Artificial intelligence in neuroradiology: a scoping review of some ethical challenges.

Artificial intelligence (AI) has great potential to increase accuracy and efficiency in many aspects of neuroradiology. It provides substantial opportunities for insights into brain pathophysiology, developing models to determine treatment decisions, and improving current prognostication as well as diagnostic algorithms. Concurrently, the autonomous use of AI models introduces ethical challenges regarding the scope of informed consent, risks associated with data privacy and protection, potential database biases, as well as responsibility and liability that might potentially arise. In this manuscript, we will first provide a brief overview of AI methods used in neuroradiology and segue into key methodological and ethical challenges. Specifically, we discuss the ethical principles affected by AI approaches to human neuroscience and provisions that might be imposed in this domain to ensure that the benefits of AI frameworks remain in alignment with ethics in research and healthcare in the future.

Evaluating the Impact of Peer Review on the Completeness of Reporting in Imaging Diagnostic Test Accuracy Research.

BACKGROUND: Despite the nearly ubiquitous reported use of peer review among reputable medical journals, there is limited evidence to support the use of peer review to improve the quality of biomedical research and in particular, imaging diagnostic test accuracy (DTA) research. PURPOSE: To evaluate whether peer review of DTA studies published by imaging journals is associated with changes in completeness of reporting, transparency for risk of bias assessment, and spin. STUDY TYPE: Retrospective cross-sectional study. STUDY SAMPLE: Cross-sectional study of articles published in Journal of Magnetic Resonance Imaging (JMRI), Canadian Association of Radiologists Journal (CARJ), and European Radiology (EuRad) before March 31, 2020. ASSESSMENT: Initial submitted and final versions of manuscripts were evaluated for completeness of reporting using the Standards for Reporting Diagnostic Accuracy Studies (STARD) 2015 and STARD for Abstracts guidelines, transparency of reporting for risk of bias assessment based on Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2), and actual and potential spin using modified published criteria. STATISTICAL TESTS: Two-tailed paired t-tests and paired Wilcoxon signed-rank tests were used for comparisons. A P value <0.05 was considered to be statistically significant. RESULTS: We included 84 diagnostic accuracy studies accepted by three journals between 2014 and 2020 (JMRI = 30, CARJ = 23, and EuRad = 31) of the 692 which were screened. Completeness of reporting according to STARD 2015 increased significantly between initial submissions and final accepted versions (average reported items: 16.67 vs. 17.47, change of 0.80 [95% confidence interval 0.25-1.17]). No significant difference was found for the reporting of STARD for Abstracts (5.28 vs. 5.25, change of -0.03 [-0.15 to 0.11], P = 0.74), QUADAS-2 (6.08 vs. 6.11, change of 0.03 [-1.00 to 0.50], P = 0.92), actual "spin" (2.36 vs. 2.40, change of 0.04 [0.00 to 1.00], P = 0.39) or potential "spin" (2.93 vs. 2.81, change of -0.12 [-1.00 to 0.00], P = 0.23) practices. CONCLUSION: Peer review is associated with a marginal improvement in completeness of reporting in published imaging DTA studies, but not with improvement in transparency for risk of bias assessment or reduction in spin. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 1.

Association of Accuracy, Conclusions, and Reporting Completeness With Acceptance by Radiology Conferences and Journals.

