Accurate detection of vascular occlusions during endovascular thrombectomy (EVT) is critical in acute ischemic stroke (AIS). Interpretation of digital subtraction angiography (DSA) sequences poses challenges due to anatomical complexity and time constraints. This work proposes OccluNet, a spatio-temporal deep learning model that integrates YOLOX, a single-stage object detector, with transformer-based temporal attention mechanisms to automate occlusion detection in DSA sequences. We compared OccluNet with a YOLOv11 baseline trained on either individual DSA frames or minimum intensity projections. Two spatio-temporal variants were explored for OccluNet: pure temporal attention and divided space-time attention. Evaluation on DSA images from the MR CLEAN Registry revealed the model’s capability to capture temporally consistent features, achieving precision and recall of 89.02% and 74.87%, respectively. OccluNet significantly outperformed the baseline models, and both attention variants attained similar performance. Source code is available here.

Endovascular thrombectomy (EVT) aims at restoring blood flow in case of acute ischemic stroke by removing the thrombus occluding a large cerebral artery. During the procedure with stent-retriever, the thrombus is captured within the device, which is then retrieved, subjecting the thrombus to several forces, potentially leading to its fragmentation. In silico studies, along with mechanical characterisation of thrombi, can enhance our understanding of the EVT, helping the development of new devices and interventional strategies. Our group previously validated a numerical approach to study EVT able to account for thrombus fragmentation. In this study, the same methodology was employed to explore the applicability of the chosen failure criterion to EVT simulations and the impact of thrombus composition on the outcome of the in silico procedure. For the first time, human clot analogues experimental data were applied to this methodology. Clot analogues of three different compositions were tested, and a material model incorporating failure was calibrated, followed by a verification analysis. Finally, the calibrated material model was used to perform EVT simulations, combining the three tested thrombus compositions with three different stent retriever models. The experimental tests confirmed a compression-tension asymmetry in the stress-strain curves, showing decreasing stiffness with increasing the red blood cell (RBC) content. Applying the resulting material models to EVT simulations demonstrated: (i) the dependency of the failure criterion on the thrombus mesh size, (ii) a greater tendency for RBC-rich thrombi to fragment, and (iii) increased difficulty in retrieving RBC-poor thrombi compared to RBC-rich thrombi.

Intracranial vessel perforation is a peri-procedural complication during endovascular therapy (EVT). Prompt recognition is important as its occurrence is strongly associated with unfavorable treatment outcomes. However, perforations can be hard to detect because they are rare, can be subtle, and the interventionalist is working under time pressure and focused on treatment of vessel occlusions. Automatic detection holds potential to improve rapid identification of intracranial vessel perforation. In this work, we present the first study on automated perforation detection and localization on X-ray digital subtraction angiography (DSA) image series. We adapt several state-of-the-art single-frame detectors and further propose temporal modules to learn the progressive dynamics of contrast extravasation. Application-tailored loss function and post-processing techniques are designed. We train and validate various automated methods using two national multi-center datasets (i.e., MR CLEAN Registry and MR CLEAN-NoIV Trial), and one international multi-trial dataset (i.e., the HERMES collaboration). With ten-fold cross-validation, the proposed methods achieve an area under the curve (AUC) of the receiver operating characteristic of 0.93 in terms of series level perforation classification. Perforation localization precision and recall reach 0.83 and 0.70 respectively. Furthermore, we demonstrate that the proposed automatic solutions perform at similar level as an expert radiologist.

The presence of collaterals and high thrombus permeability are associated with good functional outcomes after an acute ischaemic stroke. We aim to understand the combined effect of the collaterals and thrombus permeability on cerebral blood flow during an acute ischaemic stroke. A cerebral blood flow model including the leptomeningeal collateral circulation is used to simulate cerebral blood flow during an acute ischaemic stroke. The collateral circulation is varied to capture the collateral scores: absent, poor, moderate and good. Measurements of the transit time, void fraction and thrombus length in acute ischaemic stroke patients are used to estimate thrombus permeability. Estimated thrombus permeability ranges between 10-7 and 10-4 mm2. Measured flow rates through the thrombus are small and the effect of a permeable thrombus on brain perfusion during stroke is small compared with the effect of collaterals. Our simulations suggest that the collaterals are a dominant factor in the resulting infarct volume after a stroke.

The collateral score is an important biomarker in decision making for endovascular treatment (EVT) of patients with ischemic stroke. The existing collateral grading systems are based on visual inspection and prone to subjective interpretation and interobserver variation. The purpose of our work is the development of an automatic collateral scoring method. In this work, we present a method that is inspired by human collateral scoring. Firstly, we define an anatomical region by atlas-based registration and extract vessel structures using a deep convolutional neural network. From this, high-level features based on the ratios of vessel length and volume of the occluded and the contralateral side are defined. Multi-class classification models are used to map the feature space to a four-grade collateral score and a quantitative score. The dataset used for training, validation and testing is from a registry of images acquired in clinical routine at multiple medical centers. The model performance is tested on 269 subjects, achieving an accuracy of 0.8. The dichotomized collateral score accuracy is 0.9. The error is comparable to the interobserver variation, the results are comparable to the performance of two radiologists with 10 to 30 years of experience.

