Comparing two artificial intelligence software packages for normative brain volumetry in memory clinic imaging

Purpose To compare two artificial intelligence software packages performing normative brain volumetry and explore whether they could differently impact dementia diagnostics in a clinical context. Methods Sixty patients (20 Alzheimer’s disease, 20 frontotemporal dementia, 20 mild cognitive impairment) and 20 controls were included retrospectively. One MRI per subject was processed by software packages from two proprietary manufacturers, producing two quantitative reports per subject. Two neuroradiologists assigned forced-choice diagnoses using only the normative volumetry data in these reports. They classified the volumetric profile as “normal,” or “abnormal”, and if “abnormal,” they specified the most likely dementia subtype. Differences between the packages’ clinical impact were assessed by comparing (1) agreement between diagnoses based on software output; (2) diagnostic accuracy, sensitivity, and specificity; and (3) diagnostic confidence. Quantitative outputs were also compared to provide context to any diagnostic differences. Results Diagnostic agreement between packages was moderate, for distinguishing normal and abnormal volumetry (K = .41–.43) and for specific diagnoses (K = .36–.38). However, each package yielded high inter-observer agreement when distinguishing normal and abnormal profiles (K = .73–.82). Accuracy, sensitivity, and specificity were not different between packages. Diagnostic confidence was different between packages for one rater. Whole brain intracranial volume output differed between software packages (10.73%, p < .001), and normative regional data interpreted for diagnosis correlated weakly to moderately (rs = .12–.80). Conclusion Different artificial intelligence software packages for quantitative normative assessment of brain MRI can produce distinct effects at the level of clinical interpretation. Clinics should not assume that different packages are interchangeable, thus recommending internal evaluation of packages before adoption.

Efficient non-contrast enhanced 3D Cartesian cardiovascular magnetic resonance angiography of the thoracic aorta in 3 min

Background The application of cardiovascular magnetic resonance angiography (CMRA) for the assessment of thoracic aortic disease is often associated with prolonged and unpredictable acquisition times and residual motion artefacts. To overcome these limitations, we have integrated undersampled acquisition with image-based navigators and inline non-rigid motion correction to enable a free-breathing, contrast-free Cartesian CMRA framework for the visualization of the thoracic aorta in a short and predictable scan of 3 min. Methods 35 patients with thoracic aortic disease (36 ± 13y, 14 female) were prospectively enrolled in this single-center study. The proposed 3D T2-prepared balanced steady state free precession (bSSFP) sequence with image-based navigator (iNAV) was compared to the clinical 3D T2-prepared bSSFP with diaphragmatic-navigator gating (dNAV), in terms of image acquisition time. Three cardiologists blinded to iNAV vs. dNAV acquisition, recorded image quality scores across four aortic segments and their overall diagnostic confidence. Contrast ratio (CR) and relative standard deviation (RSD) of signal intensity (SI) in the corresponding segments were estimated. Co-axial aortic dimensions in six landmarks were measured by two readers to evaluate the agreement between the two methods, along with inter-observer and intra-observer agreement. Kolmogorov–Smirnov test, Mann–Whitney U (MWU), Bland–Altman analysis (BAA), intraclass correlation coefficient (ICC) were used for statistical analysis. Results The scan time for the iNAV-based approach was significantly shorter (3.1 ± 0.5 min vs. 12.0 ± 3.0 min for dNAV, P = 0.005). Reconstruction was performed inline in 3.0 ± 0.3 min. Diagnostic confidence was similar for the proposed iNAV versus dNAV for all three reviewers (Reviewer 1: 3.9 ± 0.3 vs. 3.8 ± 0.4, P = 0.7; Reviewer 2: 4.0 ± 0.2 vs. 3.9 ± 0.3, P = 0.4; Reviewer 3: 3.8 ± 0.4 vs. 3.7 ± 0.6, P = 0.3). The proposed method yielded higher image quality scores in terms of artefacts from respiratory motion, and non-diagnostic images due to signal inhomogeneity were observed less frequently. While the dNAV approach outperformed the iNAV method in the CR assessment, the iNAV sequence showed improved signal homogeneity along the entire thoracic aorta [RSD SI 5.1 (4.4, 6.5) vs. 6.5 (4.6, 8.6), P = 0.002]. BAA showed a mean difference of < 0.05 cm across the 6 landmarks between the two datasets. ICC showed excellent inter- and intra-observer reproducibility. Conclusions Thoracic aortic iNAV-based CMRA with fast acquisition (~ 3 min) and inline reconstruction (3 min) is proposed, resulting in high diagnostic confidence and reproducible aortic measurements.

