DETAILED ACTIONS
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
Claims 1-20 are pending.
Response to Amendment
The amendment filed 06/22/2026 has been entered in full.
Response to Arguments
Applicant's arguments filed 03/20/2026 have been fully considered but they are not persuasive.
On pages 9-11 of the Remarks, Applicants contend that Coulombe does not teach or suggest “predicting, for each 2D slice of the plurality of 2D slices, using a trained machine learning (ML) model, a 2D information inferred from the 2D slice using the 2D slice as an input“ as required by amended independent claim 1. Applicants argue that Coulombe’s use of descriptor matrices for matching does not disclose the limitation in question. Applicants also argue that Coulombe’s Fourier neural operator is not a trained machine learning model The Examiner respectfully disagrees with this characterization of Coulombe and submits that the reference does indeed disclose the limitation in question.
The Fourier neural operator that was used by Coulombe to output a 2D matrix information which is analogous to the 2D information in the limitation is an example of a trained machine learning model. A neural operator is a specialized class of deep learning architecture which is designed to learn maps between infinite-dimensional function spaces (https://en.wikipedia.org/wiki/Neural_operators). Therefore, under BRI the Fourier neural operator is a kind of trained machine learning model. Coulombe’s Fourier neural operator predicts or outputs a 2D descriptor matrix. Coulombe also says generating a description matrix for each slice of the STL file, [0088], “descriptor matrix comprises a number of the plurality of cross-section slices and a second y dimension of the first descriptor matrix comprising a number of the plurality of radial lengths in each slice”. Then, Coulombe further discloses converting the plurality of 2D description matrix into a 3D model, [0093], “Image E in FIG. 6 is an illustration of a visualization map 210 of the radial encoder output 2D matrix 208 visualized as a pixel array. The visualization map 210 of the radial encoder 106 output can be visualized on a graphical user interface, where each entry in the 2D matrix 208 is converted to a pixel value, referred to herein as a visualization map 210. Image E illustrates an example of a greyscale (black and white range) visualization mapping of matrix 208, however it is understood that the same can be displayed in color as desired. The visualization mapping and/or 2D matrix can then be converted back into a 3D representation of the tooth and displayed on a graphical user interface during the dental procedure.” Therefore, Coulombe teaches the limitation “predicting, for each 2D slice of the plurality of 2D slices, using a trained machine learning (ML) model, a 2D information inferred from the 2D slice using the 2D slice as an input“.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-5, 11-13, and 18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Coulombe et al., (WO 2024/103143 A1), hereinafter referred to as Coulombe.
Claim 1
Coulombe discloses a method (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), comprising:
computing a central axis for a three-dimensional (3D) model (Coulombe, [0120], “Assigning and aligning the z-axis of each tooth or crown to a central axis and using the same dental file segmentation algorithm along the same z-axis with the same common centroid for all descriptor matrixes enables matching with other crown and tooth descriptor matrixes with the same reference loci”, [0010], “scanning a tooth with a scanner to obtain a three dimensional (3D) image of the tooth”)
generating a plurality of two-dimensional (2D) slices from the 3D model (Coulombe, [0072], “One example method of dental file segmentation for creating a descriptor matrix of a tooth comprises slicing a three-dimension (3D) representation of the tooth into a number of two- dimension (2D) cross-sectional slices, and for each 2D cross-sectional slice determining an indexing centroid and a plurality of radial lengths measured from the slicing centroid to the cross-sectional boundary.”, [0029], “an indexed slicer for slicing the mesh file into a plurality of slices, each slice comprising a cross-sectional boundary of the dental object; a radial encoder assigning an indexing centroid and measuring a plurality of rays from the indexing centroid to the cross-sectional boundary”), the central axis passing through each of the plurality of 2D slices (Coulombe, [0089], “The present method can be achieved by mathematically slicing the three-dimensional image of tooth 200 into equally angled slicing planes by the indexed slider 104 as shown in image B. To do this the 3D image representation of the tooth 200 is passed through the indexed slicer 104 where, in this embodiment, the 3D