Incorrect imaging protocol selection can lead to important clinical findings being missed, contributing to both wasted health care resources and patient harm. We present a machine learning method for ...analyzing the unstructured text of clinical indications and patient demographics from magnetic resonance imaging (MRI) orders to automatically protocol MRI procedures at the sequence level. We compared 3 machine learning models - support vector machine, gradient boosting machine, and random forest - to a baseline model that predicted the most common protocol for all observations in our test set. The gradient boosting machine model significantly outperformed the baseline and demonstrated the best performance of the 3 models in terms of accuracy (95%), precision (86%), recall (80%), and Hamming loss (0.0487). This demonstrates the feasibility of automating sequence selection by applying machine learning to MRI orders. Automated sequence selection has important safety, quality, and financial implications and may facilitate improvements in the quality and safety of medical imaging service delivery.
Flow diversion for posterior circulation aneurysms performed using the Pipeline embolization device (PED) constitutes an increasingly common off-label use for otherwise untreatable aneurysms. The ...safety and efficacy of this treatment modality has not been assessed in a multicenter study.
A retrospective review of prospectively maintained databases at 8 academic institutions was performed for the years 2009 to 2016 to identify patients with posterior circulation aneurysms treated with PED placement.
A total of 129 consecutive patients underwent 129 procedures to treat 131 aneurysms; 29 dissecting, 53 fusiform, and 49 saccular lesions were included. At a median follow-up of 11 months, complete and near-complete occlusion was recorded in 78.1%. Dissecting aneurysms had the highest occlusion rate and fusiform the lowest. Major complications were most frequent in fusiform aneurysms, whereas minor complications occurred most commonly in saccular aneurysms. In patients with saccular aneurysms, clopidogrel responders had a lower complication rate than did clopidogrel nonresponders. The majority of dissecting aneurysms were treated in the immediate or acute phase following subarachnoid hemorrhage, a circumstance that contributed to the highest mortality rate in those aneurysms.
In the largest series to date, fusiform aneurysms were found to have the lowest occlusion rate and the highest frequency of major complications. Dissecting aneurysms, frequently treated in the setting of subarachnoid hemorrhage, occluded most often and had a low complication rate. Saccular aneurysms were associated with predominantly minor complications, particularly in clopidogrel nonresponders.
Abstract
BACKGROUND
Flow diversion for basilar apex aneurysms has rarely been reported.
OBJECTIVE
To assess flow diversion for basilar apex aneurysms in a multicenter cohort.
METHODS
Retrospective ...review of prospectively maintained databases at 8 academic institutions was performed from 2009 to 2016 to identify patients with basilar apex aneurysms treated with flow diversion. Clinical and radiographic data were analyzed.
RESULTS
Sixteen consecutive patients (median age 54.5 yr) underwent 18 procedures to treat 16 basilar apex aneurysms with either the Pipeline Embolization Device (Medtronic Inc, Dublin, Ireland) or Flow Redirection Endoluminal Device (Microvention, Tustin, California). Five aneurysms (31.3%) were treated in the setting of subarachnoid hemorrhage. Seven aneurysms (43.8%) were treated with flow diversion alone, while 9 (56.2%) underwent flow diversion and adjunctive coiling. At a median follow-up of 6 mo, complete (100%) and near-complete (90%-99%) occlusion was noted in 11 (68.8%) aneurysms. Incomplete occlusion occurred more commonly in patients treated with flow diversion alone compared to those with adjunctive coiling. Patients with partial occlusion were significantly younger. Retreatment with an additional flow diverter and adjunctive coiling occurred in 2 aneurysms with wide necks. There was 1 mortality in a patient (6.3%) who experienced posterior cerebral artery and cerebellar strokes as well as subarachnoid hemorrhage after the placement of a flow diverter. Minor complications occurred in 2 patients (12.5%).
CONCLUSION
Flow diversion for the treatment of basilar apex aneurysms results in acceptable occlusion rates in highly selected cases. Both primary flow diversion and rescue after failed clipping or coiling resulted in a modified Rankin Scale score that was either equal or better than at presentation and the technology represents a viable alternative or adjunctive option.
Incorrect imaging protocol selection can contribute to increased healthcare cost and waste. To help healthcare providers improve the quality and safety of medical imaging services, we developed and ...evaluated three natural language processing (NLP) models to determine whether NLP techniques could be employed to aid in clinical decision support for protocoling and prioritization of magnetic resonance imaging (MRI) brain examinations.
To test the feasibility of using an NLP model to support clinical decision making for MRI brain examinations, we designed three different medical imaging prediction tasks, each with a unique outcome: selecting an examination protocol, evaluating the need for contrast administration, and determining priority. We created three models for each prediction task, each using a different classification algorithm-random forest, support vector machine, or k-nearest neighbor-to predict outcomes based on the narrative clinical indications and demographic data associated with 13,982 MRI brain examinations performed from January 1, 2013 to June 30, 2015. Test datasets were used to calculate the accuracy, sensitivity and specificity, predictive values, and the area under the curve.
Our optimal results show an accuracy of 82.9%, 83.0%, and 88.2% for the protocol selection, contrast administration, and prioritization tasks, respectively, demonstrating that predictive algorithms can be used to aid in clinical decision support for examination protocoling.
NLP models developed from the narrative clinical information provided by referring clinicians and demographic data are feasible methods to predict the protocol and priority of MRI brain examinations.
