Salary Equity in Academic Radiology Relative to Other Clinical Specialties.

BACKGROUND: Equity in faculty compensation in U.S. academic radiology physicians relative to other specialties is not well known. OBJECTIVE: The aim of this study is to assess salary equity in U.S. academic radiology physicians at different ranks relative to other clinical specialties. METHODS: The American Association of Medical Colleges (AAMC) Faculty Salary Survey was used to collect information for full-time faculty at U.S. medical schools. Financial compensation data were collected for 2023 for faculty with MD or equivalent degree in medical specialties, stratified by gender and rank. RESULTS: The AAMC Faculty Salary Survey data for 2023 included responses for 97,224 faculty members in clinical specialties, with 5847 faculty members in Radiology departments. In radiology, compared to men (n = 3839), the women faculty members (n = 1763) had a lower median faculty compensation by 6% at the rank of Assistant Professor, 3% for Associate Professors, 4% for Professors and 6% for Section Chief positions. Surgery had the highest difference in median compensation with 21%, 24%, 22% and 19% lower faculty compensation, respectively, for women faculty members at corresponding ranks. Pathology had the lowest percent difference (<1%) in median compensation for all professor ranks. Salary inequity in radiology was lower compared to most other specialties. From assistant to full professors, all other clinical specialties except Pathology and Psychiatry, had a greater salary inequity than Radiology. CONCLUSION: The salary inequity in academic radiology faculty is lower than most other specialties. Further efforts should be made to reduce salary inequities as broader efforts to provide a more diverse, equitable and inclusive environment. SUMMARY STATEMENT: Salary inequity in academic radiology faculty is lower than most other specialties.

Predictors for large vessel recanalization before stroke thrombectomy: the HALT score.

BACKGROUND: Large vessel recanalization (LVR) before endovascular therapy (EVT) for acute large vessel ischemic strokes is a poorly understood phenomenon. Better understanding of predictors for LVR is important for optimizing stroke triage and patient selection for bridging thrombolysis. METHODS: In this retrospective cohort study, consecutive patients presenting to a comprehensive stroke center for EVT treatment were identified from 2018 to 2022. Demographic information, clinical characteristics, intravenous thrombolysis (IVT) use, and LVR before EVT were recorded. Factors independently associated with different rates of LVR were identified, and a prediction model for LVR was constructed. RESULTS: 640 patients were identified. 57 (8.9%) patients had LVR before EVT. A minority (36.4%) of LVR patients had significant improvements in National Institutes of Health Stroke Scale. Independent predictors for LVR were identified and used to construct the 8-point HALT score: hyperlipidemia (1 point), atrial fibrillation (1 point), location of vascular occlusion (internal carotid: 0 points, M1: 1 point, M2: 2 points, vertebral/basilar: 3 points), and thrombolysis at least 1.5 hours before angiography (3 points). The HALT score had an area under the receiver-operating curve (AUC) of 0.85 (95% CI 0.81 to 0.90, P<0.001) for predicting LVR. LVR before EVT occurred in only 1 of 302 patients (0.3%) with low (0-2) HALT scores. CONCLUSIONS: IVT at least 1.5 hours before angiography, site of vascular occlusion, atrial fibrillation, and hyperlipidemia are independent predictors for LVR. The 8-point HALT score proposed in this study may be a valuable tool for predicting LVR before EVT.

Analysis of Medical Malpractice Claims Involving Interventional Radiologists: A Comprehensive Analysis From Two National Legal Databases.