digital representation of the tooth 200 is sliced radially. In a preferred embodiment with radial slicing, each slice will pass through the dental object common centroid on the z-axis 212, however it is noted that slicing can also be done in, for example, parallel or near parallel planes, or using angular indexing..”), and 2D points of the plurality of 2D slices corresponding to 3D points from the 3D model via a 2D point - 3D point correspondence (Coulombe, [0091], “Image C shows a 2D cross section of one slicing plane shown in image B. As shown in image C, the distance from the indexing centroid 214 to the intersection of each cross-sectional point or tooth cross sectional boundary 222 on the circumference of the tooth image slice, or the length of each indexing ray 216, is measured by the radial encoder. The indexing centroid 214 is the centroid of the cross-sectional plane of a single radial slicing plane generated from radial slicing of the tooth image through a radial slicing plane. The radial encoder will generate a plurality of indexing rays 216 originating at the indexing centroid 214, where the distance between the indexing centroid 214 and the circumference of the cross-section of the radial slicing plane at the edge of the dental object can be generated from the slicing plane is measured by the radial encoder 106 to map the circumference of the dental object in the slicing plane. In a radial slicing method, the slicing centroid 206 can also be at the same location as the indexing centroid of each cross-section generated from slicing plane 202a, 202b, 202c. The radial encoder 106 will generate indexing rays 216 from the indexing centroid 214 which maps the length of each indexing ray from the indexing centroid to the dental object cross-sectional boundary 222 in each radial slice”);
computing 3D information about the 3D model (Coulombe, [0093], “The visualization mapping and/or 2D matrix can then be converted back into a 3D representation of the tooth and displayed on a graphical user interface during the dental procedure.”)by:
proposing for each 2D slice of the plurality of 2D slices, using a trained machine learning (ML) model, 2D information about the 2D slice using the 2D slice as input, to obtain a plurality of 2D information for the plurality of 2D slices (Coulombe, [0088], “The indexed slicer 104 receives an STL or mesh file 102 as an input source and the Fourier neural operator 108 in the system 100 generates 2D descriptor matrix 208 output which can be visualized on a visualization unit 110.”, [0092], “Image D illustrates an example of a radial encoder output of the Fourier neural operator 108 as a 2D descriptor matrix 208 for the tooth shown in images A and B. A 2D descriptor matrix 208 is generated for each radial portion or slicing plane 202a, 202b, 202c of the tooth dental object shown in image B”), and
converting the plurality of 2D information to the 3D information about the 3D model based on the 2D point - 3D point correspondence (Coulombe, [0093], “Image E in FIG. 6 is an illustration of a visualization map 210 of the radial encoder output 2D matrix 208 visualized as a pixel array. The visualization map 210 of the radial encoder 106 output can be visualized on a graphical user interface, where each entry in the 2D matrix 208 is converted to a pixel value, referred to herein as a visualization map 210. Image E illustrates an example of a greyscale (black and white range) visualization mapping of matrix 208, however it is understood that the same can be displayed in color as desired. The visualization mapping and/or 2D matrix can then be converted back into a 3D representation of the tooth and displayed on a graphical user interface during the dental procedure.”).
Claim 2
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), wherein the 3D information about the 3D model comprises a portion of the 3D model that at least partially surrounds the central axis (Coulombe, [0093], “Image E in FIG. 6 is an illustration of a visualization map 210 of the radial encoder output 2D matrix 208 visualized as a pixel array. The visualization map 210 of the radial encoder 106 output can be visualized on a graphical user interface, where each entry in the 2D matrix 208 is converted to a pixel value, referred to herein as a visualization map 210. Image E illustrates an example of a greyscale (black and white range) visualization mapping of matrix 208, however it is understood that the same can be displayed in color as desired. The visualization mapping and/or 2D matrix can then be converted back into a 3D representation of the tooth and displayed on a graphical user interface during the dental procedure.”, [0108], “ Accordingly the z-axis can serve as a common reference locus for each slicing plane and there is no need to locate a single common centroid for the dental object as a whole.”).
