Classification and 3D modelling of 3D dento-maxillofacial structures using deep learning methods

A computer-implemented method for training a 3D deep neural network system to process 3D image data of a dento-maxillofacial structure comprising:

• a computer receiving training data, the training data including 3D input data defining one or more voxel representations of one or more dento-maxillofacial structures respectively, a voxel being associated with a radiation intensity value, the voxels of a voxel representation defining an image volume;
• processing the one or more voxel representations of the one or more dento-maxillofacial structures respectively to determine one or more 3D positional features for voxels in the one or more voxel representations, a 3D positional feature defining information about a position of at least one voxel of a voxel representation of a dento-maxillofacial structures relative to the position of a dental reference plane or the position of a dental reference object in the image volume;
• and,
• training the 3D deep neural network to classify voxels into jaw, teeth, or nerve voxels based on the training data and the one or more 3D positional features.

A computer system configured to process 3D image data of a dento-maxillofacial structure comprising:

• a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code including a pre-processing algorithm and a first first deep neural network;
• and
• a processor coupled to the computer readable storage medium, wherein responsive to executing the computer readable program code, the processor is configured to perform executable operations comprising:
• receiving 3D input data including a first voxel representation of the dento-maxillofacial structure, a voxel being associated with a radiation intensity value, the voxels of the voxel representation defining an image volume;
• determining one or more 3D positional features of the dento-maxillofacial structure based on the 3D input data, a 3D positional feature defining information about positions of voxels of the first voxel representation relative to the position of a dental reference plane, or relative to the position of a dental reference object;
• providing the first voxel representation and the one or more 3D positional features associated with the first voxel representation to the input of a first 3D deep neural network, the first deep neural network being configured to classify voxels of the first voxel representation into at least jaw, teeth, or nerve voxels,
• the first neural network being trained on the basis of a training set, the training set including 3D image data of dento-maxillofacial structures and, one or more 3D positional features derived from the 3D image data of the training set;
• receiving classified voxels of the first voxel representation from the output of the first 3D deep neural network and determining a voxel representation of at least one of the jaw, teeth or nerve tissue of the dento-maxillofacial structure on the basis of the classified voxels.