Journal of NeuroInterventional Surgery最新文献

{"title":"Tirofiban versus aspirin to prevent in-stent thrombosis after emergent carotid artery stenting in acute ischemic stroke.","authors":"Manuel Medina-Rodriguez, Diego Villagran, Antonio Cristobal Luque-Ambrosiani, Juan Antonio Cabezas-Rodríguez, Leire Ainz-Gómez, Pablo Baena Palomino, Blanca Pardo-Galiana, Aynara Zamora, Asier de Albóniga-Chindurza, Marta Aguilar-Perez, Alejandro González, Francisco Moniche, Elena Zapata-Arriaza","doi":"10.1136/jnis-2024-021845","DOIUrl":"10.1136/jnis-2024-021845","url":null,"abstract":"<p><strong>Background: </strong>Several antithrombotic treatments during emergent carotid artery stenting (eCAS) have been proposed, but an appropriate protocol to balance risk-benefit is not well known.</p><p><strong>Objective: </strong>To investigate the efficacy and safety of tirofiban compared with aspirin in patients with acute ischemic stroke undergoing eCAS.</p><p><strong>Methods: </strong>We conducted a retrospective single-center study of the prospective ARTISTA Registry, including patients with atherosclerotic internal carotid artery occlusion treated with eCAS. Two groups, according to antiplatelet drug, were studied: aspirin (250-500 mg single-dose) versus tirofiban (500 μg bolus+200 μg/h). Primary outcomes were the rate of in-stent thrombosis and symptomatic intracranial hemorrhage (sICH) within the first 24 hours.</p><p><strong>Results: </strong>During the period 2019-2023, 181 patients were included, 103 received aspirin, 78 tirofiban; 149 (82.3%) had tandem lesions. The primary efficacy outcome occurred in 9 (9.4%) in the aspirin group, as compared with 1 (1.3%) in the tirofiban group (adjusted odds ratio (aOR)=0.11, 95% CI 0.01 to 0.98; P=0.048). The primary safety outcome was detected in 12 (11.7%) in the aspirin group, as compared with 2 (2.6%) in the tirofiban group (aOR=0.16, 95% CI 0.03 to 0.87; P=0.034). The tirofiban group presented a lower risk of parenchymal hemorrhage (18 (17.4%) vs 4 (5.2%), aOR=0.27, 95% CI 0.09 to 0.88; P=0.029) and an increased rate of excellent recanalization (expanded Treatment in Cerebral Infarction (eTICI) 2c-3) (50 (48.5%) vs 54 (69.2%); aOR=2.15, 95% CI 1.12 to 4.13; P=0.02). There were no differences in functional outcomes or mortality at 3 months.</p><p><strong>Conclusions: </strong>Periprocedural antithrombotic therapy with tirofiban was associated with a lower risk of in-stent thrombosis and sICH at 24 hours from eCAS compared with aspirin. Prospective randomized clinical trials are needed to confirm our results.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"697-702"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141436981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pioneering artificial intelligence-based real time assistance for intracranial liquid embolization in humans: an initial experience.","authors":"Yuya Sakakura, Osamu Masuo, Takeshi Fujimoto, Tomoaki Terada, Kenichi Kono","doi":"10.1136/jnis-2024-022001","DOIUrl":"10.1136/jnis-2024-022001","url":null,"abstract":"<p><strong>Background: </strong>Liquid embolization in neuroendovascular procedures carries the risk of embolizing an inappropriate vessel. Operators must pay close attention to multiple vessels during the procedure to avoid ischemic complications. We report our experience with real time artificial intelligence (AI) assisted liquid embolization and evaluate its performance.</p><p><strong>Methods: </strong>An AI-based system (Neuro-Vascular Assist, iMed technologies, Tokyo, Japan) was used in eight endovascular liquid embolization procedures in two institutions. The software automatically detects liquid embolic agent on biplane fluoroscopy images in real time and notifies operators when the agent reaches a predefined area. Safety, efficacy, and accuracy of the notifications were evaluated using recorded videos.</p><p><strong>Results: </strong>Onyx or n-butyl-2-cyanoacrylate (NBCA) was used in the treatment of arteriovenous malformation, dural arteriovenous fistula, meningioma, and chronic subdural hematoma. The mean number of true positive and false negative notifications per case was 31.8 and 2.8, respectively. No false positive notifications occurred. The precision and recall of the notifications were 100% and 92.0%, respectively. In 28.3% of the true positive notifications, the operator immediately paused agent injection after receiving the notification, which demonstrates the potential effectiveness of the AI-based system. No adverse events were associated with the notifications.