Tooth decay diagnostics using artificial intelligence
A device is disclosed for diagnosing a dental condition. The device captures image data representative of a tooth of a patient based on data obtained from a hardware device that scans the tooth. The device inputs the image data into a first supervised machine learning model, and receives, as output from the first supervised machine learning model, a plurality of biomarkers, each biomarker corresponding to a different location of the tooth. The device inputs the plurality of biomarkers into a second supervised machine learning model, and receives, as output from the second supervised machine learning model, a diagnosis of a dental condition.
1. A method for diagnosing a dental condition, the method comprising:
capturing image data representative of a tooth of a patient based on data obtained from a hardware device that scans the tooth, wherein the image data obtained from the hardware device that scans the tooth comprises Deep Penetration Optical Coherence Tomography (DPOCT) data, and wherein the image data comprises an intensity map reflective of one or more optical properties of the tooth based on the DPOCT data;
inputting the image data into a first supervised machine learning model, wherein the first supervised machine learning model is trained to detect changes in intensity between regions in the intensity map and to output respective classifications for each different location of the tooth based on the changes in intensity, each respective classification forming a biomarker, wherein the first supervised machine learning model additionally takes as input a color image of the tooth, and wherein the first supervised machine learning model is additionally trained to output biomarkers based on both the color image and the intensity map, the output of the first machine learning model excluding color data from the color image;
receiving, as output from the first supervised machine learning model, a plurality of biomarkers, each biomarker corresponding to a different location of the tooth;
inputting the plurality of biomarkers into a second supervised machine learning model; and
receiving, as output from the second supervised machine learning model, a diagnosis of a dental condition.
2. The method of claim 1 , the method further comprising accessing historical biomarkers of the tooth, wherein the historical biomarkers are input with the plurality of biomarkers into the second supervised machine learning model, and wherein the second supervised machine learning model outputs the diagnosis on the basis of both the historical biomarkers and the plurality of biomarkers.
3. The method of claim 2 , wherein inputting the historical biomarkers with the plurality of biomarkers into the second supervised machine learning model comprises computing an intensity difference for each biomarker of the plurality of biomarkers relative to the historical biomarkers and inputting each intensity difference into the second supervised machine learning model.
4. The method of claim 1 , wherein receiving, as output from the second supervised machine learning model, the diagnosis of the dental condition comprises receiving a plurality of diagnoses, each diagnosis of the plurality of diagnoses corresponding to a different one of the plurality of biomarkers.
5. A non-transitory computer-readable medium comprising memory with instructions encoded thereon for diagnosing a dental condition, the instructions, when executed, causing one or more processors to perform operations, the instructions comprising instructions to:
capture image data representative of a tooth of a patient based on data obtained from a hardware device that scans the tooth, wherein the image data obtained from the hardware device that scans the tooth comprises Deep Penetration Optical Coherence Tomography (DPOCT) data, and wherein the image data comprises an intensity map reflective of one or more optical properties of the tooth based on the DPOCT data;
input the image data into a first supervised machine learning model, wherein the first supervised machine learning model is trained to detect changes in intensity between regions in the intensity map and to output respective classifications for each different location of the tooth based on the changes in intensity, each respective classification forming a biomarker, wherein the first supervised machine learning model additionally takes as input a color image of the tooth, and wherein the first supervised machine learning model is additionally trained to output biomarkers based on both the color image and the intensity map, the output of the first machine learning model excluding color data from the color image;
receive, as output from the first supervised machine learning model, a plurality of biomarkers, each biomarker corresponding to a different location of the tooth;
input the plurality of biomarkers into a second supervised machine learning model; and
receive, as output from the second supervised machine learning model, a diagnosis of a dental condition.
6. The non-transitory computer-readable medium of claim 5 , the instructions further comprise instructions to access historical biomarkers of the tooth, wherein the historical biomarkers are input with the plurality of biomarkers into the second supervised machine learning model, and wherein the second supervised machine learning model outputs the diagnosis on the basis of both the historical biomarkers and the plurality of biomarkers.
7. The non-transitory computer-readable medium of claim 6 , wherein the instructions to input the historical biomarkers with the plurality of biomarkers into the second supervised machine learning model comprise instructions to compute an intensity difference for each biomarker of the plurality of biomarkers relative to the historical biomarkers and inputting each intensity difference into the second supervised machine learning model.
8. The non-transitory computer-readable medium of claim 5 , wherein the instructions to receive, as output from the second supervised machine learning model, the diagnosis of the dental condition comprise instructions to receive a plurality of diagnoses, each diagnosis of the plurality of diagnoses corresponding to a different one of the plurality of biomarkers.
9. A system comprising:
memory with instructions encoded thereon for diagnosing a dental condition; and
one or more processors that, when executing the instructions, are caused to perform operations comprising:
capturing image data representative of a tooth of a patient based on data obtained from a hardware device that scans the tooth, wherein the image data obtained from the hardware device that scans the tooth comprises Deep Penetration Optical Coherence Tomography (DPOCT) data, and wherein the image data comprises an intensity map reflective of one or more optical properties of the tooth based on the DPOCT data;
inputting the image data into a first supervised machine learning model, wherein the first supervised machine learning model is trained to detect changes in intensity between regions in the intensity map and to output respective classifications for each different location of the tooth based on the changes in intensity, each respective classification forming a biomarker, wherein the first supervised machine learning model additionally takes as input a color image of the tooth, and wherein the first supervised machine learning model is additionally trained to output biomarkers based on both the color image and the intensity map, the output of the first machine learning model excluding color data from the color image;
receiving, as output from the first supervised machine learning model, a plurality of biomarkers, each biomarker corresponding to a different location of the tooth;
inputting the plurality of biomarkers into a second supervised machine learning model; and
receiving, as output from the second supervised machine learning model, a diagnosis of a dental condition.
10. The system of claim 9 , the operations further comprising accessing historical biomarkers of the tooth, wherein the historical biomarkers are input with the plurality of biomarkers into the second supervised machine learning model, and wherein the second supervised machine learning model outputs the diagnosis on the basis of both the historical biomarkers and the plurality of biomarkers.
11. The system of claim 10 , wherein inputting the historical biomarkers with the plurality of biomarkers into the second supervised machine learning model comprises computing an intensity difference for each biomarker of the plurality of biomarkers relative to the historical biomarkers and inputting each intensity difference into the second supervised machine learning model.