System and method for classifying a tooth condition based on landmarked anthropomorphic measurements

1. A method for classifying a dental or maxillofacial condition, said method comprising the steps of:

parsing a received volumetric image into at least a single image frame;

localizing a landmark in a patient's dental or maxillofacial region depicted in the at least single image frame, wherein localizing the landmark is achieved by applying a deep neural network to obtain any kind of probability distribution of location of any one of the landmark, said landmark comprising of a feature or features with anatomical or pathological significance; and

classifying a dental or maxillofacial condition based on measuring a relation between two or more features within and/or between landmarks, wherein the measured relation is based on at least one of: measuring a distance between at least two points and comparing to a reference range; measuring an angle between three points in three-dimensional space and comparing to a reference range; or measuring a distance between a plane defined by three points and comparing to a reference range.

2. The method of claim 1 , wherein the landmark features localized are a cement-enamel junction (CEJ) point and bone attachment points (BAP) of a tooth and a distance measured between the two features above or below a threshold determine periodontal bone loss.

3. The method of claim 1 , wherein the landmark features localized are cephalometric features on a maxillofacial region and an angle measured between cephalometric features above or below a threshold determine an orthodontic condition.

4. The method of claim 1 , wherein the at least one received volumetric image comprises a 3-D pixel array.

5. The method of claim 1 , further comprising converting the 3-D pixel array into an array of Hounsfield Unit (HU) radio intensity measurements for pre-processing the received image.

6. The method of claim 5 , further comprising rescaling using linear interpolation for pre-processing the received image.

7. The method of claim 5 , further comprising normalization schemes to account for variations in intensity for pre-processing the received image.

8. The method of claim 1 , further comprising a step of: finding a minimum bounding rectangle around the localized landmark to achieve extraction for classification of a condition.

9. A system for classifying a dental or maxillofacial condition, said method comprising the steps of:

a processor;

a non-transitory storage element coupled to the processor;

encoded instructions stored in the non-transitory storage element, wherein the encoded instructions when implemented by the processor, configure the system to:

parse a received volumetric image into at least a single image frame; localize a landmark in a patient's dental or maxillofacial region depicted in the at least single image frame, wherein localizing the landmark is achieved by applying a deep neural network to obtain any kind of probability distribution of location of any one of the landmark, said landmark comprising of a feature or features with anatomical or pathological significance; and

classify a dental or maxillofacial condition based on measuring a relation between two or more features within and/or between landmarks, wherein the measured relation is based on at least one of: measuring a distance between at least two points and comparing to a reference range; measuring an angle between three points in three-dimensional space and comparing to a reference range; or measuring a distance between a plane defined by three points and comparing to a reference range.

10. The system of claim 9 , wherein the landmark features localized are a cement-enamel junction (CEJ) point and bone attachment points (BAP) of a tooth and a distance measured between the two features above or below a threshold determine periodontal bone loss.

11. The system of claim 9 , wherein the landmark features localized are cephalometric features on a maxillofacial region and an angle measured between cephalometric features above or below a threshold determine an orthodontic condition.

12. The system of claim 9 , wherein the at least one received volumetric image comprises a 3-D pixel array.

13. The system of claim 9 , further comprising converting the 3-D pixel array into an array of Hounsfield Unit (HU) radio intensity measurements for pre-processing the received image.

14. The system of claim 9 , further comprising rescaling using linear interpolation for pre-processing the received image.

15. The system of claim 9 , further comprising normalization schemes to account for variations in intensity for pre-processing the received image.

16. The system of claim 9 , further comprising finding a minimum bounding rectangle around the localized landmark for achieving extraction for classification of a condition.

17. A system for classifying a dental or maxillofacial condition, said system comprising:

a localization layer;

a classification layer;

a processor;

a non-transitory storage element coupled to the processor;

encoded instructions stored in the non-transitory storage element, wherein the encoded instructions when implemented by the processor, configure the system to:

localize a landmark in a patient's dental or maxillofacial region depicted in a parsed image frame by the localization layer, wherein the localizing is indicated by a minimum bounding rectangle around the landmark, said landmark comprising of a feature or features with anatomical or pathological significance; and

classify a dental or maxillofacial condition based on measuring a relation between two or more features within and/or between landmarks, wherein the measured relation is based on at least one of: measuring a distance between at least two points and comparing to a reference range; measuring an angle between three points in three-dimensional space and comparing to a reference range; or measuring a distance between a plane defined by three points and comparing to a reference range.