Claim 3
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), wherein the 3D information about the 3D model comprises a portion of the 3D model that at least fully surrounds the central axis (Coulombe, [0093], “Image E in FIG. 6 is an illustration of a visualization map 210 of the radial encoder output 2D matrix 208 visualized as a pixel array. The visualization map 210 of the radial encoder 106 output can be visualized on a graphical user interface, where each entry in the 2D matrix 208 is converted to a pixel value, referred to herein as a visualization map 210. Image E illustrates an example of a greyscale (black and white range) visualization mapping of matrix 208, however it is understood that the same can be displayed in color as desired. The visualization mapping and/or 2D matrix can then be converted back into a 3D representation of the tooth and displayed on a graphical user interface during the dental procedure.”, [0108], “ Accordingly the z-axis can serve as a common reference locus for each slicing plane and there is no need to locate a single common centroid for the dental object as a whole.”).
Claim 4
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), wherein the 2D information about the 2D slice comprises a measurable property of the 2D slice ([0081], “The mesh file is then converted into one or more 2D descriptor matrixes which contains information on the exterior of the surface of the tooth, and also on other surfaces surrounding the tooth, such as the surfaces of adjacent teeth, the gumline, and occlusal surfaces on opposing or facing teeth.”).
Claim 5
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), wherein the trained ML model is first trained using a plurality of 2D training slices obtained from 3D training models as training inputs and a plurality of 2D training information about the 2D training slices as training outputs ([0099], “The dental descriptor database 124 can, for example, be a database comprising a plurality of dental descriptor mesh or mesh files and/or matrix data files that can be called upon for matching, comparison, diagnostic, artificial intelligence or machine learning training or testing sets, or for other comparative purposes”).
Claims 11-13 are rejected for similar reasons as those described in claims 1-3. The additional elements in Claims 11-13 (Coulombe) discloses includes: a computing apparatus (Coulombe, [0068], “n, the terms “component,” “system,” “platform,” “layer,” “controller,” “terminal,” “station,” “node,” “interface” are intended to refer to a computer-related entity or an entity related to, or that is part of, an operational apparatus with one or more specific functionalities”) comprising: a processor; and a memory storing instructions (Coulombe, [0068], “ a component can be, but is not limited to being, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical or magnetic storage medium) including affixed or removably affixed solid-state storage drives, an object, an executable, a computer-executable program, and/or a computer. By way of illustration, both an application running on a server and the server can be a component.”).
Claim 18 is rejected for similar reasons as those described in claim 1. The additional elements in Claim 18 (Coulombe) discloses includes: a non-transitory computer-readable storage medium including instruction (Coulombe, [0068], “ a component can be, but is not limited to being, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical or magnetic storage medium) including affixed or removably affixed solid-state storage drives, an object, an executable, a computer-executable program, and/or a computer. By way of illustration, both an application running on a server and the server can be a component.”).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 6-9, 14-17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Coulombe in view of Ezhov et al., (US 2021/0217170 A1), hereinafter referred to as Ezhov.
Claim 6
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”).
Coulombe discloses that the 2D descriptor matrixes contains information on the exterior of the surface of the tooth, and also on other surfaces surrounding the tooth, such as the surfaces of adjacent teeth, the gumline, and occlusal surfaces on opposing or facing teeth. However, Coulombe does not explicitly disclose further comprising locating a 3D cementa-enamel junction (CEJ) around a tooth by: proposing 2D locations of the CEJ by the trained ML model, and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence.