Purpose: Thrombus perviousness has been associated with favorable functional outcome in acute ischemic stroke (AIS) patients. Measuring thrombus perviousness on CTA may be suboptimal due to potential delay in contrast agent arrival in occluded arteries at the moment of imaging. Dynamic sequences acquired over time can potentially overcome this issue. We investigate if dynamic CTA has added value in assessing thrombus perviousness. Methods: Prospectively collected image data of AIS patients with proven occlusion of the anterior or posterior circulation with thin-slice multi-phase CTA (MCTA) and non-contrast CT were co-registered (n = 221). Thrombus attenuation increase (TAI; a perviousness measure) was measured for the arterial, venous, and delayed phase of the MCTA and time-invariant CTAs (TiCTA). Associations with favorable clinical outcome (90-day mRS ≤ 2) were assessed using univariate and multivariable regressions and calculating areas under receiver operating curves (AUC). Results: TAI determined from the arterial phase CTA was superior in the association with favorable outcome with OR = 1.21 per 10 HU increase (95%CI 1.04–1.41, AUC 0.62, p = 0.014) compared to any other phase (venous 1.14(95%CI 1.01–1.30, AUC 0.58, p = 0.033), delayed 1.046(95%CI 0.919–1.19, AUC 0.53, p = 0.50)), and TiCTA (1.15(95%CI 1.02–1.30, AUC 0.60, p = 0.022). In the multivariable model, only TAI on arterial phase was significantly associated with favorable outcome (aOR 1.59, 95%CI 1.04–2.43, p = 0.032). Conclusion: Association between TAI with functional outcome was optimal on arterial-phase CTA such that dynamic CTA imaging has no additional benefits in current thrombus perviousness assessment, thereby suggesting that the delay of contrast arrival at the clot is a key variable for patient functional outcome.

Background and Purpose - Preclinical studies showed that thrombus permeability improves recombinant tissue-type plasminogen activator (r-tPA) efficacy. We hypothesize that thrombus permeability estimated from radiological imaging is associated with improved recanalization after treatment with intravenously administered r-tPA (r-tPA) and with better functional outcome. Methods - We assessed thrombus attenuation increase (TAI) in patients from the Dutch Acute Stroke Study with an occlusion of an intracranial artery on computed tomographic angiography. Patients were included within 9 hours after the stroke onset. After dichotomization of TAI as pervious or impervious, logistic regressions analyses were performed to estimate associations of intravenous r-tPA therapy with complete recanalization and with favorable functional outcome (modified Rankin Scale score of ≤2). Results - Three hundred eight patients matched the inclusion criteria. The median TAI was 20.1 (interquartile range, 8.5-37.8) Hounsfield unit (HU). We found a significant increase in the odds of complete recanalization with increasing TAI for patients treated with intravenous r-tPA (P=0.030). One hundred thirty-one (42%) thrombi were classified as pervious with TAI of ≥23 HU. In patients with a pervious thrombus, complete recanalization was more frequent after treatment with intravenous r-tPA than after conservative treatment (odds ratio, 6.26; 95% confidence interval, 2.4-16.8; P

BACKGROUNDANDPURPOSE: Hemodynamics are thought to play an important role in the rupture of intracranial aneurysms.Wetested whether hemodynamics, determined from computational fluid dynamics models, have additional value in discriminating ruptured and unruptured aneurysms. Such discriminative power could provide better prediction models for rupture. MATERIALS AND METHODS: A cross-sectional study was performed on patients eligible for endovascular treatment, including 55 ruptured and 62 unruptured aneurysms. Association with rupture status was tested for location, aneurysm type, and 4 geometric and 10 hemodynamic parameters. Patient-specific spatiotemporal velocities measured with phase-contrast MR imaging were used as inflow conditions for computational fluid dynamics. To assess the additional value of hemodynamic parameters, we performed 1 univariate and 2 multivariate analyses: 1 traditional model including only location and geometry and 1 advanced model that included patient-specific hemodynamic parameters. RESULTS: In the univariate analysis, high-risk locations (anterior cerebral arteries, posterior communicating artery, and posterior circulation), daughter sacs, unstable inflow jets, impingements at the aneurysm body, and unstable complex flow patterns were significantly present more often in ruptured aneurysms. In both multivariate analyses, only the high-risk location (OR, 3.92; 95% CI, 1.77-8.68) and the presence of daughter sacs (OR, 2.79; 95% CI, 1.25-6.25) remained as significant independent determinants. CONCLUSIONS: In this study population of patients eligible for endovascular treatment, we found no independent additional value of aneurysmal hemodynamics in discriminating rupture status, despite high univariate associations. Only traditional parameters (high-risk location and the presence of daughter sacs) were independently associated with ruptured aneurysms.