• Home
  • Chemistry
  • Astronomy
  • Energy
  • Nature
  • Biology
  • Physics
  • Electronics
  • AI Model Predicts Optimal Design for Next-Generation Stents
    A new model developed at the Georgia Institute of Technology could provide a roadmap to developing better metal stents, which are used to prop open narrowed arteries and blood vessels to keep them flowing freely.

    The model is based on a type of topological analysis that measures the complexity of the stent's geometry. The greater the complexity, the less uniform stresses will be experienced across the stent when implanted.

    "Our model can predict which geometrical structures will cause regions of stress concentration," said Yongjie Jessica Zhang, an assistant professor in the George W. Woodruff School of Mechanical Engineering at Georgia Tech. "Using this information, it's possible to redesign the stent geometry so these stress concentrations are eliminated, reducing the likelihood of fatigue failures."

    The research is reported June 1, 2022, in the journal Acta Biomaterialia. The first author of the paper is Jiahan Zhou, a Ph.D. student working with Zhang.

    Metal stents are common medical devices used to treat arterial and venous diseases. However, the long-term effectiveness is plagued by the complications associated with their structural failures, such as thrombosis (blood clots), restenosis (blocked arteries), and stent fractures.

    The stent's geometry has been identified as a critical factor in determining its structural stability and functionality. However, predicting how a specific stent geometry will impact performance is a challenge because it requires the evaluation of extremely complex structures.

    "The challenge here is that the geometries are very intricate," Zhou said. "The traditional way to analyze and improve them has mostly relied on trial-and-error experiments. This is time-consuming and costly."

    To overcome these challenges, Zhang and Zhou turned to a geometric analysis method known as "persistent homology." Unlike a typical analysis that only looks at the spatial geometry, persistent homology captures not only the geometry but also its topology, which refers to essential features that can't be changed through deformation or stretching.

    "We look into how the geometry is arranged and how those structural features impact the stress across the material," Zhang said.

    In this study, the team used persistent homology analysis to create a topology-stress map of different stent geometries. They looked at 10 variations of a widely used self-expandable stent called a Palmaz-Schatz stent. Their models predicted that increasing the complexity of the stent's geometry increased stress concentrations.

    The team is now working on developing strategies to reduce stress concentrations in the stent geometry. They are also applying the topological analysis method to study the effects of the artery wall's properties on stent performance.

    Science Discoveries © www.scienceaq.com