However, Ezhov teaches further comprising locating a 3D cementa-enamel junction (CEJ) around a tooth (Ezhov, [0080], “Diagnosis is performed by segmenting 1) tooth's body, 2) enamel, 3) alveolar bone (tooth's bony envelope), and then algorithmically measuring what part of a tooth between apex and CEJ is covered by the bone.”) by: proposing 2D locations of the CEJ (Ezhov, [0122], “step 1318, the identified landmark/features and surrounding context within the localized i/v.i are extracted; all p/v belonging to the localized landmark structure is selected by finding a minimal bounding rectangle around the p/v and the surrounding region for cropping as a defined landmark structure by the localization layer.”, [0124], “in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth, the distance measured between the two features above or below a threshold may determine periodontal bone loss”) by the trained ML model (Ezhov, [0036], “ localization is achieved using any one of fully convolutional network or plain classification convolutional neural network (FCN/CNN), such as a V-Net-based fully convolutional neural network. In one embodiment, the V-Net is a 3D generalization of UNeT”, [0035], “the input data is a 2-Dimensional (2D) image data.”, [0042], “the system could apply either segmentation or object detection in 2D, to segment axial slices”, [0039], “the system is configured to classify each p/v as one of 32 teeth or background and resulting segmentation assigns each p/v to one of 33 classes. In another embodiment, the system is configured to classify each p/v as either tooth or other anatomical structure of interest.”), and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence (Ezhov, [0042], “In another embodiment, the system could apply either segmentation or object detection in 2D, to segment axial slices. This would allow to process images in original resolution (albeit in 2D instead of 3D) and then infer 3D shape from 2D segmentation.”, [0075], “Using state-of-the-art 2D instance segmentation deep learning models (R-CNN detectors), 1) localize teeth in 2D bounding boxes; 2) assign a number to each detected tooth in accordance with the dental formula; and 3) provide accurate 2D masks to each detected tooth. To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator. With the assistance of the 3D panoramic ribbon, retrieve 3D bounding boxes, inferred from the 2D panoramic instances, defining correspondence between 2D and 3D bounding boxes coordinates. The 3D bounding boxes (as regions of interest) are further submitted to 3D segmentation module, to obtain accurate 3D teeth masks in the original CBCT fine scale.”).
Coulombe and Ezhov are both considered to be analogous to the claimed invention because they are in the same field of dental image analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Coulombe to incorporate the teachings of Ezhov of locating a 3D cementa-enamel junction (CEJ) around a tooth by: proposing 2D locations of the CEJ by the trained ML model, and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to help diagnose gum disease (Ezhov, [0080]).
Claim 7
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”).
Coulombe discloses that the 2D descriptor matrixes contains information on the exterior of the surface of the tooth, and also on other surfaces surrounding the tooth, such as the surfaces of adjacent teeth, the gumline, and occlusal surfaces on opposing or facing teeth. However, Coulombe does not explicitly disclose further comprising locating a 3D alveolar crest level (AC) around a tooth by: proposing 2D locations of the AC by the trained ML model, and converting the 2D locations of the AC of the plurality of slices are 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence.
However, Ezhov teaches further comprising locating a 3D alveolar crest level (AC) around a tooth (Ezhov, [0080], “Diagnosis is performed by segmenting 1) tooth's body, 2) enamel, 3) alveolar bone (tooth's bony envelope), and then algorithmically measuring what part of a tooth between apex and CEJ is covered by the bone.”) by: proposing 2D locations of the AC (Ezhov, [0122], “step 1318, the identified landmark/features and surrounding context within the localized i/v.i are extracted; all p/v belonging to the localized landmark structure is selected by finding a minimal bounding rectangle around the p/v and the surrounding region for cropping as a defined landmark structure by the localization layer.”, [0124], “in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth”, the bone attachment point is analogous to the alveolar bone) by the trained ML model (Ezhov, [0036], “ localization is achieved using any one of fully convolutional network or plain classification convolutional neural network (FCN/CNN), such as a V-Net-based fully convolutional neural network. In one embodiment, the V-Net is a 3D generalization of UNeT”, [0035], “the input data is a 2-Dimensional (2D) image data.”, [0042], “the system could apply either segmentation or object detection in 2D, to segment axial slices”, [0039], “the system is configured to classify each p/v as one of 32 teeth or background and resulting segmentation assigns each p/v to one of 33 classes. In another embodiment, the system is configured to classify each p/v as either tooth or other anatomical structure of interest.”), and converting the 2D locations of the AC of the plurality of slices are to 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence (Ezhov, [0042], “In another embodiment, the system could apply either segmentation or object detection in 2D, to segment axial slices. This would allow to process images in original resolution (albeit in 2D instead of 3D) and then infer 3D shape from 2D segmentation.”, [0075], “Using state-of-the-art 2D instance segmentation deep learning models (R-CNN detectors), 1) localize teeth in 2D bounding boxes; 2) assign a number to each detected tooth in accordance with the dental formula; and 3) provide accurate 2D masks to each detected tooth. To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator. With the assistance of the 3D panoramic ribbon, retrieve 3D bounding boxes, inferred from the 2D panoramic instances, defining correspondence between 2D and 3D bounding boxes coordinates. The 3D bounding boxes (as regions of interest) are further submitted to 3D segmentation module, to obtain accurate 3D teeth masks in the original CBCT fine scale.”).