PURPOSE: The nature of lawsuits involving interventional radiologists (IRs) is not well understood. The purposes of this article are to provide an overview of the causes of action underlying medical malpractice lawsuits related to IRs and to characterize the associated factors and outcomes. METHODS: Two large legal databases were used to search for US legal cases in which there were jury awards and settlements involving IRs in the United States. Cases were screened to include only those cases in which the cause of action involved negligence on the part of IRs. RESULTS: A total of 389 published case summaries were identified, of which 93 were eligible to be included in the analysis. In 46% of the cases (43 of 93), medical malpractice was alleged against an individual IR, whereas in 43% (40 of 93), it was alleged against both an individual IR and a health care institution. Thirty-five percent of IR malpractice cases (33 of 93) involved the performance of a vascular procedure, most commonly embolization procedures (30% [n = 10]), stenting or angioplasty (21% [n = 7]), and diagnostic arteriography and angiography (18% [n = 6]). Twenty-six percent of cases (24 of 93) involved IR performance of a biopsy. Eighteen percent of cases (17 of 93) involved a failure to gain informed consent in addition to an allegation of medical negligence during treatment. Eleven percent of cases (10 of 93) were resolved by settlement, with an average settlement amount of $877,500 (range, $200,000-$2,700,000). Among the 72 cases that went to trial, 74% (53 of 72) resulted in judgments for the defendants, and 26% (19 of 72) resulted in judgements for the plaintiffs, with an average award of $2,012,243 (range, $101,667-$6,400,000). CONCLUSIONS: Vascular procedures and biopsies were the most frequent reasons for malpractice lawsuits involving IRs. Failure to gain informed consent in addition to an allegation of medical negligence during treatment was not infrequent. Although the majority of published medical malpractice claims involving IRs resulted in judgments in favor of the defendants, the average amount awarded to plaintiffs was higher compared with previous data reported for all physicians.

Clinical implications of Peri-hematomal edema microperfusion fraction in intracerebral hemorrhage intravoxel incoherent motion imaging - A pilot study.

BACKGROUND AND PURPOSE: Perihematomal edema (PHE) represents the secondary brain injury after intracerebral hemorrhage (ICH). However, neurobiological characteristics of post-ICH parenchymal injury other than PHE volume have not been fully characterized. Using intravoxel incoherent motion imaging (IVIM), we explored the clinical correlates of PHE diffusion and (micro)perfusion metrics in subacute ICH. MATERIALS AND METHODS: In 41 consecutive patients scanned 1-to-7 days after supratentorial ICH, we determined the mean diffusion (D), pseudo-diffusion (D*), and perfusion fraction (F) within manually segmented PHE. Using univariable and multivariable statistics, we evaluated the relationship of these IVIM metrics with 3-month outcome based on the modified Rankin Scale (mRS). RESULTS: In our cohort, the average (± standard deviation) age of patients was 68.6±15.6 years, median (interquartile) baseline National Institute of Health Stroke Scale (NIHSS) was 7 (3-13), 11 (27 %) patients had poor outcomes (mRS>3), and 4 (10 %) deceased during the follow-up period. In univariable analyses, admission NIHSS (p < 0.001), ICH volume (p = 0.019), ICH+PHE volume (p = 0.016), and average F of the PHE (p = 0.005) had significant correlation with 3-month mRS. In multivariable model, the admission NIHSS (p = 0.006) and average F perfusion fraction of the PHE (p = 0.003) were predictors of 3-month mRS. CONCLUSION: The IVIM perfusion fraction (F) maps represent the blood flow within microvasculature. Our pilot study shows that higher PHE microperfusion in subacute ICH is associated with worse outcomes. Once validated in larger cohorts, IVIM metrics may provide insight into neurobiology of post-ICH secondary brain injury and identify at-risk patients who may benefit from neuroprotective therapy.

Utility of Gadolinium-Based Contrast in Initial Evaluation of Seizures in Children Presenting Emergently.