BACKGROUND: Preferential publication of studies with positive findings can lead to overestimation of diagnostic test accuracy (i.e. publication bias). Understanding the contribution of the editorial process to publication bias could inform interventions to optimize the evidence guiding clinical decisions. PURPOSE/HYPOTHESIS: To evaluate whether accuracy estimates, abstract conclusion positivity, and completeness of abstract reporting are associated with acceptance to radiology conferences and journals. STUDY TYPE: Meta-research. POPULATION: Abstracts submitted to radiology conferences (European Society of Gastrointestinal and Abdominal Radiology (ESGAR) and International Society for Magnetic Resonance in Medicine (ISMRM)) from 2008 to 2018 and manuscripts submitted to radiology journals (Radiology, Journal of Magnetic Resonance Imaging [JMRI]) from 2017 to 2018. Primary clinical studies evaluating sensitivity and specificity of a diagnostic imaging test in humans with available editorial decisions were included. ASSESSMENT: Primary variables (Youden's index [YI > 0.8 vs. <0.8], abstract conclusion positivity [positive vs. neutral/negative], number of reported items on the Standards for Reporting of Diagnostic Accuracy Studies [STARD] for Abstract guideline) and confounding variables (prospective vs. retrospective/unreported, sample size, study duration, interobserver agreement assessment, subspecialty, modality) were extracted. STATISTICAL TESTS: Multivariable logistic regression to obtain adjusted odds ratio (OR) as a measure of the association between the primary variables and acceptance by radiology conferences and journals; 95% confidence intervals (CIs) and P-values were obtained; the threshold for statistical significance was P < 0.05. RESULTS: A total of 1000 conference abstracts (500 ESGAR and 500 ISMRM) and 1000 journal manuscripts (505 Radiology and 495 JMRI) were included. Conference abstract acceptance was not significantly associated with YI (adjusted OR = 0.97 for YI > 0.8; CI = 0.70-1.35), conclusion positivity (OR = 1.21 for positive conclusions; CI = 0.75-1.90) or STARD for Abstracts adherence (OR = 0.96 per unit increase in reported items; CI = 0.82-1.18). Manuscripts with positive abstract conclusions were less likely to be accepted by radiology journals (OR = 0.45; CI = 0.24-0.86), while YI (OR = 0.85; CI = 0.56-1.29) and STARD for Abstracts adherence (OR = 1.06; CI = 0.87-1.30) showed no significant association. Positive conclusions were present in 86.7% of submitted conference abstracts and 90.2% of journal manuscripts. DATA CONCLUSION: Diagnostic test accuracy studies with positive findings were not preferentially accepted by the evaluated radiology conferences or journals. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 2.

Quantitative analysis of repaired rabbit supraspinatus tendons (± channeling) using magnetic resonance imaging at 7 Tesla.

BACKGROUND: The quantitative assessment of supraspinatus tendons by conventional magnetic resonance is limited by low contrast-to-noise ratio (CNR). Magnetic resonance imaging (MRI) scanners operating at 7 Tesla offer high signal-to noise ratio (SNR), low CNR and high spatial resolution that are well-suited for rapidly relaxing tissues like tendons. Few studies have applied T2 and T2* mapping to musculoskeletal imaging and to the rotator cuff tendons. Our objective was to analyze the T2 and T2* relaxation times from surgically repaired supraspinatus tendons and the effect of bone channeling. METHODS: One supraspinatus tendon of 112 adult female New Zealand white rabbits was surgically detached and repaired one week later. Rabbits were randomly assigned to channeling (n=64) or control (n=48) groups and harvested at 0, 1, 2, and 4 weeks. A 7T magnet was used for signal acquisition. For T2 mapping, a sagittal multi slice 2D multi-echo spin-echo (MESE) CPMG sequence with fat saturation was applied and T2* mapping was performed using a 3D UTE sequence. Magnetic resonance images from supraspinatus tendons were analyzed by two raters. Three regions of interest were manually drawn on the first T2-weighted dataset. For T2 and T2*, different ROI masks were generated to obtain relaxation times. RESULTS: T2-weighted maps but not T2*-weighted maps generated reliable signals for relaxation time measurement. Torn supraspinatus tendons had lower T2 than controls at the time of repair (20.0±3.4 vs. 25.6±3.9 ms; P<0.05). T2 increased at 1, 2 and 4 postoperative weeks: 22.7±3.1, 23.3±3.9 and 24.0±5.1 ms, respectively, and values were significantly different from contralateral supraspinatus tendons (24.8±3.1; 26.8±4.3 and 26.5±3.6 ms; all P<0.05). Bone channeling did not affect T2 (P>0.05). CONCLUSIONS: Supraspinatus tendons detached for 1 week had shorter T2 relaxation time compared to contralateral as measured with 7T MRI.

Impact of Education-based HEART Score Pathway on Coronary Computed Tomography Angiography Utilization and Yield in the Emergency Department.