Coulombe and Ezhov are both considered to be analogous to the claimed invention because they are in the same field of dental image analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Coulombe to incorporate the teachings of Ezhov of locating a 3D alveolar crest level (AC) around a tooth by: proposing 2D locations of the AC by the trained ML model, and converting the 2D locations of the AC of the plurality of slices are 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to help diagnose gum disease (Ezhov, [0080]).
Claim 8
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”).
Coulombe discloses that the 2D descriptor matrixes contains information on the exterior of the surface of the tooth, and also on other surfaces surrounding the tooth, such as the surfaces of adjacent teeth, the gumline, and occlusal surfaces on opposing or facing teeth. However, Coulombe does not explicitly disclose wherein a 3D periodontal bone loss (PBL) is generated by: detecting of a 3D cementa-enamel junction (CEJ) around a tooth by: proposing 2D locations of the CEJ by the trained ML model, and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence, detecting a 3D alveolar crest level (AC) around a tooth by: proposing 2D locations of the AC by the trained ML model, and converting the 2D locations of the AC of the plurality slices are 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence, and generating a difference between the 3D CEJ and the 3D AC.
However, Ezhov teaches wherein a 3D periodontal bone loss (PBL) (Ezhov, [0124], “The CEJ point and BAP of a tooth alongside 2, 4 or more sides of the tooth (i.e. buccal, lingual, mesial, distal, buccal-distal, buccal-mesial, or any kind of intermediate points) may be measured in terms of distance between points to define a BAP-CEJ value, which may be indicative of the following diagnosis: “No periodontal bone loss” if <1mm; “Mild periodontal bone loss” if 1 to 3 mm; “Moderate periodontal bone loss” if 3 to 5 mm; “Severe periodontal bone loss” if more than 5 mm. Actual ranges are extracted from literature and are subject to review/change/tuning. Furthermore, another landmark/landmark feature subject to localization/diagnostic may be cephalometric features on a maxillofacial region, whereby an angle measured between cephalometric features above or below a threshold indicate an orthodontic condition. The anthropomorphic diagnostic may be in lieu of, or in conjunction with, deep neural network applications to further validate a diagnosis—whether periodontal or orthodontic.”) is generated by: detecting of a 3D cementa-enamel junction (CEJ) (Ezhov, [0080], “Diagnosis is performed by segmenting 1) tooth's body, 2) enamel, 3) alveolar bone (tooth's bony envelope), and then algorithmically measuring what part of a tooth between apex and CEJ is covered by the bone.”) around a tooth by: proposing 2D locations of the CEJ (Ezhov, [0122], “step 1318, the identified landmark/features and surrounding context within the localized i/v.i are extracted; all p/v belonging to the localized landmark structure is selected by finding a minimal bounding rectangle around the p/v and the surrounding region for cropping as a defined landmark structure by the localization layer.”, [0124], “in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth, the distance measured between the two features above or below a threshold may determine periodontal bone loss”) by the trained ML model (Ezhov, [0036], “ localization is achieved using any one of fully convolutional network or plain classification convolutional neural network (FCN/CNN), such as a V-Net-based fully convolutional neural network. In one embodiment, the V-Net is a 3D generalization of UNeT”, [0035], “the input data is a 2-Dimensional (2D) image data.”, [0042], “the system could apply either segmentation or object detection in 2D, to segment axial slices”, [0039], “the system is configured to classify each p/v as one of 32 teeth or background and resulting segmentation assigns each p/v to one of 33 classes. In another embodiment, the system is configured to classify each p/v as either tooth or other anatomical structure of interest.”), and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence (Ezhov, [0042], “In another embodiment, the system