BACKGROUND AND PURPOSE: The frequency and utility of gadolinium in evaluation of acute pediatric seizure presentation is not well known. The purpose of this study was to assess the utility of gadolinium-based contrast agents in MR imaging performed for the evaluation of acute pediatric seizure presentation. MATERIALS AND METHODS: We identified consecutive pediatric patients with new-onset seizures from October 1, 2016, to September 30, 2021, who presented to the emergency department and/or were admitted to the inpatient unit and had an MR imaging of the brain for the evaluation of seizures. The clinical and imaging data were recorded, including the patient's age and sex, the use of IV gadolinium, and the underlying cause of epilepsy when available. RESULTS: A total of 1884 patients were identified for inclusion. Five hundred twenty-four (28%) patients had potential epileptogenic findings on brain MR imaging, while 1153 (61%) patients had studies with normal findings and 207 (11%) patients had nonspecific signal changes. Epileptogenic findings were subclassified as the following: neurodevelopmental lesions, 142 (27%); intracranial hemorrhage (traumatic or germinal matrix), 89 (17%); ischemic/hypoxic, 62 (12%); hippocampal sclerosis, 44 (8%); neoplastic, 38 (7%); immune/infectious, 20 (4%); phakomatoses, 19 (4%); vascular anomalies, 17 (3%); metabolic, 3 (<1%); and other, 90 (17%). Eight hundred seventy-four (46%) patients received IV gadolinium. Of those, only 48 (5%) cases were retrospectively deemed to have necessitated the use of IV gadolinium: Fifteen of 48 (31%) cases were subclassified as immune/infectious, while 33 (69%) were neoplastic. Of the 1010 patients with an initial noncontrast study, 15 (1.5%) required repeat MR imaging with IV contrast to further evaluate the findings. CONCLUSIONS: Gadolinium contrast is of limited additive benefit in the imaging of patients with an acute onset of pediatric seizures in most instances.

Dural arteriovenous fistula onyx embolization through a non-visible previously embolized pedicle.

Cerebral dural arteriovenous fistulas (dAVFs) are commonly treated with endovascular embolization, either through arterial or venous routes in single or multi-staged procedures. Treatment outcomes depend on multiple factors, including the patient's clinical status, the anatomy of the malformation, and the operator's experience, but technical success is also highly dependent on choice of neurovascular devices and embolic agents. When transvenous routes are not feasible and the most obvious transarterial routes do not appear safe, deep knowledge of the dynamic fluid properties of liquid embolics can be a valuable asset. We present a case(video 1) in which a complex skull-base dAVF was completely occluded through a non-visualized previously partially embolized arterial feeder. It is a unique case in which the embolization takes advantage of the peculiar fluid dynamic properties of non-adhesive embolic agents (Onyx-Medtronic, Minnesota, USA) 18 and 34.1-3neurintsurg;jnis-2023-020506v1/V1F1V1Video 1 Technical video presenting a dAVF completely cured through a non-visible previously embolized arterial pedicle.

Distribution and Disparities of Industry Payments to Radiologists (2016-2020).

BACKGROUND: The frequency, magnitude, and distribution of industry payments to radiologists are not well understood. RATIONALE AND OBJECTIVES: The aim of this study was to analyze the distribution of industry payments to physicians working in diagnostic radiology, interventional radiology, and radiation oncology, study the categories of payments and determine their correlation. MATERIALS AND METHODS: The Open Payments Database from the Centers for Medicare & Medicaid Services was accessed and analyzed for the period from January 1, 2016 to December 31, 2020. Payments were grouped into six categories: consulting fees, education, gifts, research, speaker fees, and royalties/ownership. The total amount and types of industry payments going to the top 5% group were determined overall and for each category of payment. RESULTS: From 2016 to 2020, a total of 513 020 payments, amounting to $370 782 608, were made to 28 739 radiologists suggesting that approximately 70% of the 41 000 radiologists in the US received at least one industry payment during the 5-year period. The median payment value was $27 (IQR: $15-$120) and the median number of payments per physician over the 5-year period was 4 (IQR: 1-13). Gifts were the most frequent payment type made (76.4%), but accounted for only 4.8% of payment value. The median total value of payments earned by members of the top 5% group over the 5-year period was $58 878 (IQR: $29 686-$162 425) ($11 776 per year) compared to $172 (IQR: $49-877) ($34 per year) in the bottom 95% group. Members of the top 5% group received a median of 67 (IQR: 26-147) individual payments (13 payments per year) while members of the bottom 95% group received a median of 3 (IQR: 1-11) (0.6 payments per year). CONCLUSION: Between 2016 and 2020, industry payments to radiologists were highly concentrated both in terms of number/frequency and value of payments.

PET/CT May Assist in Avoiding Pointless Thyroidectomy in Indeterminate Thyroid Nodules: A Narrative Review.

