 45.83 ± 25.89, while in the case of DLM reconstructions, it was 35.61 ± 24.28 (p < 0.05). Subjective image quality assessment revealed no statistically significant differences between native and DLM reconstructions.

CONCLUSIONS

The use of deep learning-based image reconstruction algorithms for CBCT imaging of the oral cavity can improve image quality by enhancing the CNR and lowering the noise.

Improving diagnostic confidence in low-dose dual-energy CTE with low energy level and deep learning reconstruction

OBJECTIVE

To demonstrate the value of using 50 keV virtual monochromatic images with deep learning image reconstruction (DLIR) in low-dose dual-energy CT enterography (CTE).

METHODS

In this prospective study, 114 participants (62 % M; 41.9 ± 16 years) underwent dual-energy CTE. The early-enteric phase was performed using standard-dose (noise index (NI): 8) and images were reconstructed at 70 keV and 50 keV with 40 % strength ASIR-V (ASIR-V40%). The late-enteric phase used low-dose (NI: 12) and images were reconstructed at 50 keV with ASIR-V40%, and DLIR at medium (DLIR-M) and high strength (DLIR-H). Image standard deviation (SD), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), edge-rise-slope (ERS) were computed. The quantitative comb sign score was calculated for the 27 patients with Crohn's disease. The subjective noise, image contrast, display of rectus artery were scored using a 5-point scale by two radiologists blindly.

RESULTS

Effective dose was reduced by 50 % (P < 0.001) in the late-enteric phase to 3.26 mSv. The lower-dose 50 keV-DLIR-H images (SD:17.7 ± 0.5HU) had similar image noise (P = 0.97) as the standard-dose 70 keV-ASIR-V40% images (SD:17.7 ± 0.73HU), but with higher (P < 0.001) SNR, CNR, ERS and quantitative comb sign score (5.7 ± 0.17, 1.8 ± 0.12, 156.04 ± 5.21 and 5.05 ± 0.73, respectively). Furthermore, the lower-dose 50 keV-DLIR-H images obtained the highest score in the rectus artery visibility (4.27 ± 0.6).

CONCLUSIONS

The 50 keV images in dual-energy CTE with DLIR provides high-quality images, with a 50 % reduction in radiation dose. Images with high contrast and density resolutions significantly enhance the diagnostic confidence of Crohn's disease and are essential for the clinical development of individualized treatment plans.

Improved overall image quality in low-dose dual-energy computed tomography enterography using deep-learning image reconstruction

OBJECTIVE

To demonstrate the clinical advantages of a deep-learning image reconstruction (DLIR) in low-dose dual-energy computed tomography enterography (DECTE) by comparing images with standard-dose adaptive iterative reconstruction-Veo (ASIR-V) images.

METHODS

In this Institutional review board approved prospective study, 86 participants who underwent DECTE were enrolled. The early-enteric phase scan was performed using standard-dose (noise index: 8) and images were reconstructed at 5 mm and 1.25 mm slice thickness with ASIR-V at a level of 40% (ASIR-V40%). The late-enteric phase scan used low-dose (noise index: 12) and images were reconstructed at 1.25 mm slice thickness with ASIR-V40%, and DLIR at medium (DLIR-M) and high (DLIR-H). The 70 keV monochromatic images were used for image comparison and analysis. For objective assessment, image noise, artifact index, SNR and CNR were measured. For subjective assessment, subjective noise, image contrast, bowel wall sharpness, mesenteric vessel clarity, and small structure visibility were scored by two radiologists blindly. Radiation dose was compared between the early- and late-enteric phases.

RESULTS

Radiation dose was reduced by 50% in the late-enteric phase [(6.31 ± 1.67) mSv] compared with the early-enteric phase [(3.01 ± 1.09) mSv]. For the 1.25 mm images, DLIR-M and DLIR-H significantly improved both objective and subjective image quality compared to those with ASIR-V40%. The low-dose 1.25 mm DLIR-H images had similar image noise, SNR, CNR values as the standard-dose 5 mm ASIR-V40% images, but significantly higher scores in image contrast [5(5-5), P < 0.05], bowel wall sharpness [5(5-5), P < 0.05], mesenteric vessel clarity [5(5-5), P < 0.05] and small structure visibility [5(5-5), P < 0.05].

CONCLUSIONS

DLIR significantly reduces image noise at the same slice thickness, but significantly improves spatial resolution and lesion conspicuity with thinner slice thickness in DECTE, compared to conventional ASIR-V40% 5 mm images, all while providing 50% radiation dose reduction.

Deep Learning-Based Reconstruction Algorithm With Lung Enhancement Filter for Chest CT: Effect on Image Quality and Ground Glass Nodule Sharpness

OBJECTIVE

To assess the effect of a new lung enhancement filter combined with deep learning image reconstruction (DLIR) algorithm on image quality and ground-glass nodule (GGN) sharpness compared to hybrid iterative reconstruction or DLIR alone.

MATERIALS AND METHODS

Five artificial spherical GGNs with various densities (-250, -350, -450, -550, and -630 Hounsfield units) and 10 mm in diameter were placed in a thorax anthropomorphic phantom. Four scans at four different radiation dose levels were performed using a 256-slice CT (Revolution Apex CT, GE Healthcare). Each scan was reconstructed using three different reconstruction algorithms: adaptive statistical iterative reconstruction-V at a level of 50% (AR50), Truefidelity (TF), which is a DLIR method, and TF with a lung enhancement filter (TF + Lu). Thus, 12 sets of reconstructed images were obtained and analyzed. Image noise, signal-to-noise ratio, and contrast-to-noise ratio were compared among the three reconstruction algorithms. Nodule sharpness was compared among the three reconstruction algorithms using the full-width at half-maximum value. Furthermore, subjective image quality analysis was performed.

