Filters

Settings

Now showing1 - 2 of 2
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Publication, Access status: without files ,
    Quantitative analysis of periprocedural thrombus fragmentation using an automated optical detection system in a comprehensive stroke intervention training platform
    (American Society of Mechanical Engineers (ASME), 2025-04)
    Schmiech, Jonte
    ;
    De Sousa Guerreiro, Helena I.
    ;
    MacMillan, Nadine
    ;
    Sobirey, Eve
    ;
    Ramdani, Nora
    Mechanical thrombectomy for occluded large brain vessels in acute ischemic stroke has proven highly effective, but periprocedural thrombus fragmentation (PTF) remains a significant challenge that can lead to downstream embolization and incomplete recanalization. While laboratory methods exist for detailed PTF analysis, practical solutions are needed for quantitative assessment of simulated thrombectomy procedures performed on physical training models. We present a novel measurement system that enables automated detection and quantification of thrombus fragmentation events during simulated thrombectomy procedures on the HANNES neurointerventional simulator. The system employs a six-channel measurement chamber with integrated UV illumination and an optical detection setup to track fluorescent thrombus fragments. Fragments are automatically detected, assigned IDs, and measured in real-time using computer vision techniques. Validation studies comparing the system's measurements to microscopic analysis demonstrated strong correlation (Pearson's r=0.9939, p<0.001). In controlled testing, the system achieved 100% sensitivity and specificity for single fragment detection, successfully measuring fragments as small as 642 μm in radius. While adhering fragments are currently treated as single entities, the system's ability to automatically track and quantify PTF events in real-time provides an objective basis for evaluating procedural performance and comparing thrombectomy techniques in training scenarios. This novel measurement system represents a practical advance for studying thrombus fragmentation in educational settings. Integration into comprehensive training platforms like HANNES could improve understanding and management of PTF risks, potentially leading to better patient outcomes through enhanced operator training. Future studies correlating PTF metrics with physician experience and clinical results may establish performance benchmarks to help quantify individual proficiency and predict clinical competency.
    Publicationtype: Conference Paper
      40
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Publication, Access status: without files ,
    Developing and testing 3D printed models of arteriovenous malformations for embolization training in interventional neuroradiology
    (American Society of Mechanical Engineers (ASME), 2025-04)
    Sobirey, Eve
    ;
    Schmiech, Jonte
    ;
    Flottmann, Fabian
    ;
    Bechstein, Matthias
    ;
    Jungnitz, Maximilian
    The advancement of technology and the development of smaller treatment instruments has enabled the utilization of neurointerventional procedures to treat conditions affecting small vessels (>2mm). One example is an arteriovenous malformation (AVM), which can be treated by embolization. The embolization procedure is technically challenging and requires extensive training of the physicians. A number of training models have been developed which simulate the treatment of AVM in a flow model with original instruments, using the embolic agent Onyx. These AVM models are frequently intricate and time-consuming to fabricate. This paper presents the simple development of four AVM models for embolization simulations, which have been designed for integration into the existing neurointerventional training simulator, HANNES. The development and testing of the models were carried out by an interdisciplinary team of physicians and engineers. The application test was conducted in a real angiography suite with original treatment instruments and the embolic agent PHIL<sup>TM</sup> by an experienced neurointerventional physician to evaluate the AVM models in regard to geometric mapping, haptics and embolization behaviour. The models offer a range of AVM embolization simulations within the HANNES environment, catering to diverse proficiency levels, from novice to advanced and expert users.
    Publicationtype: Conference Paper
      37