</p><p><strong>Conclusions: </strong>To the best of our knowledge, this is the first report of real time AI assistance with liquid embolization procedures in humans. The system demonstrated high notification accuracy, safety, and potential clinical usefulness in liquid embolization procedures. Further research is warranted to validate its impact on clinical outcomes. AI-based real time surgical support has the potential to advance neuroendovascular treatment.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"748-752"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141468696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The changing Food and Drug Administration landscape and what it means for neurointervention.","authors":"Matthew R Fusco","doi":"10.1136/jnis-2025-023765","DOIUrl":"https://doi.org/10.1136/jnis-2025-023765","url":null,"abstract":"","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":"17 7","pages":"669-670"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144310045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparing the safety and effectiveness of overlapping stents with flow diverters for unruptured vertebral artery dissecting aneurysms.","authors":"Hyung Jun Kim, Na Rae Yang, Tae Keun Jee, Je-Young Yeon, Keon-Ha Kim, Jong-Soo Kim, Woo-Keun Seo, Pyoung Jeon","doi":"10.1136/jnis-2024-021762","DOIUrl":"10.1136/jnis-2024-021762","url":null,"abstract":"<p><strong>Background: </strong>Endovascular treatment for vertebral artery dissecting aneurysms (VADAs) includes overlapping stents and flow diverters. This study compared the safety and effectiveness of overlapping stents and flow diverters for unruptured VADAs.</p><p><strong>Methods: </strong>We retrospectively enrolled patients with unruptured VADAs who underwent overlapping stents or flow diverters at two tertiary hospitals in South Korea. The primary clinical outcome was the occurrence of stroke. The primary angiographic outcomes (>12 months) were categorized as regression, no decrease in size, recanalization, or stent occlusion, of which only regression was defined as a favorable angiographic outcomes.</p><p><strong>Results: </strong>Of the 146 patients with VADAs, 25 (17.1%) had flow diverters and 121 (82.9%) had overlapping stents. For the primary angiographic outcomes over 12 months, the rate of favorable angiographic outcomes for flow diverters was 81.8% and for overlapping stents (triple stents) was 98.8% (P=0.006). In the multivariale analysis, after adjusting for partially thrombosed aneurysms, aneurysm shape, non-dominant vessel, posterior inferior cerebellar artery involvement, and procedure type, overlapping stents (triple stents) was not associated with favorable angiographic outcomes compared with flow diverters (OR 7.040, 95% CI 0.549 to 90.294; P=0.134), but partially thrombosed aneurysms was inversely associated with favorable angiographic outcomes (OR 0.056, 95% CI 0.005 to 0.589; P=0.016). The primary clinical outcome followed up to the last angiography did not occur in all patients.</p><p><strong>Conclusion: </strong>There was no difference in safety and effectiveness between overlapping stents and flow diverters in unruptured VADAs. Further endovascular treatment studies are needed regarding the association of partially thrombosed aneurysms with unfavorable angiographic outcomes.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"725-730"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141446317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Two-stage convolutional neural network for segmentation and detection of carotid web on CT angiography.","authors":"Hulin Kuang, Xianzhen Tan, Fouzi Bala, Jialiang Huang, Jianhai Zhang, Ibrahim Alhabli, Faysal Benali, Nishita Singh, Aravind Ganesh, Shelagh B Coutts, Mohammed A Almekhlafi, Mayank Goyal, Michael D Hill, Wu Qiu, Bijoy K Menon","doi":"10.1136/jnis-2024-021782","DOIUrl":"10.1136/jnis-2024-021782","url":null,"abstract":"<p><strong>Background: </strong>Carotid web (CaW) is a risk factor for ischemic stroke, mainly in young patients with stroke of undetermined etiology. Its detection is challenging, especially among non-experienced physicians.</p><p><strong>Methods: </strong>We included patients with CaW from six international trials and registries of patients with acute ischemic stroke. Identification and manual segmentations of CaW were performed by three trained radiologists. We designed a two-stage segmentation strategy based on a convolutional neural network (CNN). At the first stage, the two carotid arteries were segmented using a U-shaped CNN. At the second stage, the segmentation of the CaW was first confined to the vicinity of the carotid arteries. Then, the carotid bifurcation region was localized by the proposed carotid bifurcation localization algorithm followed by another U-shaped CNN. A volume threshold based on the derived CaW manual segmentation statistics was then used to determine whether or not CaW was present.