RESULTS

AR50 demonstrated the highest level of noise, which was decreased by using TF + Lu and TF alone (P = 0.001). TF + Lu significantly improved nodule sharpness at all radiation doses compared to TF alone (P = 0.001). The nodule sharpness of TF + Lu was similar to that of AR50. Using TF alone resulted in the lowest nodule sharpness.

CONCLUSION

Adding a lung enhancement filter to DLIR (TF + Lu) significantly improved the nodule sharpness compared to DLIR alone (TF). TF + Lu can be an effective reconstruction technique to enhance image quality and GGN evaluation in ultralow-dose chest CT scans.

Automated Opportunistic Osteoporosis Screening Using Low-Dose Chest CT among Individuals Undergoing Lung Cancer Screening in a Korean Population

Opportunistic osteoporosis screening using deep learning (DL) analysis of low-dose chest CT (LDCT) scans is a potentially promising approach for the early diagnosis of this condition. We explored bone mineral density (BMD) profiles across all adult ages and prevalence of osteoporosis using LDCT with DL in a Korean population. This retrospective study included 1915 participants from two hospitals who underwent LDCT during general health checkups between 2018 and 2021. Trabecular volumetric BMD of L1-2 was automatically calculated using DL and categorized according to the American College of Radiology quantitative computed tomography diagnostic criteria. BMD decreased with age in both men and women. Women had a higher peak BMD in their twenties, but lower BMD than men after 50. Among adults aged 50 and older, the prevalence of osteoporosis and osteopenia was 26.3% and 42.0%, respectively. Osteoporosis prevalence was 18.0% in men and 34.9% in women, increasing with age. Compared to previous data obtained using dual-energy X-ray absorptiometry, the prevalence of osteoporosis, particularly in men, was more than double. The automated opportunistic BMD measurements using LDCT can effectively predict osteoporosis for opportunistic screening and identify high-risk patients. Patients undergoing lung cancer screening may especially profit from this procedure requiring no additional imaging or radiation exposure.

Clinical value of deep learning image reconstruction on the diagnosis of pulmonary nodule for ultra-low-dose chest CT imaging

PURPOSE

To compare the image quality and pulmonary nodule detectability between deep learning image reconstruction (DLIR) and adaptive statistical iterative reconstruction-Veo (ASIR-V) in ultra-low-dose CT (ULD-CT).

METHODS

142 participants required lung examination who underwent simultaneously ULD-CT (UL-A, 0.57 ± 0.04 mSv or UL-B, 0.33 ± 0.03 mSv), and standard CT (SDCT, 4.32 ± 0.33 mSv) plain scans were included in this prospective study. SDCT was the reference standard using ASIR-V at 50% strength (50%ASIR-V). ULD-CT was reconstructed with 50%ASIR-V, DLIR at medium and high strength (DLIR-M, DLIR-H). The noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and subjective scores were measured. The presence and accuracy of nodules were analyzed using a combination of a deep learning-based nodule evaluation system and a radiologist.

RESULTS

A total of 710 nodules were detected by SDCT, including 358 nodules in UL-A and 352 nodules in UL-B. DLIR-H exhibited superior noise, SNR, and CNR performance, and achieved comparable or even higher subjective scores compared to 50%ASIR-V in ULD-CT. Nodules sensitivity detection of 50%ASIR-V, DLIR-M, and DLIR-H in ULD-CT were identical (96.90%). In multivariate analysis, body mass index (BMI), nodule diameter, and type were independent predictors for the sensitivity of nodule detection (p<.001). DLIR-H provided a lower absolute percent error (APE) in volume (3.10% ± 95.11% vs 8.29% ± 99.14%) compared to 50%ASIR-V of ULD-CT (P<.001).

CONCLUSIONS

ULD-CT scanning has a high sensitivity for detecting pulmonary nodules. Compared with ASIR-V, DLIR can significantly reduce image noise, and improve image quality, and accuracy of the nodule measurement in ULD-CT.

The Impact of AI on Metal Artifacts in CBCT Oral Cavity Imaging

OBJECTIVE

This study aimed to assess the impact of artificial intelligence (AI)-driven noise reduction algorithms on metal artifacts and image quality parameters in cone-beam computed tomography (CBCT) images of the oral cavity.

MATERIALS AND METHODS

This retrospective study included 70 patients, 61 of whom were analyzed after excluding those with severe motion artifacts. CBCT scans, performed using a Hyperion X9 PRO 13 × 10 CBCT machine, included images with dental implants, amalgam fillings, orthodontic appliances, root canal fillings, and crowns. Images were processed with the ClariCT.AI deep learning model (DLM) for noise reduction. Objective image quality was assessed using metrics such as the differentiation between voxel values (ΔVVs), the artifact index (AIx), and the contrast-to-noise ratio (CNR). Subjective assessments were performed by two experienced readers, who rated overall image quality and artifact intensity on predefined scales.

RESULTS

Compared with native images, DLM reconstructions significantly reduced the AIx and increased the CNR (p < 0.001), indicating improved image clarity and artifact reduction. Subjective assessments also favored DLM images, with higher ratings for overall image quality and lower artifact intensity (p < 0.001). However, the ΔVV values were similar between the native and DLM images, indicating that while the DLM reduced noise, it maintained the overall density distribution. Orthodontic appliances produced the most pronounced artifacts, while implants generated the least.

CONCLUSIONS

AI-based noise reduction using ClariCT.AI significantly enhances CBCT image quality by reducing noise and metal artifacts, thereby improving diagnostic accuracy and treatment planning. Further research with larger, multicenter cohorts is recommended to validate these findings.

Diagnostic performance and image quality of an image-based denoising algorithm applied to radiation dose-reduced CT in diagnosing acute appendicitis

PURPOSE

To evaluate diagnostic performance and image quality of ultralow-dose CT (ULDCT) in diagnosing acute appendicitis with an image-based deep-learning denoising algorithm (IDLDA).