could apply either segmentation or object detection in 2D, to segment axial slices. This would allow to process images in original resolution (albeit in 2D instead of 3D) and then infer 3D shape from 2D segmentation.”, [0075], “Using state-of-the-art 2D instance segmentation deep learning models (R-CNN detectors), 1) localize teeth in 2D bounding boxes; 2) assign a number to each detected tooth in accordance with the dental formula; and 3) provide accurate 2D masks to each detected tooth. To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator. With the assistance of the 3D panoramic ribbon, retrieve 3D bounding boxes, inferred from the 2D panoramic instances, defining correspondence between 2D and 3D bounding boxes coordinates. The 3D bounding boxes (as regions of interest) are further submitted to 3D segmentation module, to obtain accurate 3D teeth masks in the original CBCT fine scale.”), detecting a 3D alveolar crest level (AC) (Ezhov, [0080], “Diagnosis is performed by segmenting 1) tooth's body, 2) enamel, 3) alveolar bone (tooth's bony envelope), and then algorithmically measuring what part of a tooth between apex and CEJ is covered by the bone.”) by: proposing 2D locations of the AC (Ezhov, [0122], “step 1318, the identified landmark/features and surrounding context within the localized i/v.i are extracted; all p/v belonging to the localized landmark structure is selected by finding a minimal bounding rectangle around the p/v and the surrounding region for cropping as a defined landmark structure by the localization layer.”, [0124], “in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth”, the bone attachment point is analogous to the alveolar bone) by the trained ML model (Ezhov, [0036], “ localization is achieved using any one of fully convolutional network or plain classification convolutional neural network (FCN/CNN), such as a V-Net-based fully convolutional neural network. In one embodiment, the V-Net is a 3D generalization of UNeT”, [0035], “the input data is a 2-Dimensional (2D) image data.”, [0042], “the system could apply either segmentation or object detection in 2D, to segment axial slices”, [0039], “the system is configured to classify each p/v as one of 32 teeth or background and resulting segmentation assigns each p/v to one of 33 classes. In another embodiment, the system is configured to classify each p/v as either tooth or other anatomical structure of interest.”), and converting the 2D locations of the AC of the plurality of slices are to 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence (Ezhov, [0042], “In another embodiment, the system could apply either segmentation or object detection in 2D, to segment axial slices. This would allow to process images in original resolution (albeit in 2D instead of 3D) and then infer 3D shape from 2D segmentation.”, [0075], “Using state-of-the-art 2D instance segmentation deep learning models (R-CNN detectors), 1) localize teeth in 2D bounding boxes; 2) assign a number to each detected tooth in accordance with the dental formula; and 3) provide accurate 2D masks to each detected tooth. To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator. With the assistance of the 3D panoramic ribbon, retrieve 3D bounding boxes, inferred from the 2D panoramic instances, defining correspondence between 2D and 3D bounding boxes coordinates. The 3D bounding boxes (as regions of interest) are further submitted to 3D segmentation module, to obtain accurate 3D teeth masks in the original CBCT fine scale.”), and generating a difference between the 3D CEJ and the 3D AC (Ezhov, [0124], “For instance, in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth, the distance measured between the two features above or below a threshold may determine periodontal bone loss. The CEJ point and BAP of a tooth alongside 2, 4 or more sides of the tooth (i.e. buccal, lingual, mesial, distal, buccal-distal, buccal-mesial, or any kind of intermediate points) may be measured in terms of distance between points to define a BAP-CEJ value, which may be indicative of the following diagnosis: “No periodontal bone loss” if <1mm; “Mild periodontal bone loss” if 1 to 3 mm; “Moderate periodontal bone loss” if 3 to 5 mm; “Severe periodontal bone loss” if more than 5 mm.”).