Indeterminate thyroid nodules (ITN) are commonly encountered among the general population, with a malignancy rate of 10 to 40%. However, many patients may be overtreated with futile surgery for benign ITN. To avoid unnecessary surgery, PET/CT scan is a possible alternative to help differentiate between benign and malignant ITN. In this narrative review, the major results and limitations of the most recent studies on PET/CT efficacy (from PET/CT visual assessment to quantitative PET parameters and recent radiomic features analysis) and on cost-effectiveness (compared to other alternatives (such as surgery)) are presented. PET/CT can reduce futile surgery with visual assessment (around 40%; if ITN ≥ 10 mm). Moreover, PET/CT conventional parameters and radiomic features extracted from PET/CT imaging can be associated together in a predictive model to rule out malignancy in ITN, with a high NPV (96%) when certain criteria are met. Even though promising results were obtained in these recent PET/CT studies, further studies are needed to enable PET/CT to become the definitive diagnostic tool once a thyroid nodule is identified as indeterminate.

National Trends of Cerebral Venous Sinus Stenting for the Treatment of Idiopathic Intracranial Hypertension.

OBJECTIVES: Cerebral venous sinus stenting (VSS) has emerged as a new surgical procedure for the treatment of severe idiopathic intracranial hypertension (IIH), and its popularity has been anecdotally on the rise. This study explores recent temporal trends of VSS and other surgical IIH treatments in the United States. METHODS: Adult patients with IIH were identified from the 2016-2020 National Inpatient Sample databases, and surgical procedures and hospital characteristics were recorded. Temporal trends of procedure numbers for VSS, CSF shunts, and optic nerve sheath fenestrations (ONSFs) were assessed and compared. RESULTS: A total of 46,065 (95% CI 44,710-47,420) patients with IIH were identified, of whom 7,535 patients (95% CI 6,982-8,088) received surgical IIH treatments. VSS procedures increased 80% (150 [95% CI 55-245] to 270 [95% CI 162-378] per year, p < 0.001). Concurrently, the number of CSF shunts decreased by 19% (1,365 [95% CI 1,126-1,604] to 1,105 [95% CI 900-1,310] per year, p < 0.001), and ONSF procedures decreased by 54% (65 [95% CI 20-110] to 30 [95% CI 6-54] per year, p < 0.001). DISCUSSION: Practice patterns for surgical IIH treatment in the United States are rapidly evolving, and VSS is becoming increasingly common. These findings highlight the urgency of randomized controlled trials to investigate the comparative effectiveness and safety of VSS, CSF shunts, ONSF, and standard medical treatments.

Follow-Up Infarct Volume Prediction by CTP-Based Hypoperfusion Index, and the Discrepancy between Small Follow-Up Infarct Volume and Poor Functional Outcome-A Multicenter Study.

(1) Background: Follow-up infarct volume (FIV) may have implications for prognostication in acute ischemic stroke patients. Factors predicting the discrepancy between FIV and 90-day outcomes are poorly understood. We aimed to develop a comprehensive predictive model of FIV and explore factors associated with the discrepancy. (2) Methods: Patients with acute anterior circulation large vessel occlusion were included. Baseline clinical and CT features were extracted and analyzed, including the CTP-based hypoperfusion index (HI) and the NCCT-based e-ASPECT, measured by automated software. FIV was assessed on follow-up NCCT at 3−7 days. Multiple linear regression was used to construct the predictive model. Subgroup analysis was performed to explore factors associated with poor outcomes (90-mRS scores 3−6) in small FIV (<70 mL). (3) Results: There were 170 patients included. Baseline e-ASPECT, infarct core volume, hypoperfusion volume, HI, baseline international normalized ratio, and successful recanalization were associated with FIV and included in constructing the predictive model. Baseline NIHSS, baseline hypertension, stroke history, and current tobacco use were associated with poor outcomes in small FIV. (4) Conclusions: A comprehensive predictive model (including HI) of FIV was constructed. We also emphasized the importance of hypertension and smoking status at baseline for the functional outcomes in patients with a small FIV.

Excessive platelet inhibition following Pipeline embolization of intracranial aneurysms.