METHODS

This retrospective multicenter study included 180 patients (mean ± standard deviation, 29 ± 9 years; 91 female) who underwent contrast-enhanced 2-mSv CT for suspected appendicitis from February 2014 to August 2016. We simulated ULDCT from 2-mSv CT, reducing the dose by at least 50%. Then we applied an IDLDA on ULDCT to produce denoised ULDCT (D-ULDCT). Six radiologists with different experience levels (three board-certified radiologists and three residents) independently reviewed the ULDCT and D-ULDCT. They rated the likelihood of appendicitis and subjective image qualities (subjective image noise, diagnostic acceptability, and artificial sensation). One radiologist measured image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). We used the receiver operating characteristic (ROC) analyses, Wilcoxon's signed-rank tests, and paired t-tests.

RESULTS

The area under the ROC curves (AUC) for diagnosing appendicitis ranged 0.90-0.97 for ULDCT and 0.94-0.97 for D-ULDCT. The AUCs of two residents were significantly higher on D-ULDCT (AUC difference = 0.06 [95% confidence interval, 0.01-0.11; p = .022] and 0.05 [0.00-0.10; p = .046], respectively). D-ULDCT provided better subjective image noise and diagnostic acceptability to all six readers. However, the response of board-certified radiologists and residents differed in artificial sensation (all p ≤ .003). D-ULDCT showed significantly lower image noise, higher SNR, and higher CNR (all p < .001).

CONCLUSION

An IDLDA can provide better ULDCT image quality and enhance diagnostic performance for less-experienced radiologists.

Assessment of image quality and impact of deep learning-based software in non-contrast head CT scans

In this retrospective study, we aimed to assess the objective and subjective image quality of different reconstruction techniques and a deep learning-based software on non-contrast head computed tomography (CT) images. In total, 152 adult head CT scans (77 female, 75 male; mean age 69.4 ± 18.3 years) obtained from three different CT scanners using different protocols between March and April 2021 were included. CT images were reconstructed using filtered-back projection (FBP), iterative reconstruction (IR), and post-processed using a deep learning-based algorithm (PS). Post-processing significantly reduced noise in FBP-reconstructed images (up to 15.4% reduction) depending on the protocol, leading to improvements in signal-to-noise ratio of up to 19.7%. However, when deep learning-based post-processing was applied to FBP images compared to IR alone, the differences were inconsistent and partly non-significant, which appeared to be protocol or site specific. Subjective assessments showed no significant overall improvement in image quality for all reconstructions and post-processing. Inter-rater reliability was low and preferences varied. Deep learning-based denoising software improved objective image quality compared to FBP in routine head CT. A significant difference compared to IR was observed for only one protocol. Subjective assessments did not indicate a significant clinical impact in terms of improved subjective image quality, likely due to the low noise levels in full-dose images.

Noise-Optimized CBCT Imaging of Temporomandibular Joints-The Impact of AI on Image Quality

Background: Temporomandibular joint disorder (TMD) is a common medical condition. Cone beam computed tomography (CBCT) is effective in assessing TMD-related bone changes, but image noise may impair diagnosis. Emerging deep learning reconstruction algorithms (DLRs) could minimize noise and improve CBCT image clarity. This study compares standard and deep learning-enhanced CBCT images for image quality in detecting osteoarthritis-related degeneration in TMJs (temporomandibular joints). This study analyzed CBCT images of patients with suspected temporomandibular joint degenerative joint disease (TMJ DJD). Methods: The DLM reconstructions were performed with ClariCT.AI software. Image quality was evaluated objectively via CNR in target areas and subjectively by two experts using a five-point scale. Both readers also assessed TMJ DJD lesions. The study involved 50 patients with a mean age of 28.29 years. Results: Objective analysis revealed a significantly better image quality in DLM reconstructions (CNR levels; p < 0.001). Subjective assessment showed high inter-reader agreement (κ = 0.805) but no significant difference in image quality between the reconstruction types (p = 0.055). Lesion counts were not significantly correlated with the reconstruction type (p > 0.05). Conclusions: The analyzed DLM reconstruction notably enhanced the objective image quality in TMJ CBCT images but did not significantly alter the subjective quality or DJD lesion diagnosis. However, the readers favored DLM images, indicating the potential for better TMD diagnosis with CBCT, meriting more study.

Deep Learning-Based Kernel Adaptation Enhances Quantification of Emphysema on Low-Dose Chest CT for Predicting Long-Term Mortality

OBJECTIVES

The aim of this study was to ascertain the predictive value of quantifying emphysema using low-dose computed tomography (LDCT) post deep learning-based kernel adaptation on long-term mortality.

MATERIALS AND METHODS

This retrospective study investigated LDCTs obtained from asymptomatic individuals aged 60 years or older during health checkups between February 2009 and December 2016. These LDCTs were reconstructed using a 1- or 1.25-mm slice thickness alongside high-frequency kernels. A deep learning algorithm, capable of generating CT images that resemble standard-dose and low-frequency kernel images, was applied to these LDCTs. To quantify emphysema, the lung volume percentage with an attenuation value less than or equal to -950 Hounsfield units (LAA-950) was gauged before and after kernel adaptation. Low-dose chest CTs with LAA-950 exceeding 6% were deemed emphysema-positive according to the Fleischner Society statement. Survival data were sourced from the National Registry Database at the close of 2021. The risk of nonaccidental death, excluding causes such as injury or poisoning, was explored according to the emphysema quantification results using multivariate Cox proportional hazards models.

RESULTS

The study comprised 5178 participants (mean age ± SD, 66 ± 3 years; 3110 males). The median LAA-950 (18.2% vs 2.6%) and the proportion of LDCTs with LAA-950 exceeding 6% (96.3% vs 39.3%) saw a significant decline after kernel adaptation. There was no association between emphysema quantification before kernel adaptation and the risk of nonaccidental death. Nevertheless, after kernel adaptation, higher LAA-950 (hazards ratio for 1% increase, 1.01; P = 0.045) and LAA-950 exceeding 6% (hazards ratio, 1.36; P = 0.008) emerged as independent predictors of nonaccidental death, upon adjusting for age, sex, and smoking status.