Addition of CT to Improve the Diagnostic Confidence for the Detection of Sacroiliac Joint Erosions in Patients with Equivocal MRI Findings

Purpose To determine if CT can improve the diagnostic confidence for the detection of sacroiliac joint (SIJ) erosions in patients with equivocal MRI findings. Methods A retrospective analysis of adult patients who had an SIJ MRI and a subsequent SIJ CT within 12 months was conducted. Using a 5-point Likert scale, two reviewers evaluated the de-identified MRI and CT images in randomized order and in separate sessions to answer the question: “Does the patient have SIJ erosions?”. A Fisher’s exact test was used to analyze the difference in diagnostic confidence, and intraclass correlation coefficient (ICC) was used to determine interrater reliability. Results 54 patients were included in the analysis (average age, 43.9 years). The average time interval between initial SIJ MRI and subsequent CT was 14.4 weeks (range, 5.6–50.3 weeks). CT resulted in significantly more cases with definitive diagnostic confidence than cases with probable or equivocal confidence compared to MRI ( P < .001). Amongst cases with equivocal findings on MRI, 73.2% of cases had definitive diagnoses on CT. There was moderate interrater agreement for MRI, with an ICC of .490 [95% CI, .258–.669], and excellent agreement for CT, with an ICC of .832 [95% CI, .728–.899]. Conclusion Overall, CT led to significantly increased diagnostic confidence and higher interrater reliability for the detection of SIJ erosions compared to MRI. Judicious use of CT may be useful in detecting SIJ erosions in patients with equivocal MRI findings.

Variability in endoscopic assessment of Nissen fundoplication wrap integrity and hiatus herniation

Summary Background Upper endoscopy (EGD) is frequently performed in patients with esophageal complaints following anti-reflux surgery such as fundoplication. Endoscopic evaluation of fundoplication wrap integrity can be challenging. Our primary aim in this pilot study was to evaluate the accuracy and confidence of assessing Nissen fundoplication integrity and hiatus herniation among gastroenterology (GI) fellows, subspecialists, and foregut surgeons. Methods Five variations of post-Nissen fundoplication anatomy were included in a survey of 20 sets of EGD images that was completed by GI fellows, general GI attendings, esophagologists, and foregut surgeons. Accuracy, diagnostic confidence, and inter-rater agreement across providers were evaluated. Results There were 31 respondents in the final cohort. Confidence in pre-survey diagnostics significantly differed by provider type (mean confidence out of 5 was 1.8 for GI fellows, 2.7 for general GI attendings, 3.6 for esophagologists, and 3.6 for foregut surgeons, P = 0.01). The mean overall accuracy was 45.9%, which significantly differed by provider type with the lowest rate among GI fellows (37%) and highest among esophagologists (53%; P = 0.01). The accuracy was highest among esophagologists across all wrap integrity variations. Inter-rater agreement was low across wrap integrity variations (Krippendorf’s alpha &lt;0.30), indicating low to no agreement between providers. Conclusion In this multi-center survey study, GI fellows had the lowest accuracy and confidence in assessing EGD images after Nissen fundoplication, whereas esophagologists had the highest. Diagnostic confidence varied considerably and inter-rater agreement was poor. These findings suggest experience may improve confidence, but highlight the need to improve the evaluation of fundoplication wraps.

