Image-based aligner fit evaluation

1. A method for image-based orthodontic treatment, the method comprising:

receiving image data of a patient's dentition and an orthodontic appliance;

identifying, from the image data, the orthodontic appliance by performing semantic segmentation to classify each pixel of the image data into one of a plurality of classes;

training a neural network to perform the semantic segmentation by:

inputting an image data set for semantic segmentation by the neural network;

computing an error between an output of the neural network and a mask data set corresponding to the image data set; and

adjusting parameters of the neural network to reduce the error;

calculating a misalignment height of a misalignment of the orthodontic appliance with respect to the patient's dentition;

determining whether the misalignment height satisfies a misalignment threshold; and

in response to satisfying the misalignment threshold, providing a notification.

2. A method for image-based orthodontic treatment, the method comprising: receiving image data of a patient's dentition and an orthodontic appliance;

identifying, from the image data, the orthodontic appliance by performing semantic segmentation to classify each pixel of the image data into one of a plurality of classes by, for each pixel:

determining a probability of the pixel matching each of the plurality of classes; and

classifying the pixel into the one of the plurality of classes based on a corresponding highest probability value;

calculating a misalignment height of a misalignment of the orthodontic appliance with respect to the patient's dentition;

determining whether the misalignment height satisfies a misalignment threshold; and

in response to satisfying the misalignment threshold, providing a notification.

3. The method of claim 2 , wherein the misalignment height is calculated from aggregating a plurality of identified misalignments.

4. The method of claim 3 , wherein calculating the misalignment height further comprises:

determining a height value at a particular percentile of the plurality of identified misalignments; and

setting the misalignment height based on the determined height value.

5. The method of claim 2 , wherein calculating the misalignment height further comprises subtracting a thickness offset to simulate a material thickness of the orthodontic appliance.

6. The method of claim 5 , wherein the thickness offset is obtained from a treatment plan for the patient.

7. The method of claim 2 , further comprising tracking the misalignment height over time using image data over time.

8. The method of claim 7 , further comprising:

identifying a misalignment trend from the tracked misalignment height;

determining whether the misalignment trend satisfies a misalignment trend threshold; and

in response to satisfying the misalignment trend threshold, providing a notification.

9. The method of claim 2 , further comprising providing a visual overlay of the misalignment.

10. The method of claim 2 , wherein identifying the orthodontic appliance further comprises:

applying one or more filters to the image data to determine a tooth edge and an orthodontic appliance edge.

11. The method of claim 2 , wherein identifying the orthodontic appliance further comprises:

evaluating a color value of each pixel to identify a tooth portion without the orthodontic appliance and a tooth portion with the orthodontic appliance.

12. A method for image-based orthodontic treatment, the method comprising:

receiving image data of a patient's dentition and an orthodontic appliance;

identifying, from the image data, the orthodontic appliance by performing semantic segmentation to classify each pixel of the image data into one of a plurality of classes, wherein the plurality of classes includes a tooth class indicating teeth of the patient, a gap class indicating a gap between the orthodontic appliance and a corresponding gingival edge, and a space class indicating a space between an incisal edge of the orthodontic appliance and an incisal edge of a corresponding tooth, and wherein the gap class and the space class correspond to a misalignment of the orthodontic appliance with respect to the patient's dentition,

calculating a misalignment height of the misalignment of the orthodontic appliance with respect to the patient's dentition;

determining whether the misalignment height satisfies a misalignment threshold; and

in response to satisfying the misalignment threshold, providing a notification.

13. The method of claim 12 , wherein calculating the misalignment height further comprises subtracting a thickness offset to simulate a material thickness of the orthodontic appliance.

14. The method of claim 12 , further comprising:

tracking the misalignment height over time using image data over time;

identifying a misalignment trend from the tracked misalignment height;

determining whether the misalignment trend satisfies a misalignment trend threshold; and

in response to satisfying the misalignment trend threshold, providing a notification.

15. A method for image-based orthodontic treatment, the method comprising:

receiving image data of a patient's dentition and an orthodontic appliance;

identifying, from the image data, the orthodontic appliance,

calculating a misalignment height of a misalignment of the orthodontic appliance with respect to the patient's dentition from a pixel height of the misalignment by:

determining a pixel height of an incisor identified in the image data;

obtaining an incisor measurement in a standard unit of measurement;

determining, using the pixel height of the incisor and the incisor measurement, a conversion factor between pixels and the standard unit of measurement; and

converting, using the conversion factor, the misalignment height from pixels to the standard unit of measurement;

determining whether the misalignment height satisfies a misalignment threshold; and

in response to satisfying the misalignment threshold, providing a notification.

16. The method of claim 15 , wherein the incisor measurement is obtained from a treatment plan for the patient.

17. The method of claim 15 , further comprising:

tracking the misalignment height over time using image data over time;

identifying a misalignment trend from the tracked misalignment height;

determining whether the misalignment trend satisfies a misalignment trend threshold; and

in response to satisfying the misalignment trend threshold, providing a notification.

18. A method for image-based orthodontic treatment, the method comprising: further comprising

receiving image data of a patient's dentition and an orthodontic appliance;

identifying, from the image data, the orthodontic appliance,

calculating a misalignment height of a misalignment of the orthodontic appliance with respect to the patient's dentition from a pixel height of the misalignment;

determining whether the misalignment height satisfies at least one of a plurality of misalignment thresholds;

in response to satisfying the misalignment threshold, providing a notification; and

providing a visual overlay of the misalignment, wherein each range between each of the plurality of misalignment thresholds corresponds to a unique color.

19. The method of claim 18 , wherein calculating the misalignment height further comprises subtracting a thickness offset to simulate a material thickness of the orthodontic appliance.

20. The method of claim 18 , further comprising:

tracking the misalignment height over time using image data over time;

identifying a misalignment trend from the tracked misalignment height;

determining whether the misalignment trend satisfies a misalignment trend threshold; and

in response to satisfying the misalignment trend threshold, providing a notification.