Abstract: Technologies as described herein are directed towards generating optimized stent designs for treatment of tubular anatomical structures. In an exemplary computing system, the system obtains imaging data comprising images representative of anatomical structures. The system then generates a plurality of models based upon the imaging data. The models are then clustered based upon similar characteristics of the models. Certain fit parameters are applied to the models in each cluster and an optimized stent shape is generated for each cluster, wherein each optimized stent shape is representative of the models within the cluster.

Abstract: A method for determining a lumen model determines training data having a plurality of lumen centerlines. The method trains a neural network using the training data. The method determines a voxel data set corresponding to a voxel of a lumen model. The method inputs the voxel data set into the neural network. The method outputs a centerline status for the voxel from the neural network. The method determines a centerline of the lumen model using the centerline status.

Abstract: A stent design system has a memory device to store a set of instructions which, when executed by a processing device, determine a lumen cross-section of a lumen model. The stent design system determines a center point of the lumen cross-section. The stent design system determines a plurality of surface points of a surface of the lumen cross-section. The stent design system determines an updated center point using a centroid of the plurality of surface points.

Abstract: A method for designing a stent provides a user interface to display a 3D lumen model. The method receives a user input from the user interface indicating a selection of a point of the 3D lumen model. The method determines a 2D cursor position on the user interface corresponding to the selection. The method translates the 2D cursor position to a 3D lumen model position. The method determines a center point of the 3D lumen model based on a proximity to the 3D lumen model position. The method determines a diameter for a sphere based on the center point. The method positions a center of the sphere at the center point.

Abstract: A bronchial stent includes a first branch configured to widen, open, and/or mechanically support a first airway; an obstructive portion that, when the stent is deployed in the first airway, obstructs a second airway, the second airway forming a branching connection with the first airway; and a feature proximal to the obstructive portion, the feature configured to facilitate opening of the obstructive portion.