CONCLUSIONS

The application of deep learning for kernel adaptation proves instrumental in quantifying pulmonary emphysema on LDCTs, establishing itself as a potential predictive tool for long-term nonaccidental mortality in asymptomatic individuals.

Image Quality and Diagnostic Performance of Low-Dose Liver CT with Deep Learning Reconstruction versus Standard-Dose CT

Purpose To compare the image quality and diagnostic capability in detecting malignant liver tumors of low-dose CT (LDCT, 33% dose) with deep learning-based denoising (DLD) and standard-dose CT (SDCT, 100% dose) with model-based iterative reconstruction (MBIR). Materials and Methods In this prospective, multicenter, noninferiority study, individuals referred for liver CT scans were enrolled from three tertiary referral hospitals between February 2021 and August 2022. All liver CT scans were conducted using a dual-source scanner with the dose split into tubes A (67% dose) and B (33% dose). Blended images from tubes A and B were created using MBIR to produce SDCT images, whereas LDCT images used data from tube B and were reconstructed with DLD. The noise in liver images was measured and compared between imaging techniques. The diagnostic performance of each technique in detecting malignant liver tumors was evaluated by three independent radiologists using jackknife alternative free-response receiver operating characteristic analysis. Noninferiority of LDCT compared with SDCT was declared when the lower limit of the 95% CI for the difference in figure of merit (FOM) was greater than -0.10. Results A total of 296 participants (196 men, 100 women; mean age, 60.5 years ± 13.3 [SD]) were included. The mean noise level in the liver was significantly lower for LDCT (10.1) compared with SDCT (10.7) (P < .001). Diagnostic performance was assessed in 246 participants (108 malignant tumors in 90 participants). The reader-averaged FOM was 0.880 for SDCT and 0.875 for LDCT (P = .35). The difference fell within the noninferiority margin (difference, -0.005 [95% CI: -0.024, 0.012]). Conclusion Compared with SDCT with MBIR, LDCT using 33% of the standard radiation dose had reduced image noise and comparable diagnostic performance in detecting malignant liver tumors. Keywords: CT, Abdomen/GI, Liver, Comparative Studies, Diagnosis, Reconstruction Algorithms Clinical trial registration no. NCT05804799 © RSNA, 2024 Supplemental material is available for this article.

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.

AntiHalluciNet: A Potential Auditing Tool of the Behavior of Deep Learning Denoising Models in Low-Dose Computed Tomography

Gaining the ability to audit the behavior of deep learning (DL) denoising models is of crucial importance to prevent potential hallucinations and adversarial clinical consequences. We present a preliminary version of AntiHalluciNet, which is designed to predict spurious structural components embedded in the residual noise from DL denoising models in low-dose CT and assess its feasibility for auditing the behavior of DL denoising models. We created a paired set of structure-embedded and pure noise images and trained AntiHalluciNet to predict spurious structures in the structure-embedded noise images. The performance of AntiHalluciNet was evaluated by using a newly devised residual structure index (RSI), which represents the prediction confidence based on the presence of structural components in the residual noise image. We also evaluated whether AntiHalluciNet could assess the image fidelity of a denoised image by using only a noise component instead of measuring the SSIM, which requires both reference and test images. Then, we explored the potential of AntiHalluciNet for auditing the behavior of DL denoising models. AntiHalluciNet was applied to three DL denoising models (two pre-trained models, RED-CNN and CTformer, and a commercial software, ClariCT.AI [version 1.2.3]), and whether AntiHalluciNet could discriminate between the noise purity performances of DL denoising models was assessed. AntiHalluciNet demonstrated an excellent performance in predicting the presence of structural components. The RSI values for the structural-embedded and pure noise images measured using the 50% low-dose dataset were 0.57 ± 31 and 0.02 ± 0.02, respectively, showing a substantial difference with a p-value < 0.0001. The AntiHalluciNet-derived RSI could differentiate between the quality of the degraded denoised images, with measurement values of 0.27, 0.41, 0.48, and 0.52 for the 25%, 50%, 75%, and 100% mixing rates of the degradation component, which showed a higher differentiation potential compared with the SSIM values of 0.9603, 0.9579, 0.9490, and 0.9333. The RSI measurements from the residual images of the three DL denoising models showed a distinct distribution, being 0.28 ± 0.06, 0.21 ± 0.06, and 0.15 ± 0.03 for RED-CNN, CTformer, and ClariCT.AI, respectively. AntiHalluciNet has the potential to predict the structural components embedded in the residual noise from DL denoising models in low-dose CT. With AntiHalluciNet, it is feasible to audit the performance and behavior of DL denoising models in clinical environments where only residual noise images are available.

Deep Learning-Based Motion Correction in Projection Domain for Coronary Computed Tomography Angiography: A Clinical Evaluation

OBJECTIVE

This study aimed to evaluate the clinical performance of a deep learning-based motion correction algorithm (MCA) in projection domain for coronary computed tomography angiography (CCTA).

METHODS

A total of 192 patients who underwent CCTA examinations were included and divided into 2 groups based on the average heart rate (HR): group 1, 82 patients with HR of <75 beats per minute; group 2, 110 patients with HR of ≥75 beats per minute. The CCTA images were reconstructed with and without MCA. The subjective image quality was graded in terms of vessel visualization, sharpness, diagnostic confidence, and overall image quality using a 5-point scale, where cases with all scores of ≥3 were deemed interpretable. Objective image quality was measured through signal-to-noise ratio and contrast-to-noise ratio in regions relative to the vessels. The image quality scores for 2 reconstructions and effective dose between 2 groups were compared.