BACKGROUND: High levels of platelet inhibition have been associated with hemorrhagic complications following Pipeline embolization of intracranial aneurysms. We therefore titrate clopidogrel dosing to maintain a moderate level of platelet inhibition using the VerifyNow P2Y12 assay. However, many patients demonstrate dramatic increases in platelet inhibition following treatment despite being on a consistent antiplatelet regimen. We therefore elected to explore the incidence of this phenomenon and possible predisposing factors. METHODS: All successful Pipeline aneurysm treatments performed at our institution from 2011 to 2019 with moderate procedure-day platelet inhibition levels as indicated by a VerifyNow PRU of 60-235 were included. Patients who received glycoprotein IIb/IIIa inhibitors and those treated for ruptured/symptomatic lesions were excluded. The incidence of excessive platelet inhibition defined by a PRU<60 within 8 weeks of treatment was noted. Multivariable logistic regression was performed to determined independent predictors of the phenomenon. RESULTS: Some 190 treatments were performed in 178 qualifying patients. A post-procedure PRU <60 occurred following 79% of treatments, documented on average after 8.5 (range 1-47) days. A higher procedure day hematocrit level (P=0.003, OR 1.09, 95% CI 1.029 to 1.152) was an independent predictor of reaching a PRU <60, while intra-procedural midazolam exposure (P=0.044, OR 0.44, 95% CI 0.201 to 0.980) and a higher procedure-day PRU (P=0.047, OR 0.99, 95% CI 0.982 to 1.000) were associated with a reduced odds. Time-since-procedure and hematocrit levels were associated with excessive platelet inhibition when excluding patients who initially demonstrated hyperresponse. CONCLUSION: Elevations in platelet inhibition were frequently observed following flow diversion with Pipeline.

Higher Hospital Frailty Risk Score is associated with increased complications and healthcare resource utilization after endovascular treatment of ruptured intracranial aneurysms.

AIM: To use the Hospital Frailty Risk Score (HFRS) to investigate the impact of frailty on complication rates and healthcare resource utilization in patients who underwent endovascular treatment of ruptured intracranial aneurysms (IAs). METHODS: A retrospective cohort study was performed using the 2016-2019 National Inpatient Sample database. All adult patients (≥18 years) undergoing endovascular treatment for IAs after subarachnoid hemorrhage were identified using ICD-10-CM codes. Patients were categorized into frailty cohorts: low (HFRS <5), intermediate (HFRS 5-15) and high (HFRS >15). Patient demographics, adverse events, length of stay (LOS), discharge disposition, and total cost of admission were assessed. Multivariate logistic regression analysis was used to identify independent predictors of prolonged LOS, increased cost, and non-routine discharge. RESULTS: Of the 33 840 patients identified, 7940 (23.5%) were found to be low, 20 075 (59.3%) intermediate and 5825 (17.2%) high frailty by HFRS criteria. The rate of encountering any adverse event was significantly greater in the higher frailty cohorts (low: 59.9%; intermediate: 92.4%; high: 99.2%, p<0.001). There was a stepwise increase in mean LOS (low: 11.7±8.2 days; intermediate: 18.7±14.1 days; high: 26.6±20.1 days, p<0.001), mean total hospital cost (low: $62 888±37 757; intermediate: $99 670±63 446; high: $134 937±80 331, p<0.001), and non-routine discharge (low: 17.3%; intermediate: 44.4%; high: 69.4%, p<0.001) with increasing frailty. On multivariate regression analysis, a similar stepwise impact was found in prolonged LOS (intermediate: OR 2.38, p<0.001; high: OR 4.49, p<0.001)], total hospital cost (intermediate: OR 2.15, p<0.001; high: OR 3.62, p<0.001), and non-routine discharge (intermediate: OR 2.13, p<0.001; high: OR 4.17, p<0.001). CONCLUSIONS: Our study found that greater frailty as defined by the HFRS was associated with increased complications, LOS, total costs, and non-routine discharge.

Cost-effectiveness of endovascular thrombectomy in acute stroke patients with large ischemic core.

