-
公开(公告)号：EP4239538A3
公开(公告)日：2023-11-08
申请号：EP23181450.0
申请日：2019-07-29
发明人： BEN-GAL NGUYEN, Nitsan , ZIMMER, Benjamin D. , SOMASUNDARAM, Guruprasad , CUNLIFFE, Alexandra R. , STANKIEWICZ, Brian , COLLINS, Elisa J. , MAIDEN MUELLER, David T. , DEMLOW, Steven C. , OLSON, Cody J. , AFRIDI, Muhammad J. , AMATO, Nancy , THOMAS, Shawna L. , GUTIERREZ, Alexander , SANGARI, Arash , GANDRUD, Jonathan D.
摘要： The present invention relates to a computer-implemented method of generating a final setup for orthodontic treatment using supervised machine learning for direct mapping, comprising receiving, by one or more computer processors, a training set of patient case data which includes malocclusion states and ground truth images representing final setup states for the teeth of one or more patients, wherein the final setup states represent orthodontic post-treatment positions of a person's teeth and the malocclusion states are pre-treatment positions of the teeth and are represented by digital 3D models of teeth in an initial state, generating, by the one or more computer processors using a deep neural network, a generated output image, computing, by the one or more computer processors, a distance between a generated output image and the ground truth image representing the final state using a loss function, and minimizing, by the one or more computer processors, the distance to train, at least in part, the deep neural network.
-
公开(公告)号：EP4239538A2
公开(公告)日：2023-09-06
申请号：EP23181450.0
申请日：2019-07-29
发明人： BEN-GAL NGUYEN, Nitsan , ZIMMER, Benjamin D. , SOMASUNDARAM, Guruprasad , CUNLIFFE, Alexandra R. , STANKIEWICZ, Brian , COLLINS, Elisa J. , MAIDEN MUELLER, David T. , DEMLOW, Steven C. , OLSON, Cody J. , AFRIDI, Muhammad J. , AMATO, Nancy , THOMAS, Shawna L. , GUTIERREZ, Alexander , SANGARI, Arash , GANDRUD, Jonathan D.
IPC分类号： G06N20/00
摘要： The present invention relates to a computer-implemented method of generating a final setup for orthodontic treatment using supervised machine learning for direct mapping, comprising receiving, by one or more computer processors, a training set of patient case data which includes malocclusion states and ground truth images representing final setup states for the teeth of one or more patients, wherein the final setup states represent orthodontic post-treatment positions of a person's teeth and the malocclusion states are pre-treatment positions of the teeth and are represented by digital 3D models of teeth in an initial state, generating, by the one or more computer processors using a deep neural network, a generated output image, computing, by the one or more computer processors, a distance between a generated output image and the ground truth image representing the final state using a loss function, and minimizing, by the one or more computer processors, the distance to train, at least in part, the deep neural network.
-
公开(公告)号：EP3829481A1
公开(公告)日：2021-06-09
申请号：EP19844870.6
申请日：2019-07-29
发明人： BEN-GAL NGUYEN, Nitsan , ZIMMER, Benjamin D. , SOMASUNDARAM, Guruprasad , CUNLIFFE, Alexandra R. , STANKIEWICZ, Brian J. , COLLINS, Elisa J. , MAIDEN MUELLER, David T. , DEMLOW, Steven C. , OLSON, Cody J. , AFRIDI, Muhammad J. , AMATO, Nancy , THOMAS, Shawna L. , GUTIERREZ, Alexander , SANGARI, Arash , GANDRUD, Jonathan D.
-
-
搜索结果
3 条记录 (耗时 0.01 秒)