RESULTS

The mean effective dose was similar between 2 groups. Neither group showed significant difference on objective image quality for 2 reconstructions. Images reconstructed with and without MCA were both found interpretable for group 1, whereas the subjective image quality was significantly improved by the MCA for all 4 metrics in group 2, with the interpretability increased from 80.91% to 99.09%. Compared with group 1, group 2 showed similar interpretability and diagnostic confidence, despite inferior overall image quality.

CONCLUSIONS

In CCTA examinations, the deep learning-based MCA is capable of improving the image quality and diagnostic confidence for patients with increased HR to a similar level as for those with low HR.

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.

The impact of deep learning reconstruction in low dose computed tomography on the evaluation of interstitial lung disease

To evaluate the effect of the deep learning model reconstruction (DLM) method in terms of image quality and diagnostic agreement in low-dose computed tomography (LDCT) for interstitial lung disease (ILD), 193 patients who underwent LDCT for suspected ILD were retrospectively reviewed. Datasets were reconstructed using filtered back projection (FBP), adaptive statistical iterative reconstruction Veo (ASiR-V), and DLM. For image quality analysis, the signal, noise, signal-to-noise ratio (SNR), blind/referenceless image spatial quality evaluator (BRISQUE), and visual scoring were evaluated. Also, CT patterns of usual interstitial pneumonia (UIP) were classified according to the 2022 idiopathic pulmonary fibrosis (IPF) diagnostic criteria. The differences between CT images subjected to FBP, ASiR-V 30%, and DLM were evaluated. The image noise and BRISQUE scores of DLM images was lower and SNR was higher than that of the ASiR-V and FBP images (ASiR-V vs. DLM, p < 0.001 and FBP vs. DLR-M, p < 0.001, respectively). The agreement of the diagnostic categorization of IPF between the three reconstruction methods was almost perfect (κ  =  0.992, CI 0.990-0.994). Image quality was improved with DLM compared to ASiR-V and FBP.

75% radiation dose reduction using deep learning reconstruction on low-dose chest CT

OBJECTIVE

Few studies have explored the clinical feasibility of using deep-learning reconstruction to reduce the radiation dose of CT. We aimed to compare the image quality and lung nodule detectability between chest CT using a quarter of the low dose (QLD) reconstructed with vendor-agnostic deep-learning image reconstruction (DLIR) and conventional low-dose (LD) CT reconstructed with iterative reconstruction (IR).

MATERIALS AND METHODS

We retrospectively collected 100 patients (median age, 61 years [IQR, 53-70 years]) who received LDCT using a dual-source scanner, where total radiation was split into a 1:3 ratio. QLD CT was generated using a quarter dose and reconstructed with DLIR (QLD-DLIR), while LDCT images were generated using a full dose and reconstructed with IR (LD-IR). Three thoracic radiologists reviewed subjective noise, spatial resolution, and overall image quality, and image noise was measured in five areas. The radiologists were also asked to detect all Lung-RADS category 3 or 4 nodules, and their performance was evaluated using area under the jackknife free-response receiver operating characteristic curve (AUFROC).

RESULTS

The median effective dose was 0.16 (IQR, 0.14-0.18) mSv for QLD CT and 0.65 (IQR, 0.57-0.71) mSv for LDCT. The radiologists' evaluations showed no significant differences in subjective noise (QLD-DLIR vs. LD-IR, lung-window setting; 3.23 ± 0.19 vs. 3.27 ± 0.22; P = .11), spatial resolution (3.14 ± 0.28 vs. 3.16 ± 0.27; P = .12), and overall image quality (3.14 ± 0.21 vs. 3.17 ± 0.17; P = .15). QLD-DLIR demonstrated lower measured noise than LD-IR in most areas (P < .001 for all). No significant difference was found between QLD-DLIR and LD-IR for the sensitivity (76.4% vs. 72.2%; P = .35) or the AUFROCs (0.77 vs. 0.78; P = .68) in detecting Lung-RADS category 3 or 4 nodules. Under a noninferiority limit of -0.1, QLD-DLIR showed noninferior detection performance (95% CI for AUFROC difference, -0.04 to 0.06).

CONCLUSION

QLD-DLIR images showed comparable image quality and noninferior nodule detectability relative to LD-IR images.

Artificial Intelligence in Lung Cancer Screening: The Future Is Now

Lung cancer has one of the worst morbidity and fatality rates of any malignant tumour. Most lung cancers are discovered in the middle and late stages of the disease, when treatment choices are limited, and patients' survival rate is low. The aim of lung cancer screening is the identification of lung malignancies in the early stage of the disease, when more options for effective treatments are available, to improve the patients' outcomes. The desire to improve the efficacy and efficiency of clinical care continues to drive multiple innovations into practice for better patient management, and in this context, artificial intelligence (AI) plays a key role. AI may have a role in each process of the lung cancer screening workflow. First, in the acquisition of low-dose computed tomography for screening programs, AI-based reconstruction allows a further dose reduction, while still maintaining an optimal image quality. AI can help the personalization of screening programs through risk stratification based on the collection and analysis of a huge amount of imaging and clinical data. A computer-aided detection (CAD) system provides automatic detection of potential lung nodules with high sensitivity, working as a concurrent or second reader and reducing the time needed for image interpretation. Once a nodule has been detected, it should be characterized as benign or malignant. Two AI-based approaches are available to perform this task: the first one is represented by automatic segmentation with a consequent assessment of the lesion size, volume, and densitometric features; the second consists of segmentation first, followed by radiomic features extraction to characterize the whole abnormalities providing the so-called "virtual biopsy". This narrative review aims to provide an overview of all possible AI applications in lung cancer screening.

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.

Improving spatial resolution and diagnostic confidence with thinner slice and deep learning image reconstruction in contrast-enhanced abdominal CT

OBJECTIVE

To evaluate image quality and diagnostic confidence improvement using a thin slice and a deep learning image reconstruction (DLIR) in contrast-enhanced abdominal CT.