OBJECTIVE: There is a growing consensus to reduce unnecessary testing among low-risk chest pain patients. The objective of this study was to evaluate the impact of implementing an education-based HEART score pathway in the emergency department on coronary computed tomography angiography (CCTA) utilization and yield. METHODS: A retrospective before and after intervention study was conducted at a single site. Adult emergency department patients undergoing CCTA for suspected acute coronary syndrome were included. Primary outcomes were CCTA utilization and yield. Utilization was defined as the percentage of patients evaluated with CCTA and yield was calculated as the percentage of patients with a diagnosis of obstructive coronary artery disease, defined as ≥50% stenosis in any one coronary artery due to atherosclerosis. RESULTS: 1540 patients undergoing CCTAs were included. CCTA utilization before and after were 2.2% [95% confidence interval (CI) 2.0-2.3] and 2.0% (95% CI 1.9-2.2), respectively; mean difference 0.1% (95% CI -0.1 to 0.3; P = 0.21). The mean age was 53 years (SD = 11) and females were 52%. Of 1477 patients included in CCTA yield analysis, patients diagnosed with obstructive coronary artery disease before and after were 15.0% (95% CI 12.6-17.7) and 16.2% (95% CI 13.6-19.1), respectively; mean difference 1.2% (95% CI -2.6 to 5.1; P = 0.53). CONCLUSIONS: There was no significant change in the CCTA utilization or yield after the implementation of an education-based HEART pathway in a large academic center. Our findings suggest adopting a more comprehensive approach for deploying such evidence-based protocols to increase institutional compliance.

Abnormal blood-brain barrier water exchange in chronic multiple sclerosis lesions: A preliminary study.

Relapsing-remitting multiple sclerosis (RRMS) is associated with persistent blood-brain barrier (BBB) dysfunction. The impact of this persistent dysfunction in both active and chronic MS lesions has yet to be investigated due to technological challenges associated with invasive assessment of BBB water transportation (e.g. 15O-PET). The purpose of this study was to test if persistent BBB dysfunction in RRMS manifests as lower BBB water exchange in chronic lesions using a recently developed noninvasive MRI paradigm. Patients with relapsing-remitting MS and healthy subjects were recruited for this prospective study. The novel Intrinsic Diffusivity Encoding of Arterial Labeled Spins (IDEALS) MRI method was used to map BBB water extraction fraction (Ew) and water permeability surface area product (PSw), as well as cerebral blood flow (CBF). Regional differences in BBB water exchange were evaluated between MS patients (normal appearing white matter [NAWM] and normal appearing gray matter [NAGM]) and healthy subjects (white matter [WM] and gray matter [GM]) and within MS subjects in non gadolinium-based contrast-agent (GBCA) enhancing chronic lesions, perilesional areas, and NAWM. Significantly lower PSw and Ew were observed in NAWM compared to WM (ΔPSw: -11.9 mL/100 g/min, p < .05; ΔEw: -4.3%, p < .01). Significantly lower Ew was observed in NAGM compared to GM (ΔEw: -12.1%, p < .01). Significantly lower PSw and CBF were observed in non-GBCA contrast enhancing lesions compared to NAWM (ΔPSw = -11.5 mL/100 g/min, p < .05; ΔCBF = -8.1 mL/100 g/min, p < .05). Ew was significantly higher in non-GBCA enhancing chronic MS lesions compared to NAWM (ΔEw = 1.6%, p < .05). The lower BBB water exchange in chronic MS lesions is consistent with previously reported observations and may demonstrate metabolic changes associated with MS.

A Task-dependent Investigation on Dose and Texture in CT Image Reconstruction.

Localizing and characterizing clinically-significant lung nodules, a potential precursor to lung cancer, at the lowest achievable radiation dose is demanded to minimize the stochastic radiation effects from x-ray computed tomography (CT). A minimal dose level is heavily dependent on the image reconstruction algorithms and clinical task, in which the tissue texture always plays an important role. This study aims to investigate the dependence through a task-based evaluation at multiple dose levels and variable textures in reconstructions with prospective patient studies. 133 patients with a suspicious pulmonary nodule scheduled for biopsy were recruited and the data was acquired at120kVp with three different dose levels of 100, 40 and 20mAs. Three reconstruction algorithms were implemented: analytical filtered back-projection (FBP) with optimal noise filtering; statistical Markov random field (MRF) model with optimal Huber weighting (MRF-H) for piecewise smooth reconstruction; and tissue-specific texture model (MRF-T) for texture preserved statistical reconstruction. Experienced thoracic radiologists reviewed and scored all images at random, blind to the CT dose and reconstruction algorithms. The radiologists identified the nodules in each image including the 133 biopsy target nodules and 66 other non-target nodules. For target nodule characterization, only MRF-T at 40mAs showed no statistically significant difference from FBP at 100mAs. For localizing both the target nodules and the non-target nodules, some as small as 3mm, MRF-T at 40 and 20mAs levels showed no statistically significant difference from FBP at 100mAs, respectively. MRF-H and FBP at 40 and 20mAs levels performed statistically differently from FBP at 100mAs. This investigation concluded that (1) the textures in the MRF-T reconstructions improves both the tasks of localizing and characterizing nodules at low dose CT and (2) the task of characterizing nodules is more challenging than the task of localizing nodules and needs more dose or enhanced textures from reconstruction.