OBJECTIVE Treatment of wide-necked intracranial aneurysms is associated with higher recanalization and complication rates; however, the most commonly used methods are not specifically designed to ...work in bifurcation lesions. To address these issues, the authors describe the evolution in the design and use of the eCLIPs (Endovascular Clip System) device, a novel hybrid stent-like assist device with flow diverter properties that was first described in 2008. METHODS A registry was established covering 13 international centers at which patients were treated with the second-generation eCLIPs device. Aneurysm morphology and rupture status, device neck coverage, coil retention, and procedural and late morbidity and mortality were recorded. For those patients who had undergone successful implantation more than 6 months earlier, the final imaging and clinical follow-up results and need for re-treatment were recorded. RESULTS Thirty-three patients were treated between June 2013 and September 2015. Twenty-five (76%) patients had successful placement of an eCLIPs device; 23 (92%) of these 25 patients had complete data. Eight cases of nondeployment occurred during the 1st year of use, consistent with a learning curve; no failures of deployment occurred thereafter. Two periprocedural transient ischemic attacks and 2 asymptomatic thrombotic events occurred. Twenty-one (91%) of 23 patients underwent follow-up at an average of 8 months (range 3-18 months); 9 (42.9%) of these 21 patients demonstrated an improvement in Raymond grade at follow-up; no cases of worsening Raymond grade were recorded, and 17 (81.0%) patients sustained a modified Raymond-Roy Classification class of I or II angiographic result at follow-up. Two delayed ruptures were recorded, both in previously coiled, symptomatic giant aneurysms where the device was used as a part of a salvage strategy. CONCLUSIONS The second-generation eCLIPs device is a viable treatment option for bifurcation aneurysms. The aneurysm occlusion rates in this initial clinical series are comparable to the initial experience with other bifurcation support devices.
Guidelines recommend the use of perfusion computed tomography (CT) to identify emergent large vessel ischemic stroke (ELVIS) patients who are likely to benefit from endovascular thrombectomy (EVT) if ...they present within 6-24 hour (late window) of stroke onset. We aim to determine if the interrater and intrarater reliability among physicians when recommending EVT is significantly different when perfusion CT or non-perfusion CT is reviewed.
A total of 30 non-consecutive patients will be selected from our institutional database comprising 3144 cranial CT scans performed for acute stroke symptoms January 2018 to August 2022. The clinical and radiologic data of the 30 patients will be presented in random order to a group of 29 physicians in two separate sessions at least three weeks apart. In each session, the physicians will evaluate each patient once with automated perfusion images and once without. We will use non-overlapping 95% confidence intervals and difference in agreement classification as criteria to suggest a difference between the Gwet AC1 statistics (κG).
The results obtained from this study, combined with the clinical outcomes data of patients categorized through the two imaging techniques and a cost-effectiveness analysis, will offer a comprehensive evaluation of the clinical utility of perfusion CT neuroimaging. Should there be no significant disparity in the reliability of decisions made by clinicians using the two neuroimaging protocols, it may be necessary to revise existing recommendations regarding neuroimaging in the later time window to align with these findings.
Spontaneous intracerebral hemorrhage (ICH), defined as nontraumatic bleeding into the brain parenchyma, is the second most common subtype of stroke, with 5.3 million cases and over 3 million deaths ...reported worldwide in 2010. Case fatality is extremely high (reaching approximately 60 % at 1 year post event). Only 20 % of patients who survive are independent within 6 months. Factors such as chronic hypertension, cerebral amyloid angiopathy, and anticoagulation are commonly associated with ICH. Chronic arterial hypertension represents the major risk factor for bleeding. The incidence of hypertension-related ICH is decreasing in some regions due to improvements in the treatment of chronic hypertension. Anticoagulant-related ICH (vitamin K antagonists and the newer oral anticoagulant drugs) represents an increasing cause of ICH, currently accounting for more than 15 % of all cases. Although questions regarding the optimal medical and surgical management of ICH still remain, recent clinical trials examining hemostatic therapy, blood pressure control, and hematoma evacuation have advanced our understanding of ICH management. Timely and aggressive management in the acute phase may mitigate secondary brain injury. The initial management should include: initial medical stabilization; rapid, accurate neuroimaging to establish the diagnosis and elucidate an etiology; standardized neurologic assessment to determine baseline severity; prevention of hematoma expansion (blood pressure management and reversal of coagulopathy); consideration of early surgical intervention; and prevention of secondary brain injury. This review aims to provide a clinical approach for the practicing clinician.
Guidelines recommend the treatment of emergent large vessel ischemic stroke (ELVIS) patients presenting beyond 6 hours of last known well time with endovascular thrombectomy (EVT) based on perfusion ...computed tomography (CT) neuroimaging. We compared the outcomes (long-term good clinical outcomes, symptomatic intracranial hemorrhage (sICH), and mortality) of ELVIS patients according to the type of CT neuroimaging they underwent.
We searched the following databases: Medline, Embase, CENTRAL, and Scopus from January 1, 2015, to June 14, 2023. We included studies of late-presenting ELVIS patients undergoing EVT that had with data for non-perfusion and perfusion CT neuroimaging. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. Data were pooled using a random effects model.
We found 7 observational cohorts. Non-perfusion versus perfusion CT was not statistically significantly different for both long-term clinical (n = 3,224; RR: 0.96; 95% CI 0.86 to 1.06; I2 = 18%) and sICH (n = 3,724; RR: 1.08 95% CI 0.60 to 1.94; I2 = 76%). Perfusion CT had less mortality (n = 3874; RR: 1.22; 95% CI 1.07 to 1.40; I2 = 0%). The certainty of these findings is very low because of limitations in the risk of bias, indirectness, and imprecision domains of the Grading of Recommendations, Assessment, Development and Evaluations.
The use of either non-perfusion or perfusion CT neuroimaging may have little to no effect on long-term clinical outcomes and sICH for late-presenting EVT patients. Perfusion CT neuroimaging may be associated with a reduced the risk of mortality. Evidence uncertainty warrants randomized trial data.