METHODS

Forty patients with hepatic lesions in enhanced abdominal CT were retrospectively analyzed. Images in the portal phase were reconstructed at 5 mm and 1.25 mm slice thickness using the 50% adaptive statistical iterative reconstruction (ASIR-V) (ASIR-V50%) and at 1.25 mm using DLIR at medium (DLIR-M) and high (DLIR-H) settings. CT number and standard deviation of the hepatic parenchyma, spleen, portal vein, and subcutaneous fat were measured, and contrast-to-noise ratio (CNR) was calculated. Edge-rise-slope (ERS) was measured on the portal vein to reflect spatial resolution and the CT number skewness on liver parenchyma was calculated to reflect image texture. Two radiologists blindly assessed the overall image quality including subjective noise, image contrast, visibility of small structures using a 5-point scale, and object sharpness and lesion contour using a 4-point scale.

RESULTS

For the 1.25-mm images, DLIR significantly reduced image noise, improved CNR and overall subjective image quality compared to ASIR-V50%. Compared to the 5-mm ASIR-V50% images, DLIR images had significantly higher scores in the visibility and contour for small structures and lesions; as well as significantly higher ERS and lower CT number skewness. At a quarter of the signal strength, the 1.25-mm DLIR-H images had a similar subjective noise score as the 5-mm ASIR-V50% images.

CONCLUSION

DLIR significantly reduces image noise and maintains a more natural image texture; image spatial resolution and diagnostic confidence can be improved using thin slice images and DLIR in abdominal CT.

KEY POINTS

• DLIR further reduces image noise compared with ASIR-V while maintaining favorable image texture. • In abdominal CT, thinner slice images improve image spatial resolution and small object visualization but suffer from higher image noise. • Thinner slice images combined with DLIR in abdominal CT significantly suppress image noise for detecting low-density lesions while significantly improving image spatial resolution and overall image quality.

Lung-Optimized Deep-Learning-Based Reconstruction for Ultralow-Dose CT

RATIONALE AND OBJECTIVES

To evaluate the image properties of lung-specialized deep-learning-based reconstruction (DLR) and its applicability in ultralow-dose CT (ULDCT) relative to hybrid- (HIR) and model-based iterative-reconstructions (MBIR).

MATERIALS AND METHODS

An anthropomorphic chest phantom was scanned on a 320-row scanner at 50-mA (low-dose-CT 1 [LDCT-1]), 25-mA (LDCT-2), and 10-mA (ULDCT). LDCT were reconstructed with HIR; ULDCT images were reconstructed with HIR (ULDCT-HIR), MBIR (ULDCT-MBIR), and DLR (ULDCT-DLR). Image noise and contrast-to-noise ratio (CNR) were quantified. With the LDCT images as reference standards, ULDCT image qualities were subjectively scored on a 5-point scale (1 = substantially inferior to LDCT-2, 3 = comparable to LDCT-2, 5 = comparable to LDCT-1). For task-based image quality analyses, a physical evaluation phantom was scanned at seven doses to achieve the noise levels equivalent to chest phantom; noise power spectrum (NPS) and task-based transfer function (TTF) were evaluated. Clinical ULDCT (10-mA) images obtained in 14 nonobese patients were reconstructed with HIR, MBIR, and DLR; the subjective acceptability was ranked.

RESULTS

Image noise was lower and CNR was higher in ULDCT-DLR and ULDCT-MBIR than in LDCT-1, LDCT-2, and ULDCT-HIR (p < 0.01). The overall quality of ULDCT-DLR was higher than of ULDCT-HIR and ULDCT-MBIR (p < 0.01), and almost comparable with that of LDCT-2 (mean score: 3.4 ± 0.5). DLR yielded the highest NPS peak frequency and TTF_50% for high-contrast object. In clinical ULDCT images, the subjective acceptability of DLR was higher than of HIR and MBIR (p < 0.01).

CONCLUSION

DLR optimized for lung CT improves image quality and provides possible greater dose optimization opportunity than HIR and MBIR.

Performance of 1-mm non-gated low-dose chest computed tomography using deep learning-based noise reduction for coronary artery calcium scoring

OBJECTIVE

To investigate performance of 1-mm, sharp kernel, low-dose chest computed tomography (LDCT) for coronary artery calcium scoring (CACS) using deep learning (DL)-based denoising technique.

METHODS

This retrospective, intra-individual comparative study consisted of four image datasets of 131 participants who underwent LDCT and calcium CT on the same day between January and February 2020; 1-mm LDCT with DL, 1-mm LDCT with iterative reconstruction (IR), 3-mm LDCT, and calcium CT. CACS from calcium CT were considered as reference and CACS were categorized as 0, 1-10, 11-100, 101-400, and > 400. We compared CACS from LDCTs with that from calcium CT.

RESULTS

Mean CACS was 104.8 ± 249.1 and proportion of positive CACS was 45% (59/131). CACS from LDCT images tended to be underestimated than those from calcium CT: 1-mm LDCT with DL (93.5 ± 249.6, p = 0.002), 1-mm LDCT with IR (94.7 ± 249.9, p < 0.001), and 3-mm LDCT (90.3 ± 245.3, p = 0.004). All LDCT datasets showed excellent agreement with calcium CT: intraclass correlation coefficient (ICC) = 0.961 (95% confidence interval (CI), 0.945-0.972) for DL, 0.969 (95% CI, 0.956-0.978) for IR, and 0.952 (95% CI, 0.932-0.966) for 3-mm LDCT; weighted kappa for CACS classification, 0.930 (95% CI, 0.893-0.966) for 1-mm LDCT with DL, 0.908 (95% CI, 0.866-0.950) for 1-mm LDCT with IR, and 0.846 (95% CI, 0.780-0.912) for 3-mm LDCT. The accuracy of CACS classification of 1-mm LDCT with DL (90%) tended to be better than 1-mm LDCT with IR (87%) and 3-mm LDCT (84.7%) (p = 0.10).

