Asymmetric Flow Diverter

One approach to treat intracranial aneurysms is to use stent-like flow diverters to create thrombogenesis conditions in the aneurysm with minimal sack manipulation. Ninety percent of IA’s occur at a bifurcation, potentially in the vicinity of smaller branches. Therefore, the devices need to be flexible for easy deliverability and porous enough to avoid adjacent arterial branches occlusion, which could cause additional strokes. One of the simple Uniform Flow Diverters (UFD) being used in restricted human cases, while effective in some sidewall IA’s, is inappropriate for usage with bifurcation aneurysms.  We have been heavily involved in a research projects to investigate, develop and optimize the design of a new Asymmetric Flow Diverter (AFD), for treatment of bifurcation IA’s without the short-comings associated with current uniform flow diverting stents. The new device would consist of a low-porosity region which will cover only the aneurysm neck, to effectively divert the flow. The rest of the device will be high-porosity to reduce the chance of blockage of adjacent arteries. We treated in-vitro and in-vivo bifurcation IA’s models using the new AFDs, and studied the flow changes in the aneurysm dome and in the adjacent branches. 

  • A−C, Self-expanding Asymmetric Flow Diverter devices. Upper: Photographs of the 3 types of these stents. Lower: Angiograms showing deployment of types A and B. Type A contains a sleeve (PTFE patch) approximately 4–5 mm wide covering the central part of the stent; types B and C have a patch covering only a given stent region. Types A and B have a closed-cell structure, whereas type C has an open-cell structure. In the photographs, black arrows indicate the platinum markers used to guide the stents during the procedure and dotted arrows indicate the marker position not visible in photographs.


Angiograms of aneurysms treated with Asymmetric Flow Diverter. Column headings indicate angiogram acquisition time (prestenting, immediately poststenting, and 4 weeks later). First row, an aneurysm treated with Asymmetric Flow Diverter -A; second row, aneurysm treated with Asymmetric Flow Diverter -B; third row, aneurysm treated with Asymmetric Flow Diverter -A showing a remnant neck. 

Phantom Development

Vascular phantoms are a very important tool for benchtop testing of new devices and procedure. 3D printing offers a great opportunity to create testing settings very similar to the clinical situations.  Using our 3D printing facility we created a plethora of patient specific arterial geometries based on CT, MRI or CBCT from patient data. Patient data acquired using various imaging modalities is loaded into a 3D station for 3D rendering and processing. We manually select the vessel of interest and perform a dynamic vessel growing and export the geometry as 3D meshes in a Stereo-Lithographic (STL) File, which will be uploaded in a mesh manipulation software. Using this software we manipulate the model to merge outlets and support in order to make a friendly benchtop model. We created full Circle of Willis models, aneurysm models, cardiac models and aortic arch models. These models can be interconnected and attached to a pump to simulate physiological flow. We used these models to simulate ischemic stroke and treatment, aneurysm treatment and diagnosis.


Flow chart of images describing the manufacturing process for a patient specific phantom of a Circle of Willis. The phantom shown is designated as Phantom I for the purpose of this study.


A segmented image of the new Circle of Willis phantom, showing the configuration of five aneurysms. (Left) The new aneurysm phantom (Right)


Aneurysm stent supported coiling example in a 3D printed phantom. (a) Fluoroscopic snapshot of the initial part of the procedure with a neurovascular stent deployed across the aneurysm neck (b) Detail of the stent deployed across the aneurysm neck. (c) Final fluoroscopic snapshot of a coil mass placed in the aneurysm dome. (d) DSA showing initial arrival of the bolus contrast. Contrast flow was not significantly reduced, indicating a need for greater occlusion through deployment of additional coils. 


Several printed phantoms connected together better simulate the traversal of a catheter from the femoral arteries to the treatment site.

  • New 3D printed cardiac vasculature phantom showing direction of flow.


A catheter was advanced into the new cardiac phantom through the aortic arch and contrast was deployed to show vessel patency. (a) Result of averaged DSA sequence. (b) Detail of the coronary arteries. (c) Sequence of DSA frames showing arrival of contrast. A cardiac phantom diversifies the suite of phantoms, which can be used for  a wide range of cardiovascular interventions.

Research Focus

Ciprian Ionita Ph.D., Stephen Rudin, Ph.D., and Hui Meng Ph.D., oversee the device development and testing group at the Toshiba Stroke and Vascular Research Center.

Over the years they developed and tested endovascular device prototypes and simulations in patient relevant physiology in a lab setting. Dr. Ionita developed the first balloon deployable asymmetric flow diverters nearly eight years ago and test them in idealized aneurysms phantoms using Particle Image Velocimetry. Over the years, at Toshiba Stroke and Vascular Research Center (TSVRC), the team developed an entire chain of manufacturing and testing of such devices, from the CAD design to an in-vivo friendly prototype.

The research team developed unique methods to develop patient specific phantoms used to test endovascular devices.  Over the years they developed an entire chain of manufacturing and testing of such phantoms, from the CT scan to a reliable and easy-to-use phantom. Dr. Ionita has dedicated a good part of his research to the idea of endovascular treatment planning using 3D printing. He developed a 3D printing facility which could serve various endovascular research and the clinical needs of TSVRC collaborators. The new workflow developed by Dr Ionita’s team allows manufacturing of complex patient specific phantoms within 24 hours from the patient volume acquisition.