</p><p><strong>Results: </strong>We included 58 patients (median (IQR) age 59 (50-75) years, 60% women). The Dice similarity coefficient and 95th percentile Hausdorff distance between manually segmented CaW and the algorithm segmented CaW were 63.20±19.03% and 1.19±0.9 mm, respectively. Using a volume threshold of 5 mm³, binary classification detection metrics for CaW on a single artery were as follows: accuracy: 92.2% (95% CI 87.93% to 96.55%), precision: 94.83% (95% CI 88.68% to 100.00%), sensitivity: 90.16% (95% CI 82.16% to 96.97%), specificity: 94.55% (95% CI 88.0% to 100.0%), F1 measure: 0.9244 (95% CI 0.8679 to 0.9692), area under the curve: 0.9235 (95%CI 0.8726 to 0.9688).</p><p><strong>Conclusions: </strong>The proposed two-stage method enables reliable segmentation and detection of CaW from head and neck CT angiography.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"769-774"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141446363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preclinical in vitro and in vivo results of the new silk vista flow diverter with P8RI coating.","authors":"Jonathan Cortese, Géraud Forestier, Sylvia M Bardet, Marie-Laure Perrin, Maxime Baudouin, Alexis Belgacem, Romain Chauvet, Voahirana Ratsimbazafy, Gregory Sasselina, Daphnée Chandellier, Jérémy Mounier, Claude Couquet, Florence Bosselut, Laurent Spelle, Charbel Mounayer, Faraj Terro, Aymeric Rouchaud","doi":"10.1136/jnis-2024-021694","DOIUrl":"10.1136/jnis-2024-021694","url":null,"abstract":"<p><strong>Background: </strong>Flow diverting stents (FDS) have transformed the treatment of intracranial aneurysms; however, their metallic structure associated with their intra-luminal positioning hamper angiographic and clinical outcomes. Therefore, there is a need to develop FDS with optimized surfaces that reduce thrombogenicity while promoting the healing process and endothelialization.</p><p><strong>Methods: </strong>P8RI, a peptide mimicking the CD31 protein, was previously developed and grafted onto Silk Vista (SV) FDS. P8RI-SV and bare-SV were used in vitro in a blood loop model to test their hemocompatibility using human whole blood and in vivo using the rabbit elastase model for optical coherence tomography (OCT) comparisons of neointimal formation at day 5 and day 28.</p><p><strong>Results: </strong>After blood loop incubation, P8RI-SV showed significant reduction in fibrin binding (p=0.004) and platelet adhesion (p=0.041) compared with bare-SV. Similarly, derivative markers measured in blood, thromboxane B2 (platelet activation) and Thrombin-Antithrombin III complexes (coagulation activation), were also significantly reduced in the P8RI-SV group (both p=0.002). In vivo, complete or near-complete occlusion was reached in all aneurysms (n=6) at day 28. Excellent rate of stent-coverage ratio was obtained at day 5 (89.3% (79.1%-98.7%)) comparable to the observation at day 28 (91.8% (79.1%-100%); p=0.44). These rates were significantly higher compared with bare-SV at day 5 (77.8% (58.3%-86.8%); p<0.001) and at day 28 (67.7% (52.6%-88.9%); p<0.0001).</p><p><strong>Conclusion: </strong>In vitro results confirm enhanced hemocompatibility with a significant anti-thrombotic effect of the P8RI-SV. In vivo results provide evidence of rapid neo-intimal growth reaching near-complete tissue healing as early as day 5 in a rabbit model.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"775-781"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141446362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cost-effectiveness of endovascular thrombectomy for acute ischemic stroke with established large infarct in Germany: a decision tree and Markov model.","authors":"Sophie Gottschalk, Hans-Helmut König, Fabien Subtil, Susanne Bonekamp, Angelique Denis, Anne Hege Aamodt, Blanca Fuentes, Elke R Gizewski, Michael D Hill, Antonin Krajina, Laurent Pierot, Claus Ziegler Simonsen, Kamil Zeleňák, Martin Bendszus, Götz Thomalla, Judith Dams","doi":"10.1136/jnis-2024-021837","DOIUrl":"10.1136/jnis-2024-021837","url":null,"abstract":"<p><strong>Background: </strong>Recent studies, including the TENSION trial, support the use of endovascular thrombectomy (EVT) in acute ischemic stroke with large infarct (Alberta Stroke Program Early Computed Tomography Score (ASPECTS) 3-5).</p><p><strong>Objective: </strong>To evaluate the cost-effectiveness of EVT compared with best medical care (BMC) alone in this population from a German healthcare payer perspective.