BACKGROUND: Evidence has shown that endovascular thrombectomy (EVT) treatment improves clinical outcomes. Yet, its benefit remains uncertain in patients with large established infarcts as defined by ASPECTS (Alberta Stroke Program Early CT Score) <6. This study evaluates the cost-effectiveness of EVT, compared with standard care (SC), in acute ischemic stroke (AIS) patients with ASPECTS 3-5. METHODS: An economic evaluation study was performed combining a decision tree and Markov model to estimate lifetime costs (2021 US$) and quality-adjusted life years (QALYs) of AIS patients with ASPECTS 3-5. Incremental cost-effectiveness ratios (ICERs), net monetary benefits (NMBs), and deterministic one-way and two-way sensitivity analyses were performed. Probabilistic sensitivity analyses were also performed to evaluate the robustness of our model. RESULTS: Compared with SC, the cost-effectiveness analyses revealed that EVT yields higher lifetime benefits (2.20 QALYs vs 1.41 QALYs) with higher lifetime healthcare cost per patient ($285 861 vs $272 954). The difference in health benefits between EVT and SC was 0.79 QALYs, equivalent to 288 additional days of healthy life per patient. Even though EVT is more costly than SC alone, it is still cost-effective given better outcomes with ICER of $16 239/QALY. The probabilistic sensitivity analyses indicated that EVT was the most cost-effective strategy in 98.8% (9882 of 10 000) of iterations at the willingness-to-pay threshold of $100 000 per QALY. CONCLUSIONS: The results of this study suggest that EVT is cost-effective in AIS patients with a large ischemic core (ASPECTS 3-5), compared with SC alone over the patient's lifetime.

Hospital Frailty Risk Score Predicts Adverse Events and Readmission Following a Ventriculoperitoneal Shunt Surgery for Normal Pressure Hydrocephalus.

OBJECTIVE: The aim of this study was to evaluate the impact of a Hospital Frailty Risk Score (HFRS) on unplanned readmission and health care resource utilization in normal pressure hydrocephalus (NPH) patients undergoing a ventriculoperitoneal (VP) shunt surgery. METHODS: A retrospective cohort study was performed using the 2016-2019 Nationwide Readmission Database. All NPH patients (≥60 years) undergoing a VP shunt surgery were identified using ICD-10-CM diagnostic and procedural codes. Patients were dichotomized into 2 cohorts as follows: Low HFRS (<5) and Intermediate-High HFRS (≥5). A multivariate logistic regression analysis was then used to identify independent predictors of adverse event (AE) and 30- and 90-day readmission. RESULTS: Of 13,262 patients, 4386 (33.1%) had an Intermediate-High HFRS score. A greater proportion of the Intermediate-High HFRS cohort experienced at least one AE (1.9 vs. 22.1, P < 0.001). The Intermediate-High HFRS cohort also had a longer length of stay (2.3 ± 2.4 days vs. 7.0 ± 7.7 days, P < 0.001), higher non-routine discharge rate (19.9% vs. 39.9%, P < 0.001), and greater admission cost ($14,634 ± 5703 vs. $21,749 ± 15,234, P < 0.001). The Intermediate-High HFRS cohort had higher rates of 30- (7.6% vs. 11.0%, P < 0.001) and 90-day (6.8% vs. 8.3%, P < 0.001) readmissions. On a multivariate regression analysis, Intermediate-High HFRS compared to Low HFRS was an independent predictor of any AE (odds ratio, 16.6; 95% confidence interval, [12.9-21.5]; P < 0.001) and 30-day readmission (odds ratio, 1.4; 95% confidence interval, [1.2-1.7]; P < 0.001). CONCLUSIONS: Our study suggests that frailty, as defined by HFRS, is associated with increased resource utilization in NPH patients undergoing VP shunt surgery. Furthermore, HFRS was an independent predictor of adverse events and 30-day hospital readmission.

Cost-effectiveness of thrombectomy in patients with minor stroke and large vessel occlusion: effect of thrombus location on cost-effectiveness and outcomes.

