Fabrication of dental appliances

1. A method comprising:

receiving data identifying approximate locations of individual teeth in a three-dimensional digital dental model representing an impressioned position of a patient's dentition;

generating component models corresponding to individual teeth for each of the identified approximate locations, the component models being disposed at initial positions based on the impressioned position of the patient's dentition, wherein generating a component model for a specific individual tooth includes:

identifying candidate vertices based on determining an on-mesh geodesic distance between the vertices and the identified approximate location for the individual tooth;

mapping the identified candidate vertices to a two-dimensional Cartesian space; and

selecting at least some of the identified candidate vertices for inclusion in the component model based on the mapping to the two-dimensional Cartesian space;

determining target positions for the component models;

generating a tooth-positioning appliance design based on the determined target positions for the component models; and

causing a tooth-positioning appliance to be fabricated based on the tooth-positioning appliance design.

2. The method of claim 1 , wherein receiving data identifying the approximate locations of individual teeth includes:

causing a user interface to be displayed that shows an occlusal view of an arch of the three-dimensional digital dental model; and

receiving user inputs via the user interface to identify points on a surface of the three-dimensional digital dental model that correspond to the approximate locations of the individual teeth.

3. The method of claim 1 , wherein receiving data identifying the approximate locations of individual teeth includes:

generating a two-dimensional image representing an occlusal view of an arch of the three-dimensional digital dental model; and

identifying the approximate locations of the individual teeth using a neural network system.

4. The method of claim 3 , wherein the neural network system includes a convolutional neural network.

5. The method of claim 1 , wherein generating the component models corresponding to individual teeth for each of the identified approximate locations includes:

generating sub-models from the three-dimensional digital dental model based on the identified approximate locations;

classifying vertices of each of the sub-models using a neural network system; and

trimming the sub-models based on the classified vertices.

6. The method of claim 1 , wherein the selecting at least some of the identified candidate vertices for inclusion in the component model based on the mapping to the two-dimensional Cartesian space includes:

finding a shortest closed loop path in the identified candidate vertices; and

using the path to identify vertices that are part of a tooth associated with the identified approximate location.

7. The method of claim 1 , wherein determining the target positions for the component models includes:

determining local coordinate systems for the component models;

identifying landmarks on the component models; and

arranging the identified landmarks with respect to a target arch form.

8. The method of claim 7 , wherein arranging the identified landmarks with respect to the target arch form includes orienting the component models so as to orient the local coordinate systems of the component models in a specific relationship to the arch form.

9. The method of claim 7 , wherein the landmarks of the component models are identified using a neural network system.

10. The method of claim 1 , wherein causing the tooth-positioning appliance to be fabricated includes transmitting data corresponding to the tooth-positioning appliance design to a rapid fabrication machine.

11. The method of claim 1 , further comprising:

determining intermediate positions of the component models along a movement path between the initial positions and the target positions; and

generating an intermediate tooth-positioning appliance based on the tooth-positioning appliance design.

12. The method of claim 1 , further comprising:

selectively adding structural elements to at least some of the component models based on differences between the initial positions and the target positions.

13. A system comprising:

at least one memory including instructions; and

at least one processor that is operably coupled to the at least one memory and that is arranged and configured to execute instructions that, when executed, cause the at least one processor to:

receive a three-dimensional digital dental model representing an impressioned position of a patient's dentition;

receive data identifying approximate locations of individual teeth within the three-dimensional digital dental model;

generate component models corresponding to individual teeth for each of the identified approximate locations, the component models being disposed at initial positions based on the impressioned position of the patient's dentition, wherein generating the component models includes:

generating sub-models from the three-dimensional digital dental model based on the identified approximate locations;

classifying vertices of each of the sub-models using a neural network system; and

trimming the sub-models based on the classified vertices;

determine target positions for the component models;

generate a tooth-positioning appliance design based on the determined target positions for the component models; and

transmit data corresponding to the tooth-positioning appliance design to a rapid fabrication machine.

14. The system of claim 13 , wherein the instructions that cause the processor to receive data identifying the approximate locations of individual teeth include instructions that cause the processor to:

cause a user interface to be displayed showing an occlusal view of an arch of the three-dimensional digital dental model; and

receive user inputs via the user interface to identify points on a surface of the three-dimensional digital dental model that correspond to the approximate locations of the individual teeth.

15. The system of claim 13 , wherein the instructions that cause the processor to receive data identifying the approximate locations of individual teeth include instructions that cause the processor to:

generate a two-dimensional image representing an occlusal view of an arch of the three-dimensional digital dental model; and

identify the approximate locations of the individual teeth using a neural network system that includes a convolutional neural network.

16. The system of claim 13 , wherein the instructions that cause the processor to generate the component models corresponding to individual teeth for each of the identified approximate locations include instructions that cause the processor to:

generate sub-models from the three-dimensional digital dental model based on the identified approximate locations;

classify vertices of each of the sub-models using a neural network system; and

trim the sub-models based on the classified vertices.

17. The system of claim 13 , wherein the instructions that cause the processor to determine the target positions for the component models include instructions that cause the processor to:

determine local coordinate systems for the component models;

identify landmarks on the component models using a neural network system; and

arrange the identified landmarks with respect to a target arch form.

18. A method comprising:

generating component models from a three-dimensional digital dental model of a patient's dentition using a neural network system, the component models corresponding to individual teeth of the patient, wherein the generating component models includes:

generating a two-dimensional image representing an occlusal view of an arch of the three-dimensional digital dental model; and

identifying approximate locations of the individual teeth in the two-dimensional image using the neural network system;

determining target positions for the component models;

generating a tooth-positioning appliance design based on the determined target positions for the component models; and

causing a tooth-positioning appliance to be fabricated based on the tooth-positioning appliance design.

19. The method of claim 18 , wherein generating the component models from the three-dimensional digital dental model includes classifying at least some of the vertices of the digital dental model using the neural network system.

20. The method of claim 18 , wherein determining the target positions for the component models includes:

determining local coordinate systems for the component models;

identifying landmarks on the component models; and

arranging the identified landmarks with respect to a target arch form.