</p><p><strong>Methods: </strong>A short-term decision tree and a long-term Markov model (lifetime horizon) were used to compare healthcare costs and quality-adjusted life years (QALYs) between EVT and BMC. The effectiveness of EVT was reflected by the 90-day modified Rankin Scale (mRS) outcome from the TENSION trial. QALYs were based on published mRS-specific health utilities (EQ-5D-3L indices). Long-term healthcare costs were calculated based on insurance data. Costs (reported in 2022 euros) and QALYs were discounted by 3% annually. Cost-effectiveness was assessed using incremental cost-effectiveness ratios (ICERs). Deterministic and probabilistic sensitivity analyses were performed to account for parameter uncertainties.</p><p><strong>Results: </strong>Compared with BMC, EVT yielded higher lifetime incremental costs (€24 257) and effects (1.41 QALYs), resulting in an ICER of €17 158/QALY. The results were robust to parameter variation in sensitivity analyses (eg, 95% probability of cost-effectiveness was achieved at a willingness to pay of >€22 000/QALY). Subgroup analyses indicated that EVT was cost-effective for all ASPECTS subgroups.</p><p><strong>Conclusions: </strong>EVT for acute ischemic stroke with established large infarct is likely to be cost-effective compared with BMC, assuming that an additional investment of €17 158/QALY is deemed acceptable by the healthcare payer.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":"683-690"},"PeriodicalIF":4.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141436957","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Outcomes of medical and surgical management in infectious intracranial aneurysms: a multicenter cohort analysis.","authors":"Ali M Alawieh, Laurie Dimisko, Youssef M Zohdy, Andrew B Koo, Hassan Saad, Bachar El Baba, Sarah Newman, Jonathan A Grossberg, Charles Matouk, C Michael Cawley, Gustavo Pradilla, Andrew Reisner, W Christopher Fox, Carlos Perez-Vega, Jan-Karl Burkhardt, Mohamed Salem, Pascal Jabbour, Kareem El Naamani, Richard F Schmidt, M Reid Gooch, Robert M Starke, Ahmed Abdelsalam, Victor M Lu, Michael Levitt, Reda M Chalhoub, Firas Kobeissy, Alejandro M Spiotta, Daniel Barrow, Brian M Howard","doi":"10.1136/jnis-2025-023421","DOIUrl":"10.1136/jnis-2025-023421","url":null,"abstract":"<p><strong>Introduction: </strong>Infectious intracranial aneurysms (IIAs) are rare but serious complications of systemic infections, particularly infective endocarditis. These aneurysms are prone to rupture, leading to significant morbidity and mortality. Management strategies lack consensus due to the rarity of the condition and reliance on small case series. This study examines the clinical management of IIAs using data from a large multicenter cohort.</p><p><strong>Methods: </strong>A retrospective registry-based cohort study was conducted across 11 tertiary care centers in the USA between 2018 and 2023. Patients with IIAs were identified based on clinical and radiographic criteria. The primary outcome was treatment failure defined as persistence, growth, or rupture of the aneurysm. Secondary outcomes were mortality and the modified Rankin Scale (mRS) score at 90 days and 1 year. Multivariate logistic regressions were used to identify outcome predictors.</p><p><strong>Results: </strong>A total of 104 patients with 166 aneurysms were included, with a median age of 43 years. Medical management was successful in 56% of cases, with failure often within 18 days of initiation. Predictors of failure included younger age, larger aneurysm size, and rupture at presentation. Surgical and endovascular interventions achieved higher success rates with better outcomes. At 90 days, 57% of patients achieved functional independence (mRS 0-2), while the mortality rate was 24%.</p><p><strong>Conclusion: </strong>This study highlights the limitations of medical management for IIAs and underscores the need for early surgical or endovascular intervention in high-risk patients. Outcome predictors aid clinical decision-making, optimizing patient management. Further research is needed to standardize management guidelines for IIAs.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":""},"PeriodicalIF":4.5,"publicationDate":"2025-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144132448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Nationwide multicenter experience of contrast-induced encephalopathy following neurointervention: clinical course and outcomes.","authors":"Frederick P Mariajoseph, Leon T Lai, Adrian Praeger, Justin Moore, Ronil V Chandra, Hamed Asadi, Peter Fawzy, Laetitia de Villiers, Tony Goldschlager, Calvin Gan, Kevin Zhou, Albert Ho Yuen Chiu, Boaz Kim, Ferdi Miteff, Ramon Martin Francisco Bañez, Davor Pavlin-Premrl, Winston Chong, Robert Fang, Kate Mahady, Sophie Dunkerton, Brendan Steinfort, Bjoern Picker, Lee-Anne Slater","doi":"10.1136/jnis-2025-023533","DOIUrl":"https://doi.org/10.1136/jnis-2025-023533","url":null,"abstract":"<p><strong>Background: </strong>Contrast-induced encephalopathy (CIE) is an increasingly observed complication following neurointervention, but remains poorly defined with limited evidence for clinical decision-making. We sought to characterize the stereotypical clinical features of CIE in a nationwide, multicenter cohort.