Clinical Evaluation of an Abbreviated Contrast-Enhanced Whole-Body MRI for Oncologic Follow-Up Imaging

Over the last decades, overall survival for most cancer types has increased due to earlier diagnosis and more effective treatments. Simultaneously, whole-body MRI-(WB-MRI) has gained importance as a radiation free staging alternative to computed tomography. The aim of this study was to evaluate the diagnostic confidence and reproducibility of a novel abbreviated 20-min WB-MRI for oncologic follow-up imaging in patients with melanoma. In total, 24 patients with melanoma were retrospectively included in this institutional review board-approved study. All patients underwent three consecutive staging examinations via WB-MRI in a clinical 3 T MR scanner with an abbreviated 20-min protocol. Three radiologists independently evaluated the images in a blinded, random order regarding image quality (overall image quality, organ-based image quality, sharpness, noise, and artifacts) and regarding its diagnostic confidence on a 5-point-Likert-Scale (5 = excellent). Inter-reader agreement and reproducibility were assessed. Overall image quality and diagnostic confidence were rated to be excellent (median 5, interquartile range [IQR] 5–5). The sharpness of anatomic structures, and the extent of noise and artifacts, as well as the assessment of lymph nodes, liver, bone, and the cutaneous system were rated to be excellent (median 5, IQR 4–5). The image quality of the lung was rated to be good (median 4, IQR 4–5). Therefore, our study demonstrated that the novel accelerated 20-min WB-MRI protocol is feasible, providing high image quality and diagnostic confidence with reliable reproducibility for oncologic follow-up imaging.

The effect of deep learning reconstruction on abdominal CT densitometry and image quality: a systematic review and meta-analysis

Objective To determine the difference in CT values and image quality of abdominal CT images reconstructed by filtered back-projection (FBP), hybrid iterative reconstruction (IR), and deep learning reconstruction (DLR). Methods PubMed and Embase were systematically searched for articles regarding CT densitometry in the abdomen and the image reconstruction techniques FBP, hybrid IR, and DLR. Mean differences in CT values between reconstruction techniques were analyzed. A comparison between signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of FBP, hybrid IR, and DLR was made. A comparison of diagnostic confidence between hybrid IR and DLR was made. Results Sixteen articles were included, six being suitable for meta-analysis. In the liver, the mean difference between hybrid IR and DLR was − 0.633 HU (p = 0.483, SD ± 0.902 HU). In the spleen, the mean difference between hybrid IR and DLR was − 0.099 HU (p = 0.925, SD ± 1.061 HU). In the pancreas, the mean difference between hybrid IR and DLR was − 1.372 HU (p = 0.353, SD ± 1.476 HU). In 14 articles, CNR was described. In all cases, DLR showed a significantly higher CNR. In 9 articles, SNR was described. In all cases but one, DLR showed a significantly higher SNR. In all cases, DLR showed a significantly higher diagnostic confidence. Conclusions There were no significant differences in CT values reconstructed by FBP, hybrid IR, and DLR in abdominal organs. This shows that these reconstruction techniques are consistent in reconstructing CT values. DLR images showed a significantly higher SNR and CNR, compared to FBP and hybrid IR. Key Points CT values of abdominal CT images are similar between deep learning reconstruction (DLR), filtered back-projection (FBP), and hybrid iterative reconstruction (IR). DLR results in improved image quality in terms of SNR and CNR compared to FBP and hybrid IR images. DLR can thus be safely implemented in the clinical setting resulting in improved image quality without affecting CT values.