In vivo evaluation of human patellar tendon microstructure and microcirculation with diffusion MRI.

BACKGROUND: Patellar tendon (PT) microstructure integrity and microcirculation status play a crucial role in the progression of tendinopathy and tendon repair. PURPOSE: To assess the feasibility and robustness of stimulated-echo based diffusion-weighted MRI with readout-segmented echo-planar imaging (ste-RS-EPI) for noninvasive assessment of microstructure and microcirculation of human PT. STUDY TYPE: Prospective. SUBJECTS: Fifteen healthy volunteers. FIELD STRENGTH/SEQUENCE: PT diffusion tensor imaging (DTI) and intravoxel incoherent motion (IVIM) were acquired with an ste-RS-EPI protocol on a 3T MRI scanner. ASSESSMENT: Subjects were positioned with their PT at the magic angle. DTI-derived parameters including axial diffusivity (AD), radial diffusivity (RD), mean diffusivity (MD), and fractional anisotropy (FA) were estimated with b-values of 0 and 800 s/mm2 and 12 diffusion directions. IVIM-derived parameters, f p , D* × f p , V b , and D* × V b were assessed in the central-third and the outer-two thirds of the PT with b-values of 0, 20, 30, 60, 80, 120, 200, 400, and 600 s/mm2 in three orthogonal directions. STATISTICAL TESTS: Paired t-tests were used to evaluate differences in IVIM parameters between the central-third and outer-two thirds regions of the patellar tendon. Paired t-tests and within-subject coefficient of variation were used to assess the intra- and intersession reproducibility of PT DTI and IVIM parameters. RESULTS: DTI parameters for healthy PT were 1.54 ± 0.09 × 10-3 mm2 /s, 1.01 ± 0.05 × 10-3 mm2 /s, 1.18 ± 0.06 × 10-3 mm2 /s, and 0.30 ± 0.04 for AD, RD, MD, and FA, respectively. Significantly higher (P < 0.05) IVIM parameters f p and D* × f p were observed in the outer-two thirds (6.1% ± 2.4% and 95.2 ± 49.6, respectively) compared with the central-third (3.8% ± 2.3% and 48.6 ± 35.2, respectively) of the PT. DATA CONCLUSION: Diffusion MRI of PT with an ste-RS-EPI protocol is clinically feasible. Both DTI- and IVIM-derived parameters of the PT demonstrated good test-retest reproducibility and interrater reliability. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2020;51:780-790.

Gender Differences in Article Citations by Authors from American Institutions in Major Radiology Journals.

Objective To investigate gender difference patterns in article citations, by first and last authors, in four radiology journals. Materials and methods Articles by authors published in four major radiology journals from 1984, 1994, 2004, and 2014 were categorized into 12 subspecialties. The number of citations, references used, co-authors, and length of the article (number of pages) were documented. The genders of first/last authors were determined. Data were analyzed using chi-square and logistic regression. Results The gender of the first author was determined in 2679 articles and that of the last author in 2717 articles. Over the selected years, 1984 to 2014, female first authorship grew from 13.0% to 31.5% (p<0.001), and female last authorship grew from 9.3% to 22.1% (p<0.001). Primary female authorship papers were cited less often as compared to males (OR 0.9972, 95% CI: 0.9948-0.9996, p=0.021), after adjusting for publication year and subspecialty. Across most subspecialties, female first authorship received fewer citations. In 1984, primary female authorship papers received on average 28.9 citations versus males at 39.1; in 1994, 50.4 versus 60.8; in 2004, 41.5 versus 44.4; and in 2014, 7.0 versus 7.8. The mean difference in the number of citations received by male and female first authors decreased from 10.47±6.09 in 1984 and 9.49±7.12 in 1994 to 1.93±5.63 in 2004 and 0.79±0.39 in 2014. However, there was no statistical difference demonstrated in article citations between male and female last authorship (OR 0.9990, 95% CI: 0.9966-1.0013, p=0.392). Conclusions Primary female authorship garnered fewer citations than men, despite the increasing frequency of authorships. However, this differential in the number of citations is narrowing.