</p><p><strong>Methods: </strong>A multicenter cohort study was conducted between 10 neurovascular sites across Australia. Patients were screened according to the previously proposed Australian diagnostic criteria. Descriptive analysis was conducted to characterize the clinical course and outcomes of CIE, and associations between clinical and radiological variables on patient outcomes were analyzed using Fisher's exact and χ² tests.</p><p><strong>Results: </strong>A total of 56 patients (median age 65 years) were included. The median contrast volume was 170 mL (IQR 140-229). Median time to symptom onset was 6 hours (IQR 1-12), with frequent symptoms including motor deficit (55.4%), dysphasia (39.3%), and confusion (35.7%). Common radiological findings included sulcal effacement (45.5%) and subarachnoid contrast staining (30.9%) on CT. Hemianopia (p=0.001) and cortical blindness (p=0.018) were associated with posterior circulation interventions, while motor deficit was correlated with anterior circulation interventions (p=0.001). At discharge, 87.5% of patients achieved complete resolution of symptoms, of which 69.4% achieved complete recovery within 72 hours.</p><p><strong>Conclusion: </strong>CIE is a recognized complication of neurointervention. Symptoms occur within hours of contrast administration and correlate with the territory of contrast administration. Most patients achieve complete symptom resolution. Ongoing investigation is required to further define CIE as a clinical entity.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":""},"PeriodicalIF":4.5,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144285009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Minimizing human-induced variability in quantitative angiography for a robust and explainable AI-based occlusion prediction in flow diverter-treated aneurysms.","authors":"Parmita Mondal, Mohammad Mahdi Shiraz Bhurwani, Swetadri Vasan Setlur Nagesh, Pui Man Rosalind Lai, Jason M Davies, Elad I Levy, Kunal Vakharia, Michael Levitt, Adnan H Siddiqui, Ciprian N Ionita","doi":"10.1136/jnis-2025-023416","DOIUrl":"10.1136/jnis-2025-023416","url":null,"abstract":"<p><strong>Background: </strong>Bias from contrast injection variability is a significant obstacle to accurate intracranial aneurysm (IA) occlusion prediction using quantitative angiography (QA) and deep neural networks (DNNs). This study explores bias removal and explainable AI (XAI) for outcome prediction.</p><p><strong>Objective: </strong>To implement an injection bias removal algorithm for reducing QA variability and examine the impact of XAI on the reliability and interpretability of deep learning models for occlusion prediction in flow diverter-treated aneurysms.</p><p><strong>Methods: </strong>This study used angiograms from 458 patients with flow diverter-treated IAs, with 6-month follow-up defining occlusion status. We minimized injection variability by deconvolving the parent artery input to isolate the aneurysm's impulse response, then reconvolving it with a standardized injection curve. A DNN trained on these QA-derived biomarkers predicted 6-month occlusion. Local Interpretable Model-Agnostic Explanations (LIME) identified the key imaging features influencing the model, ensuring transparency and clinical relevance.</p><p><strong>Results: </strong>The DNN trained with uncorrected QA parameters achieved a mean area under the receiver operating characteristic curve (AUROC) of 0.60±0.05 and an accuracy of 0.58±0.03. After correcting for injection bias by deconvolving the parent artery input and reconvolving it with a standardized injection curve, the DNN's AUCROC increased to 0.79±0.02 and accuracy to 0.73±0.01. Sensitivity and specificity were 67.61±1.93% and 76.19±1.12%, respectively. LIME plots were added for each prediction to enhance interpretability.</p><p><strong>Conclusions: </strong>Standardizing QA parameters via injection bias correction improves occlusion prediction accuracy for flow diverter-treated IAs. Adding explainable AI (eg, LIME) clarifies model decisions, demonstrating the feasibility of clinically interpretable AI-based outcome prediction.</p>","PeriodicalId":16411,"journal":{"name":"Journal of NeuroInterventional Surgery","volume":" ","pages":""},"PeriodicalIF":4.5,"publicationDate":"2025-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144285007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}