CONCLUSION

DL-based noise reduction algorithm can offer reliable calcium scores in 1-mm LDCT reconstructed with sharp kernel.

KEY POINTS

• Deep learning (DL)-based noise reduction enables calcium scoring at 1-mm, sharp kernel reconstructed low-dose chest CT (LDCT). • Both iterative reconstruction and DL-based noise reduction underestimated calcium score, but agreement were excellent with those from calcium CT. • Accuracy of categorical classification of calcium scoring tended to be highest in 1-mm LDCT with DL compared to 1-mm LDCT with IR and 3-mm LDCT (90%, 87%, and 84.7%, p = 0.10).

Measurement Accuracy and Repeatability of RECIST-Defined Pulmonary Lesions and Lymph Nodes in Ultra-Low-Dose CT Based on Deep Learning Image Reconstruction

BACKGROUND

Deep learning image reconstruction (DLIR) improves image quality. We aimed to compare the measured diameter of pulmonary lesions and lymph nodes between DLIR-based ultra-low-dose CT (ULDCT) and contrast-enhanced CT.

METHODS

The consecutive adult patients with noncontrast chest ULDCT (0.07-0.14 mSv) and contrast-enhanced CT (2.38 mSv) were prospectively enrolled. Patients with poor image quality and body mass index ≥ 30 kg/m2 were excluded. The diameter of pulmonary target lesions and lymph nodes defined by Response Evaluation Criteria in Solid Tumors (RECIST) was measured. The measurement variability between ULDCT and enhanced CT was evaluated by Bland-Altman analysis.

RESULTS

The 141 enrolled patients (62 ± 12 years) had 89 RECIST-defined measurable pulmonary target lesions (including 30 malignant lesions, mainly adenocarcinomas) and 45 measurable mediastinal lymph nodes (12 malignant). The measurement variation of pulmonary lesions between high-strength DLIR (DLIR-H) images of ULDCT and contrast-enhanced CT was 2.2% (95% CI: 1.7% to 2.6%) and the variation of lymph nodes was 1.4% (1.0% to 1.9%).

CONCLUSIONS

The measured diameters of pulmonary lesions and lymph nodes in DLIR-H images of ULDCT are highly close to those of contrast-enhanced CT. DLIR-based ULDCT may facilitate evaluating target lesions with greatly reduced radiation exposure in tumor evaluation and lung cancer screening.

A Review of Deep Learning CT Reconstruction: Concepts, Limitations, and Promise in Clinical Practice

Purpose of Review

Deep Learning reconstruction (DLR) is the current state-of-the-art method for CT image formation. Comparisons to existing filter back-projection, iterative, and model-based reconstructions are now available in the literature. This review summarizes the prior reconstruction methods, introduces DLR, and then reviews recent findings from DLR from a physics and clinical perspective.

Recent Findings

DLR has been shown to allow for noise magnitude reductions relative to filtered back-projection without suffering from “plastic” or “blotchy” noise texture that was found objectionable with most iterative and model-based solutions. Clinically, early reader studies have reported increases in subjective quality scores and studies have successfully implemented DLR-enabled dose reductions.

Summary

The future of CT image reconstruction is bright; deep learning methods have only started to tackle problems in this space via addressing noise reduction. Artifact mitigation and spectral applications likely be future candidates for DLR applications.

On the Simulation of Ultra-Sparse-View and Ultra-Low-Dose Computed Tomography with Maximum a Posteriori Reconstruction Using a Progressive Flow-Based Deep Generative Model

Ultra-sparse-view computed tomography (CT) algorithms can reduce radiation exposure for patients, but these algorithms lack an explicit cycle consistency loss minimization and an explicit log-likelihood maximization in testing. Here, we propose X2CT-FLOW for the maximum a posteriori (MAP) reconstruction of a three-dimensional (3D) chest CT image from a single or a few two-dimensional (2D) projection images using a progressive flow-based deep generative model, especially for ultra-low-dose protocols. The MAP reconstruction can simultaneously optimize the cycle consistency loss and the log-likelihood. We applied X2CT-FLOW for the reconstruction of 3D chest CT images from biplanar projection images without noise contamination (assuming a standard-dose protocol) and with strong noise contamination (assuming an ultra-low-dose protocol). We simulated an ultra-low-dose protocol. With the standard-dose protocol, our images reconstructed from 2D projected images and 3D ground-truth CT images showed good agreement in terms of structural similarity (SSIM, 0.7675 on average), peak signal-to-noise ratio (PSNR, 25.89 dB on average), mean absolute error (MAE, 0.02364 on average), and normalized root mean square error (NRMSE, 0.05731 on average). Moreover, with the ultra-low-dose protocol, our images reconstructed from 2D projected images and the 3D ground-truth CT images also showed good agreement in terms of SSIM (0.7008 on average), PSNR (23.58 dB on average), MAE (0.02991 on average), and NRMSE (0.07349 on average).

Fully automated image quality evaluation on patient CT: Multi-vendor and multi-reconstruction study

While the recent advancements of computed tomography (CT) technology have contributed in reducing radiation dose and image noise, an objective evaluation of image quality in patient scans has not yet been established. In this study, we present a patient-specific CT image quality evaluation method that includes fully automated measurements of noise level, structure sharpness, and alteration of structure. This study used the CT images of 120 patients from four different CT scanners reconstructed with three types of algorithm: filtered back projection (FBP), vendor-specific iterative reconstruction (IR), and a vendor-agnostic deep learning model (DLM, ClariCT.AI, ClariPi Inc.). The structure coherence feature (SCF) was used to divide an image into the homogeneous (RH) and structure edge (RS) regions, which in turn were used to localize the regions of interests (ROIs) for subsequent analysis of image quality indices. The noise level was calculated by averaging the standard deviations from five randomly selected ROIs on RH, and the mean SCFs on RS was used to estimate the structure sharpness. The structure alteration was defined by the standard deviation ratio between RS and RH on the subtraction image between FBP and IR or DLM, in which lower structure alterations indicate successful noise reduction without degradation of structure details. The estimated structure sharpness showed a high correlation of 0.793 with manually measured edge slopes. Compared to FBP, IR and DLM showed 34.38% and 51.30% noise reduction, 2.87% and 0.59% lower structure sharpness, and 2.20% and -12.03% structure alteration, respectively, on an average. DLM showed statistically superior performance to IR in all three image quality metrics. This study is expected to contribute to enhance the CT protocol optimization process by allowing a high throughput and quantitative image quality evaluation during the introduction or adjustment of lower-dose CT protocol into routine practice.

