Loading patent…
AssigneeAlign Technology, Inc.
InventorsGuotu LI, Chao SHI, Chad Clayton BROWN, Christopher E. CRAMER, Phillip Thomas HARRIS, Adam SILL, Will NEVILLE, Ritvik BANSAL
Estimating the term…
A system may include one or more processors and memory comprising instructions that when executed by the one or more processors cause the system to receive a first dental image depicting a patient's teeth in a first view and crop the first dental image. The system may generate a score of overall image quality for the cropped first dental image using a trained neural network. The trained neural network may be trained using a plurality of ranked sub-sets of dental images and a loss function may be configured to preserve a relative ranking order of the ranked sub-sets of dental images. The sub-sets of dental images may be ranked based on perceived overall image quality. The system may determine that the score is below a threshold and request, based on the determination, a second dental image depicting the patient's teeth in the first view.
A system of assessing an image quality of a dental image, the system comprising:
The system of claim, wherein the first dental image is not resized when generating the score of overall image quality.
The system of claim, wherein the instructions that when executed by the one or more processors further cause the system to apply a mask to the cropped first dental image to mask out regions that do not correspond to teeth or teeth and gingiva prior to generating the score of overall image quality.
The system of claim, wherein the trained neural network is trained using a plurality of ranked sub-sets of dental images that are masked to exclude regions that do not correspond to teeth or teeth and gingiva.
The system of claim, wherein the instructions that when executed by the one or more processors further cause the system to output an indicator of the score of overall image quality.
The system of claim, wherein the instructions that when executed by the one or more processors further cause the system to calculate an indicator of the score of overall image quality by applying a threshold to the score of overall image quality.
The system of claim, wherein the threshold indicates good or poor overall image quality.
The system of claim, wherein the instructions that when executed by the one or more processors further cause the system to provide a link to a video showing how to capture quality dental images.
The system of claim, wherein the trained neural network is trained using ranked sub-sets of dental images comprising ranked pairs of dental images.
The system of claim, wherein perceived overall image quality includes one or more of:
The system of claim, wherein the mask is based on a neural network trained to identify teeth within each of the dental images.
The system of claim, wherein the instructions that when executed by the one or more processors further cause the system to:
The system of claim, wherein the first view is one of a center view or a side view.
A system of assessing an image quality of a dental image, the system comprising:
The system of claim, wherein the trained neural network is trained using a plurality of ranked sub-sets of dental images that are masked to mask out regions that do not correspond to teeth or teeth and gingiva.
The system of claim, wherein the trained neural network is trained using ranked sub-sets of dental images comprising ranked pairs of dental images.
The system of claim, wherein perceived overall image quality includes one or more of:
A method of assessing an image quality of a dental image, the method comprising:
The method of claim, wherein the trained neural network is trained using a plurality of ranked sub-sets of dental images that are masked to mask out regions that do not correspond to teeth or teeth and gingiva.
The method of claim, wherein the trained neural network is trained using ranked sub-sets of dental images comprising ranked pairs of dental images.
Dependencies are inferred from the claim text.