Reduction of metal artifacts from knee tumor prostheses on CT images: value of the single energy metal artifact reduction (SEMAR) algorithm

Background To evaluate the effect of the single energy metal artifact reduction (SEMAR) algorithm with a multidetector CT (MDCT) for knee tumor prostheses. Methods First, a phantom of knee tumor prosthesis underwent a MDCT scan. The raw data was reconstructed by iterative reconstruction (IR) alone and IR plus SEMAR. The mean value of the CT number and the image noise were measured around the prosthesis at the stem level and articular level. Second, 95 consecutive patients with knee tumor prostheses underwent MDCT scans. The raw data were also reconstructed by the two methods. Periprosthetic structures were selected at the similar two levels. Four radiologists visually graded the image quality on a scale from 0 to 5. Additionally, the readers also assessed the presence of prosthetic complication and tumor recurrence on a same scale. Results In the phantom, when the SEMAR was used, the CT numbers were closer to normal value and the noise of images using soft and sharper kernel were respectively reduced by up to 77.1% and 43.4% at the stem level, and by up to 82.2% and 64.5% at the articular level. The subjective scores increased 1 ~ 3 points and 1 ~ 4 points at the two levels, respectively. Prosthetic complications and tumor recurrence were diagnosed in 66 patients. And the SEMAR increased the diagnostic confidence of prosthetic complications and tumor recurrence (4 ~ 5 vs. 1 ~ 1.5). Conclusions The SEMAR algorithm can significantly reduce the metal artifacts and increase diagnostic confidence of prosthetic complications and tumor recurrence in patients with knee tumor prostheses.

Application of deep learning image reconstruction algorithm to improve image quality in CT angiography of children with Takayasu arteritis

BACKGROUND: The inflammatory indexes of children with Takayasu arteritis (TAK) usually tend to be normal immediately after treatment, therefore, CT angiography (CTA) has become an important method to evaluate the status of TAK and sometime is even more sensitive than laboratory test results. OBJECTIVE: To evaluate image quality improvement in CTA of children diagnosed with TAK using a deep learning image reconstruction (DLIR) in comparison to other image reconstruction algorithms. METHODS: hirty-two TAK patients (9.14±4.51 years old) underwent neck, chest and abdominal CTA using 100 kVp were enrolled. Images were reconstructed at 0.625 mm slice thickness using Filtered Back-Projection (FBP), 50%adaptive statistical iterative reconstruction-V (ASIR-V), 100%ASIR-V and DLIR with high setting (DLIR-H). CT number and standard deviation (SD) of the descending aorta and back muscle were measured and contrast-to-noise ratio (CNR) for aorta was calculated. The vessel visualization, overall image noise and diagnostic confidence were evaluated using a 5-point scale (5, excellent; 3, acceptable) by 2 observers. RESULTS: There was no significant difference in CT number across images reconstructed using different algorithms. Image noise values (in HU) were 31.36±6.01, 24.96±4.69, 18.46±3.91 and 15.58±3.65, and CNR values for aorta were 11.93±2.12, 15.66±2.37, 22.54±3.34 and 24.02±4.55 using FBP, 50%ASIR-V, 100%ASIR-V and DLIR-H, respectively. The 100%ASIR-V and DLIR-H images had similar noise and CNR (all P > 0.05), and both had lower noise and higher CNR than FBP and 50%ASIR-V images (all P < 0.05). The subjective evaluation suggested that all images were diagnostic for large arteries, however, only 50%ASIR-V and DLIR-H met the diagnostic requirement for small arteries (3.03±0.18 and 3.53±0.51). CONCLUSION: DLIR-H improves CTA image quality and diagnostic confidence for TAK patients compared with 50%ASIR-V, and best balances image noise and spatial resolution compared with 100%ASIR-V.

