Stroke research with 3D printing: glass-based arterial models from CT data

A research team at the University of Sydney has developed a 3D printing method that can produce patient-specific models of carotid arteries in around two hours. The channels structured on glass slides reproduce not only the anatomy of the vessels but also the blood flow conditions. The work, published in Advanced Materials, is intended to provide new insights into the formation of blood clots and partially replace animal testing in stroke research.

“We’re not just printing blood vessels – we’re printing hope for millions at risk of stroke worldwide. With continued support and collaboration, we aim to make personalised vascular medicine accessible to every patient who needs it,” said PhD candidate Charles Zhao from the School of Biomedical Engineering, Faculty of Engineering.

“When it comes to heart attack and stroke diagnosis, speed and accuracy is key,” says Charles Zhao who is the first graduate student and founding member of the  Mechanobiology and Biomechanics Laboratory (MBL). “Clinicians typically have an approximately 12-hour decision-making window after symptom onset.”

The starting point is CT data from stroke patients, from which the geometry of the carotid arteries is reconstructed. These 3D models are scaled down to diameters of 200 to 300 micrometres, whereas real carotid arteries typically measure 5 to 7 millimetres. Instead of classic resin moulds, which according to the team can require up to ten hours of print time and have a high error rate, the vascular channels are printed directly into a layer on glass slides. From a distance they look like delicate engravings, but under the microscope they reveal a complete microfluidic system.

“This is the first-of-its-kind bioengineering endeavour in Australia, and our work is aiming to solve two crucial gaps in heart disease diagnosis and prevention, without animal testing,” said Dr Zihao Wang, the postdoctoral chief engineer of MBL group. “There is still a lot we don’t understand about the inside of our blood vessels and what creates the cascade of events that lead to blood clots, and there are currently no bespoke testing platforms that can be tailored for patients. No two patients are biologically identical, and everyone has differences in their blood vessel structure and blood, influencing their risk of blood clot disease and their treatment options.”

“Our next frontier is integrating artificial intelligence with our biofabrication platform to create true ‘digital twins’ that can predict stroke events before they happen, moving from reactive treatment to proactive prevention,” says Helen Zhao who is the postdoctoral digital scientist and operation manager of the MBL Ju lab. “Imagine a future where we can take a patient’s CT scan, rapidly print their blood vessel model, test their blood response, and use AI to predict their stroke risk years in advance.”

The researchers place particular emphasis on flow mechanics. Doctoral student Charles Zhao contributes his expertise in fluid dynamics. In the printed channels, blood flow can be simulated at different viscosities and thrombus formation can be observed in real time. The experiments show that shear stress at the vessel wall has a significant influence on platelet behaviour: in areas with high mechanical load, the team observed a seven- to tenfold increase in platelet movement – a finding associated with high blood pressure and atherosclerosis.

Professor Arnold Ju said: “We’re deeply grateful for the visionary support of the Snow Medical Research Foundation and the Snow Family through the Snow Fellowship and the National Heart Foundation Future Leader Fellowship, which has been instrumental in advancing this transformative research. My Snow Lab members have shown remarkable innovation in developing this technology, and working with our clinical partners at Royal Prince Alfred Hospital and Prince of Wales Hospital ensures our research directly addresses real patient needs. This exemplifies how engineering innovation can transform healthcare delivery, particularly aligned with future Sydney Biomedical Accelerator (SBA) goal. We’re not just printing blood vessels – we’re printing hope for millions at risk of stroke worldwide. With continued support and collaboration, we aim to make personalised vascular medicine accessible to every patient who needs it.”

In the long term, the platform is to be combined with AI methods to create digital twins from imaging data and blood parameters that can predict stroke risk. This brings closer a scenario in which doctors can print an individual “artery-on-a-chip” model within hours, test drug responses and support therapy decisions with data-driven insights.