Coulombe and Ezhov are both considered to be analogous to the claimed invention because they are in the same field of dental image analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Coulombe to incorporate the teachings of Ezhov wherein a 3D periodontal bone loss (PBL) is generated by: detecting of a 3D cementa-enamel junction (CEJ) around a tooth by: proposing 2D locations of the CEJ by the trained ML model, and converting the 2D locations of the CEJ of the plurality of slices are to 3D locations of the CEJ around the central axis of the tooth using the 2D point - 3D point correspondence, detecting a 3D alveolar crest level (AC) around a tooth by: proposing 2D locations of the AC by the trained ML model, and converting the 2D locations of the AC of the plurality slices are 3D locations of the AC around the central axis of the tooth using the 2D point - 3D point correspondence, and generating a difference between the 3D CEJ and the 3D AC. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to help diagnose gum disease (Ezhov, [0080]).
Claim 9
The combination of Coulombe in view of Ezhov discloses the method of claim 8 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), wherein a generation of the PBL further comprises computing for each pair of 2D locations of the CEJ and AC on a side of the tooth a distance between the pair in relation to the distance from the 2D CEJ location to a root tip of the tooth (Ezhov, [0124], “For instance, in the event of localizing the landmark or landmark features of a cement-enamel junction (CEJ) and bone attachment points (BAP) of a tooth, the distance measured between the two features above or below a threshold may determine periodontal bone loss. The CEJ point and BAP of a tooth alongside 2, 4 or more sides of the tooth (i.e. buccal, lingual, mesial, distal, buccal-distal, buccal-mesial, or any kind of intermediate points) may be measured in terms of distance between points to define a BAP-CEJ value, which may be indicative of the following diagnosis: “No periodontal bone loss” if <1mm; “Mild periodontal bone loss” if 1 to 3 mm; “Moderate periodontal bone loss” if 3 to 5 mm; “Severe periodontal bone loss” if more than 5 mm.”), and using the highest relative distance from the results as an indication of a maximum PBL (Ezhov, [0124], “Severe periodontal bone loss” if more than 5 mm”). The proposed combination as well as the motivation for combining the Coulombe and Ezhov references presented in the rejection of Claim 8, apply to Claim 9 and are incorporated herein by reference. Thus, the method recited in Claim 8 is met by Coulombe and Ezhov.
Claims 14-17 are rejected for similar reasons as those described in claims 6-9. The additional elements in Claims 14-17 (Coulombe and Ezhov) discloses includes: a computing apparatus (Coulombe, [0068], “n, the terms “component,” “system,” “platform,” “layer,” “controller,” “terminal,” “station,” “node,” “interface” are intended to refer to a computer-related entity or an entity related to, or that is part of, an operational apparatus with one or more specific functionalities”) comprising: a processor; and a memory storing instructions (Coulombe, [0068], “ a component can be, but is not limited to being, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical or magnetic storage medium) including affixed or removably affixed solid-state storage drives, an object, an executable, a computer-executable program, and/or a computer. By way of illustration, both an application running on a server and the server can be a component.”). The proposed combination as well as the motivation for combining the Coulombe and Ezhov references presented in the rejection of Claims 6-9, apply to Claims 14-17 and are incorporated herein by reference. Thus, the apparatus recited in Claims 14-17 is met by Coulombe and Ezhov.
Claim 19 is rejected for similar reasons as those described in claim 8. The additional elements in Claim 19 (Coulombe and Ezhov) discloses includes: a non-transitory computer-readable storage medium including instruction (Coulombe, [0068], “ a component can be, but is not limited to being, a process running on a processor, a processor, a hard disk drive, multiple storage drives (of optical or magnetic storage medium) including affixed or removably affixed solid-state storage drives, an object, an executable, a computer-executable program, and/or a computer. By way of illustration, both an application running on a server and the server can be a component.”). The proposed combination as well as the motivation for combining the Coulombe and Ezhov references presented in the rejection of Claim 8, apply to Claim 19 and are incorporated herein by reference. Thus, the medium recited in Claim 19 is met by Coulombe and Ezhov.