The Pattern of Use, Effectiveness, and Safety of Gadoteric Acid (Clariscan) in Patients Undergoing Contrast-Enhanced Magnetic Resonance Imaging: A Prospective, Multicenter, Observational Study

Objective. Contrast-enhanced MR (CE-MR) imaging is often required to improve lesion detection and characterization and to increase diagnostic confidence. This study aimed to evaluate the safety and effectiveness, as well as the use pattern, of the macrocyclic gadolinium-based contrast agent Clariscan in real-world clinical practice in Korea. Materials and Methods. This was a prospective, multicenter, observational study of patients undergoing CE-MR as part of routine clinical care at 6 university hospitals in Korea. Effectiveness was evaluated by determining diagnostic confidence and image quality; safety evaluation included the adverse event (AE) expression rate. Subgroup analyses were conducted by body regions of diagnosis (musculoskeletal, nervous system, others) and in pediatric patients (aged ≤7 years). Results. From October 2019 to September 2020, 1,376 subjects were included in the study. The mean volume of Clariscan used was 0.26 mL/kg (0.13 mmol/kg). In the overall study population and in each subgroup, diagnostic confidence increased after contrast enhancement with Clariscan. Overall, image quality was excellent in 72.5% of subjects and good-to-adequate in 27.2%. Clariscan was well tolerated (14 AEs occurred in 10 subjects); all AEs were of mild severity. Subgroup analyses showed that the mean dose of Clariscan used was ≥0.1 mmol/kg for nervous system-related diagnoses (e.g., brain) and ≤0.1 mmol/kg for musculoskeletal and pediatric-related diagnoses. All musculoskeletal and pediatric examinations were provided with a smaller package of 5 mL Clariscan. By body region of MR examination, the most common region was the nervous system in 69.0%, musculoskeletal system in 13.6%, and reproductive system in 4.9%. Conclusions. This study confirmed the use pattern of Clariscan and its excellent effectiveness and safety in the real-world clinical environment in Korea. The small-dose package indicated the possibility of increasing the convenience and efficiency of drug use.

Comparison of 68Ga-PSMA PET/CT and Multiparametric MRI for the Detection of Low- and Intermediate-risk Prostate Cancer

Purpose: To provide a direct comparison between mpMRI and 68Ga-PSMA PET/CT for the detection of low- and intermediate-risk PCa and to determine which of low- or intermediate-risk PCa is more likely to be detected by 68Ga-PSMA PET/CT.Methods: We conducted a retrospective analysis of patients who had undergone a prostate biopsy and/or radical prostatectomy, and who were scanned with 68Ga-PSMA PET/CT and mpMRI between June 2019 and March 2021. The mpMRI images were scored with the Prostate Imaging-Reporting and Data System Version 2.1 (PI-RADS) and were classified as either negative (PI-RADS 1-3) or positive (PI-RADS 4-5). Suspicious 68Ga-PSMA PET/CT lesions were reviewed for each relevant patient and classified by double-trained board-certified nuclear medicine physicians. The results were evaluated with the histopathological outcome. All patients were classified according to the D’Amico classification, and the clinical data were combined for stratified analysis.Result: 101 patients who were pathologically diagnosed with PCa were analyzed. Of the 101 patients, 88 (80.6%) patients presented with a pathologic mpMRI, and 85 (79.1%) with a pathologic 68Ga-PSMA PET/CT. In the high-risk PCa cohort, 68Ga-PSMA PET/CT was positive in 64/66 (97.0%) patients and yielded a higher detection rate than that for the mpMRI patients (58/66, 87.9%: p<0.05). However, mpMRI provided superior diagnostic confidence in identifying low- and intermediate-risk PCa (30/35, 85.7% vs. 21/35,60.0%; p<0.05). When the age threshold exceeded 62.5 years and the serum prostate specific antigen (PSA) threshold exceeded 9.4 ng/ml, a higher uptake of PSMA was more likely to occur in the lesions of low- and intermediate-risk PCa.Conclusion: In the initial diagnosis of low- and intermediate-risk PCa, we found that mpMRI showed a higher diagnostic accuracy than 68Ga-PSMA PET/CT did. Low- and intermediate-risk PCa patients with PSA≥9.4 ng/ml and age≥62.5 years were more likely to have a positive 68Ga-PSMA PET/CT result.