Abbreviated quantitative UTE imaging in anterior cruciate ligament reconstruction.

BACKGROUND: Existing ultrashort echo time magnetic resonance imaging (UTE MRI) methods require prohibitively long acquisition times (~ 20-40 min) to quantitatively assess the clinically relevant fast decay T2* component in ligaments and tendons. The purpose of this study was to evaluate the feasibility and clinical translatability of a novel abbreviated quantitative UTE MRI paradigm for monitoring graft remodeling after anterior cruciate ligament (ACL) reconstruction. METHODS: Eight patients who had Graftlink™ hamstring autograft reconstruction were recruited for this prospective study. A 3D double-echo UTE sequence at 3.0 Tesla was performed at 3- and 6-months post-surgery. An abbreviated UTE MRI paradigm was established based on numerical simulations and in vivo validation from healthy knees. This proposed approach was used to assess the T2* for fast decay component ([Formula: see text]) and bound water signal fraction (fbw) of ACL graft in regions of interest drawn by a radiologist. RESULTS: Compared to the conventional bi-exponential model, the abbreviated UTE MRI paradigm achieved low relative estimation bias for [Formula: see text] and fbw over a range of clinically relevant values for ACL grafts. A decrease in [Formula: see text] of the intra-articular graft was observed in 7 of the 8 ACL reconstruction patients from 3- to 6-months (- 0.11 ± 0.16 ms, P = 0.10). Increases in [Formula: see text] and fbw from 3- to 6-months were observed in the tibial intra-bone graft ([Formula: see text]: 0.19 ± 0.18 ms, P < 0.05; Δfbw: 4% ± 4%, P < 0.05). Lower [Formula: see text] (- 0.09 ± 0.11 ms, P < 0.05) was observed at 3-months when comparing the intra-bone graft to the graft/bone interface in the femoral tunnel. The same comparisons at the 6-months also yielded relatively lower [Formula: see text] (- 0.09 ± 0.12 ms, P < 0.05). CONCLUSION: The proposed abbreviated 3D UTE MRI paradigm is capable of assessing the ACL graft remodeling process in a clinically translatable acquisition time. Longitudinal changes in [Formula: see text] and fbw of the ACL graft were observed.

MRI biomarkers in osseous tumors.

Although radiography continues to play a critical role in osseous tumor assessment, there have been remarkable advances in cross-sectional imaging. MRI has taken a lead in this assessment due to high tissue contrast and spatial resolution, which are well suited for bone lesion assessment. More recently, although somewhat lagging other organ systems, quantitative parameters have shown promising potential as biomarkers for osseous tumors. Among these sequences are chemical shift imaging (CSI), apparent diffusion coefficient (ADC), and intravoxel incoherent motion (IVIM) from diffusion-weighted imaging (DWI), quantitative dynamic contrast enhanced (DCE)-MRI, and magnetic resonance spectroscopy (MRS). In this article, we review the background and recent roles of these quantitative MRI biomarkers for osseous tumors. Level of Evidence: 3 Technical Efficacy Stage: 3 J. MAGN. RESON. IMAGING 2019. J. Magn. Reson. Imaging 2019;50:702-718.

Merging Orthovoltage X-Ray Minibeams spare the proximal tissues while producing a solid beam at the target.

Conventional radiation therapy of brain tumors often produces cognitive deficits, particularly in children. We investigated the potential efficacy of merging Orthovoltage X-ray Minibeams (OXM). It segments the beam into an array of parallel, thin (~0.3 mm), planar beams, called minibeams, which are known from synchrotron x-ray experiments to spare tissues. Furthermore, the slight divergence of the OXM array make the individual minibeams gradually broaden, thus merging with their neighbors at a given tissue depth to produce a solid beam. In this way the proximal tissues, including the cerebral cortex, can be spared. Here we present experimental results with radiochromic films to characterize the method's dosimetry. Furthermore, we present our Monte Carlo simulation results for physical absorbed dose, and a first-order biologic model to predict tissue tolerance. In particular, a 220-kVp orthovoltage beam provides a 5-fold sharper lateral penumbra than a 6-MV x-ray beam. The method can be implemented in arc-scan, which may include volumetric-modulated arc therapy (VMAT). Finally, OXM's low beam energy makes it ideal for tumor-dose enhancement with contrast agents such as iodine or gold nanoparticles, and its low cost, portability, and small room-shielding requirements make it ideal for use in the low-and-middle-income countries.