Method for analyzing an image of a dental arch
A method for acquiring an image of a dental arch of a patient. Activation of an image acquisition apparatus so as to acquire an analysis image of the arch. Analysis of the analysis image by a deep learning device trained by a learning base. Determination, for the analysis image, as a function of the results of the analysis, of a value for an image attribute. Comparison of the image attribute value with a setpoint. Sending of an information message as a function of the comparison, the information message being related to the quality of the image acquired and/or to the position of the acquisition apparatus in relation to said arch and/or to the setting of the acquisition apparatus and/or to the opening of the mouth and/or to the wearing of a dental appliance.
1. A method for acquiring an image of a dental arch of a patient, said method comprising the following steps:
a′) activation of an image acquisition apparatus so as to acquire an image, called “analysis image”, of said arch;
b′) analysis of the analysis image by means of a deep learning device trained by means of a learning base;
c′) determination, for the analysis image, as a function of the results of the analysis in the preceding step, of a value for an image attribute;
d′) comparison of said image attribute value with a setpoint;
e′) sending of an information message as a function of said comparison, the information message being related to the quality of the image acquired or to the position of the acquisition apparatus in relation to said arch or to the setting of the acquisition apparatus or to the opening of the mouth or to the wearing of a dental appliance, or to a combination thereof,
to check whether the analysis image respects the setpoint and, if it does not respect the setpoint, to guide the operator in order for him or her to acquire a new analysis image.
2. The method as claimed in claim 1 , in which, in the step b′), at least one analysis tooth zone is identified that represents, at least partially, a tooth on said analysis image, and at least one value of a tooth attribute is determined for said at least one analysis tooth zone, and, in the step c′), the value for the image attribute is determined as a function of said tooth attribute value.
3. The method as claimed in claim 2 , in which the step b′) comprises the following steps:
1) creation of a learning base comprising more than 1000 images of dental arches, or “historical images”, each historical image comprising one or more zones each representing a tooth, or “historical tooth zones”, to each of which, for said tooth attribute, a tooth attribute value is assigned;
2) training of at least one deep learning device by means of the learning base;
3) submission of the analysis image to the deep learning device for it to determine at least one probability relating to:
the presence, at a location in said analysis image, of a zone representing, at least partially, a tooth, or “analysis tooth zone”,
the attribute value of the tooth represented on said analysis tooth zone,
4) determination, as a function of said probability, of the presence of a tooth at a position represented by said analysis tooth zone, and of the attribute value of said tooth.
4. The method as claimed in claim 3 , in which said tooth attribute is chosen from a tooth number, a tooth type, a tooth shape parameter, a tooth appearance parameter, a parameter relating to the state of the tooth, an age for the patient, or a combination of these attributes.
5. The method as claimed in claim 3 , in which, in the step 1) or 1′), a learning base is created comprising more than 10 000 historical images.
6. The method as claimed in claim 1 , in which, in the step b′), all said analysis tooth zones are identified, and at least one tooth attribute value is determined for each analysis tooth zone, and, in the step c′), the value for the image attribute is determined as a function of said tooth attribute values.
7. The method as claimed in claim 1 , in which the step b′) comprises the following steps:
1′) creation of a learning base comprising more than 1000 images of dental arches, or “historical images”, each historical image comprising an attribute value for at least one image attribute, or “image attribute value”;
2′) training of at least one deep learning device by means of the learning base;
3′) submission of the analysis image to the deep learning device for it to determine, for said analysis image, at least one probability relating to said image attribute value.
8. The method as claimed in claim 1 , in which the learning base is composed according to the following steps:
A) at an “updated” instant, production of a model of a dental arch of a patient, or “updated reference model”, and segmentation of the updated reference model so as to produce, for each tooth, a “tooth model”, and, for at least one tooth attribute, assignment of a tooth attribute value to each tooth model;
B) acquisition of at least one image of said arch, or “updated image”, in real acquisition conditions;
C) for each updated image, search for virtual acquisition conditions suitable for an acquisition of an image of the updated reference model, called “reference image”, exhibiting a maximum match with the updated image in said virtual acquisition conditions, and acquisition of said reference image;
D) identification, in the reference image, of at least one zone representing a tooth model, or “reference tooth zone” and, by comparison of the updated image and the reference image, determination, in the updated image, of a zone representing said tooth model, or “updated tooth zone”;
E) assignment, to said updated tooth zone, of the tooth attribute value of said tooth model;
F) addition of the updated image enriched with a description of said updated tooth zone and its tooth attribute value, or “historical image”, in the learning base.
9. The method as claimed in claim 1 , in which the learning base is composed according to the following steps:
A′) at an initial instant, production of a model of a dental arch of a patient, or “initial reference model”, and segmentation of the initial reference model so as to produce, for each tooth, a “tooth model”, and, for at least one tooth attribute, assignment of a tooth attribute value to each tooth model;
B′) at an updated instant, acquisition of at least one image of said arch, or “updated image”, in real acquisition conditions;
C′) search, by deformation of the initial reference model, for an updated reference model and virtual acquisition conditions suitable for an acquisition of an image of the updated reference model, called “reference image”, exhibiting a maximum match with the updated image in said virtual acquisition conditions;
D) identification, in the reference image, of at least one zone representing a tooth model, or “reference tooth zone” and, by comparison of the updated image and the reference image, determination, in the updated image, of a zone representing said tooth model, or “updated tooth zone”;
E) assignment, to said updated tooth zone, of the tooth attribute value of said tooth model;
F) addition of the updated image enriched with a description of said updated tooth zone and its tooth attribute value, or “historical image”, in the learning base.
10. The method as claimed in claim 1 , in which, to create a historical image of the learning base,
identifies one or more “historical” tooth zones on an image, called then assigns, to each identified historical tooth zone, a value for at least one tooth attribute, and/or
assigns to an image a value for at least one tooth attribute.
11. The method as claimed in claim 1 , in which the acquisition apparatus is a cellphone.
12. The method as claimed in claim 1 , in which the information message is sent by the acquisition apparatus.
13. The method as claimed in claim 1 , in which said image attribute relates to
a position or an orientation or a calibration of an acquisition apparatus used to acquire said analysis image, or a combination thereof, and/or
a quality of the analysis image, and in particular relating to the brightness, to the contrast or to the sharpness of the analysis image, and/or
the content of the analysis image, in particular to the representation of the arches, of the tongue, of the mouth, of the lips, of the jaws, of the gums, of one or more teeth or of an orthodontic appliance.
14. The method as claimed in claim 1 , in which said analysis image is analyzed and checked in real time.
15. The method as claimed in claim 1 , in which the deep learning device is a neural network.