BACKGROUND: To evaluate the cost-effectiveness of endovascular thrombectomy (EVT) to treat large vessel occlusion (LVO) in patients with acute, minor stroke (National Institute of Health Stroke Scale (NIHSS) <6) and impact of occlusion site. METHODS: A Markov decision-analytic model was constructed accounting for both costs and outcomes from a societal perspective. Two different management strategies were evaluated: EVT and medical management. Base case analysis was done for three different sites of occlusion: proximal M1, distal M1 and M2 occlusions. One-way, two-way and probabilistic sensitivity analyses were performed. RESULTS: Base-case calculation showed EVT to be the dominant strategy in 65-year-old patients with proximal M1 occlusion and NIHSS <6, with lower cost (US$37 229 per patient) and higher effectiveness (1.47 quality-adjusted life years (QALYs)), equivalent to 537 days in perfect health or 603 days in modified Rankin score (mRS) 0-2 health state. EVT is the cost-effective strategy in 92.7% of iterations for patients with proximal M1 occlusion using a willingness-to-pay threshold of US$100 000/QALY. EVT was cost-effective if it had better outcomes in 2%-3% more patients than intravenous thrombolysis (IVT) in absolute numbers (base case difference -16%). EVT was cost-effective when the proportion of M2 occlusions was less than 37.1%. CONCLUSIONS: EVT is cost-effective in patients with minor stroke and LVO in the long term (lifetime horizon), considering the poor outcomes and significant disability associated with non-reperfusion. Our study emphasizes the need for caution in interpreting previous observational studies which concluded similar results in EVT versus medical management in patients with minor stroke due to a high proportion of patients with M2 occlusions in the two strategies.

Clinical implementation of artificial intelligence in neuroradiology with development of a novel workflow-efficient picture archiving and communication system-based automated brain tumor segmentation and radiomic feature extraction.

Purpose: Personalized interpretation of medical images is critical for optimum patient care, but current tools available to physicians to perform quantitative analysis of patient's medical images in real time are significantly limited. In this work, we describe a novel platform within PACS for volumetric analysis of images and thus development of large expert annotated datasets in parallel with radiologist performing the reading that are critically needed for development of clinically meaningful AI algorithms. Specifically, we implemented a deep learning-based algorithm for automated brain tumor segmentation and radiomics extraction, and embedded it into PACS to accelerate a supervised, end-to- end workflow for image annotation and radiomic feature extraction. Materials and methods: An algorithm was trained to segment whole primary brain tumors on FLAIR images from multi-institutional glioma BraTS 2021 dataset. Algorithm was validated using internal dataset from Yale New Haven Health (YHHH) and compared (by Dice similarity coefficient [DSC]) to radiologist manual segmentation. A UNETR deep-learning was embedded into Visage 7 (Visage Imaging, Inc., San Diego, CA, United States) diagnostic workstation. The automatically segmented brain tumor was pliable for manual modification. PyRadiomics (Harvard Medical School, Boston, MA) was natively embedded into Visage 7 for feature extraction from the brain tumor segmentations. Results: UNETR brain tumor segmentation took on average 4 s and the median DSC was 86%, which is similar to published literature but lower than the RSNA ASNR MICCAI BRATS challenge 2021. Finally, extraction of 106 radiomic features within PACS took on average 5.8 ± 0.01 s. The extracted radiomic features did not vary over time of extraction or whether they were extracted within PACS or outside of PACS. The ability to perform segmentation and feature extraction before radiologist opens the study was made available in the workflow. Opening the study in PACS, allows the radiologists to verify the segmentation and thus annotate the study. Conclusion: Integration of image processing algorithms for tumor auto-segmentation and feature extraction into PACS allows curation of large datasets of annotated medical images and can accelerate translation of research into development of personalized medicine applications in the clinic. The ability to use familiar clinical tools to revise the AI segmentations and natively embedding the segmentation and radiomic feature extraction tools on the diagnostic workstation accelerates the process to generate ground-truth data.

Identifying clinically applicable machine learning algorithms for glioma segmentation: recent advances and discoveries.