Artificial intelligence in perinatal diagnosis and management of congenital heart disease

Prenatal diagnosis and management of congenital heart disease (CHD) has progressed substantially in the past few decades. Fetal echocardiography can accurately detect and diagnose approximately 85% of cardiac anomalies. The prenatal diagnosis of CHD results in improved care, with improved risk stratification, perioperative status and survival. However, there is much work to be done. A minority of CHD is actually identified prenatally. This seemingly incongruous gap is due, in part, to diminished recognition of an anomaly even when present in the images and the need for increased training to obtain specialized cardiac views. Artificial intelligence (AI) is a field within computer science that focuses on the development of algorithms that "learn, reason, and self-correct" in a human-like fashion. When applied to fetal echocardiography, AI has the potential to improve image acquisition, image optimization, automated measurements, identification of outliers, classification of diagnoses, and prediction of outcomes. Adoption of AI in the field has been thus far limited by a paucity of data, limited resources to implement new technologies, and legal and ethical concerns. Despite these barriers, recognition of the potential benefits will push us to a future in which AI will become a routine part of clinical practice.

Deep Learning Algorithm for Simultaneous Noise Reduction and Edge Sharpening in Low-Dose CT Images: A Pilot Study Using Lumbar Spine CT

Objective

The purpose of this study was to assess whether a deep learning (DL) algorithm could enable simultaneous noise reduction and edge sharpening in low-dose lumbar spine CT.

Materials and Methods

This retrospective study included 52 patients (26 male and 26 female; median age, 60.5 years) who had undergone CT-guided lumbar bone biopsy between October 2015 and April 2020. Initial 100-mAs survey images and 50-mAs intraprocedural images were reconstructed by filtered back projection. Denoising was performed using a vendor-agnostic DL model (ClariCT.AI™, ClariPI) for the 50-mAS images, and the 50-mAs, denoised 50-mAs, and 100-mAs CT images were compared. Noise, signal-to-noise ratio (SNR), and edge rise distance (ERD) for image sharpness were measured. The data were summarized as the mean ± standard deviation for these parameters. Two musculoskeletal radiologists assessed the visibility of the normal anatomical structures.

Results

Noise was lower in the denoised 50-mAs images (36.38 ± 7.03 Hounsfield unit [HU]) than the 50-mAs (93.33 ± 25.36 HU) and 100-mAs (63.33 ± 16.09 HU) images (p < 0.001). The SNRs for the images in descending order were as follows: denoised 50-mAs (1.46 ± 0.54), 100-mAs (0.99 ± 0.34), and 50-mAs (0.58 ± 0.18) images (p < 0.001). The denoised 50-mAs images had better edge sharpness than the 100-mAs images at the vertebral body (ERD; 0.94 ± 0.2 mm vs. 1.05 ± 0.24 mm, p = 0.036) and the psoas (ERD; 0.42 ± 0.09 mm vs. 0.50 ± 0.12 mm, p = 0.002). The denoised 50-mAs images significantly improved the visualization of the normal anatomical structures (p < 0.001).

Conclusion

DL-based reconstruction may enable simultaneous noise reduction and improvement in image quality with the preservation of edge sharpness on low-dose lumbar spine CT. Investigations on further radiation dose reduction and the clinical applicability of this technique are warranted.

Dose reduction potential of vendor-agnostic deep learning model in comparison with deep learning-based image reconstruction algorithm on CT: a phantom study

Objectives

To compare the dose reduction potential (DRP) of a vendor-agnostic deep learning model (DLM, ClariCT.AI) with that of a vendor-specific deep learning–based image reconstruction algorithm (DLR, TrueFidelity™).

Methods

Computed tomography (CT) images of a multi-sized image quality phantom (Mercury v4.0) were acquired under six radiation dose levels (0.48/0.97/1.93/3.87/7.74/15.47 mGy) and were reconstructed using filtered back projection (FBP) and three strength levels of the DLR (low/medium/high). The FBP images were denoised using the DLM. For all DLM and DLR images, the detectability index (d′) (a task-based detection performance metric) was obtained, under various combinations of three target sizes (10/5/1 mm), five inlets (CT value difference with the background; −895/50/90/335/1000 HU), five phantom diameters (36/31/26/21/16 cm), and six radiation dose levels. Dose reduction potential (DRP) measures the dose reduction made by using DLM or DLR, while yielding d′ equivalent to that of FBP at full dose.

Results

The DRPs of the DLM, DLR-low, DLR-medium, and DLR-high were 86% (81–88%), 60% (46–67%), 76% (60–81%), and 87% (78–92%), respectively. For 10-mm targets, the DRP of the DLM (87%) was higher than that of all DLR algorithms (58–86%). However, for smaller targets (5 mm/1 mm), the DRPs of the DLR-high (89/88%) were greater than those of the DLM (87/84%).

Conclusion

The dose reduction potential of the vendor-agnostic DLM was shown to be comparable to that of the vendor-specific DLR at high strength and superior to those of the DLRs at medium and low strengths.

Key Points

• DRP of the vendor-agnostic model was comparable to that of high-strength vendor-specific model and superior to those of medium- and low-strength models.

• Under various radiation dose levels, the deep learning model shows higher detectability indexes compared to FBP.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-08199-9.