Background: While there are innumerable machine learning (ML) research algorithms used for segmentation of gliomas, there is yet to be a US FDA cleared product. The aim of this study is to explore the systemic limitations of research algorithms that have prevented translation from concept to product by a review of the current research literature. Methods: We performed a systematic literature review on 4 databases. Of 11 727 articles, 58 articles met the inclusion criteria and were used for data extraction and screening using TRIPOD. Results: We found that while many articles were published on ML-based glioma segmentation and report high accuracy results, there were substantial limitations in the methods and results portions of the papers that result in difficulty reproducing the methods and translation into clinical practice. Conclusions: In addition, we identified that more than a third of the articles used the same publicly available BRaTS and TCIA datasets and are responsible for the majority of patient data on which ML algorithms were trained, which leads to limited generalizability and potential for overfitting and bias.

Contrast-associated acute kidney injury in acute ischemic stroke patients following multi-dose iodinated contrast.

BACKGROUND AND OBJECTIVE: lthough intravenous contrast in neuroimaging has become increasingly important in selecting patients for stroke treatment, clinical concerns remain regarding contrast-associated acute kidney injury (CA-AKI). Given the increasing utilization of CT angiography and/or perfusion coupled with cerebral angiography, the purpose of this study was to assess the association of CA-AKI and multi-dose iodinated contrast in acute ischemic stroke (AIS) patients. MATERIALS AND METHODS: etrospective review of AIS patients at a comprehensive stroke center was performed from January 2018 to December 2019. Data collection included patient demographics, stroke risk factors, stroke severity, discharge disposition, modified Rankin Scale, contrast type/volume, and creatinine levels (baseline, 48-72 h). CA-AKI was defined as creatinine increase ≥ 25 % from baseline. Bivariate analyses and multivariable logistic regression models were implemented to compare AIS patients with multi-dose and single-dose contrast. RESULTS: Of 440 AIS patients, 215 (48.9 %) were exposed to a single-dose contrast, and 225 (51.1 %) received multi-dose. In single-dose patients, CA-AKI at 48/72 h was 9.7 %/10.2 % compared to 8.0 %/8.9 % in multi-dose patients. Multi-dose patients were significantly more likely to receive a higher volume of contrast (mean 142.1 mL versus 80.8 mL; p < 0.001), but there was no significant difference in their creatinine levels or CA-AKI. NIHSS score (OR=1.08, 95 % CI=[1.04,1.13]), and patient transfer from another hospital (OR=3.84, 95 % CI=[1.94,7.62]) were significantly associated with multi-dose contrast. CONCLUSIONS: No significant association between multi-dose iodinated contrast and CA-AKI was seen in AIS patients. Concerns of CA-AKI should not deter physicians from pursuing timely and appropriate contrast-enhanced neuroimaging that may optimize treatment outcomes in AIS patients.

Machine Learning Models for Classifying High- and Low-Grade Gliomas: A Systematic Review and Quality of Reporting Analysis.

Objectives: To systematically review, assess the reporting quality of, and discuss improvement opportunities for studies describing machine learning (ML) models for glioma grade prediction. Methods: This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy (PRISMA-DTA) statement. A systematic search was performed in September 2020, and repeated in January 2021, on four databases: Embase, Medline, CENTRAL, and Web of Science Core Collection. Publications were screened in Covidence, and reporting quality was measured against the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) Statement. Descriptive statistics were calculated using GraphPad Prism 9. Results: The search identified 11,727 candidate articles with 1,135 articles undergoing full text review and 85 included in analysis. 67 (79%) articles were published between 2018-2021. The mean prediction accuracy of the best performing model in each study was 0.89 ± 0.09. The most common algorithm for conventional machine learning studies was Support Vector Machine (mean accuracy: 0.90 ± 0.07) and for deep learning studies was Convolutional Neural Network (mean accuracy: 0.91 ± 0.10). Only one study used both a large training dataset (n>200) and external validation (accuracy: 0.72) for their model. The mean adherence rate to TRIPOD was 44.5% ± 11.1%, with poor reporting adherence for model performance (0%), abstracts (0%), and titles (0%). Conclusions: The application of ML to glioma grade prediction has grown substantially, with ML model studies reporting high predictive accuracies but lacking essential metrics and characteristics for assessing model performance. Several domains, including generalizability and reproducibility, warrant further attention to enable translation into clinical practice. Systematic Review Registration: PROSPERO, identifier CRD42020209938.