Claim 20
Coulombe discloses a method (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”), comprising:
computing a central axis for a three-dimensional (3D) model (Coulombe, [0120], “Assigning and aligning the z-axis of each tooth or crown to a central axis and using the same dental file segmentation algorithm along the same z-axis with the same common centroid for all descriptor matrixes enables matching with other crown and tooth descriptor matrixes with the same reference loci”, [0010], “scanning a tooth with a scanner to obtain a three dimensional (3D) image of the tooth”)
generating one or more two-dimensional (2D) slices from the 3D model (Coulombe, [0072], “One example method of dental file segmentation for creating a descriptor matrix of a tooth comprises slicing a three-dimension (3D) representation of the tooth into a number of two- dimension (2D) cross-sectional slices, and for each 2D cross-sectional slice determining an indexing centroid and a plurality of radial lengths measured from the slicing centroid to the cross-sectional boundary.”, [0029], “an indexed slicer for slicing the mesh file into a plurality of slices, each slice comprising a cross-sectional boundary of the dental object; a radial encoder assigning an indexing centroid and measuring a plurality of rays from the indexing centroid to the cross-sectional boundary”), the central axis passing through each of the plurality of 2D slices (Coulombe, [0089], “The present method can be achieved by mathematically slicing the three-dimensional image of tooth 200 into equally angled slicing planes by the indexed slider 104 as shown in image B. To do this the 3D image representation of the tooth 200 is passed through the indexed slicer 104 where, in this embodiment, the 3D digital representation of the tooth 200 is sliced radially. In a preferred embodiment with radial slicing, each slice will pass through the dental object common centroid on the z-axis 212, however it is noted that slicing can also be done in, for example, parallel or near parallel planes, or using angular indexing..”), and 2D points of the plurality of 2D slices corresponding to 3D points from the 3D model via a 2D point - 3D point correspondence (Coulombe, [0091], “Image C shows a 2D cross section of one slicing plane shown in image B. As shown in image C, the distance from the indexing centroid 214 to the intersection of each cross-sectional point or tooth cross sectional boundary 222 on the circumference of the tooth image slice, or the length of each indexing ray 216, is measured by the radial encoder. The indexing centroid 214 is the centroid of the cross-sectional plane of a single radial slicing plane generated from radial slicing of the tooth image through a radial slicing plane. The radial encoder will generate a plurality of indexing rays 216 originating at the indexing centroid 214, where the distance between the indexing centroid 214 and the circumference of the cross-section of the radial slicing plane at the edge of the dental object can be generated from the slicing plane is measured by the radial encoder 106 to map the circumference of the dental object in the slicing plane. In a radial slicing method, the slicing centroid 206 can also be at the same location as the indexing centroid of each cross-section generated from slicing plane 202a, 202b, 202c. The radial encoder 106 will generate indexing rays 216 from the indexing centroid 214 which maps the length of each indexing ray from the indexing centroid to the dental object cross-sectional boundary 222 in each radial slice”).
Coulombe does not explicitly disclose obtaining labels of 3D information about the 3D model by: identifying 2D information and labelling the identified 2D information for each 2D slice, and converting the plurality of labelled 2D information to labelled 3D information about the 3D model based on the 2D point - 3D point correspondence, and training an ML model using the 3D information about the 3D model.
However, Ezhov teaches obtaining labels of 3D information about the 3D model (Ezhov, [0036], “ localization is achieved using any one of fully convolutional network or plain classification convolutional neural network (FCN/CNN), such as a V-Net-based fully convolutional neural network. In one embodiment, the V-Net is a 3D generalization of UNeT”, [0035], “the input data is a 2-Dimensional (2D) image data.”, [0042], “the system could apply either segmentation or object detection in 2D, to segment axial slices”, [0039], “the system is configured to classify each p/v as one of 32 teeth or background and resulting segmentation assigns each p/v to one of 33 classes. In another embodiment, the system is configured to classify each p/v as either tooth or other anatomical structure of interest.”) by: identifying 2D information and labelling the identified 2D information for each 2D slice (Ezhov, [0041], “the system provides 2-class segmentation, which includes labelling or classification, if the localization comprises tooth or not”), and converting the plurality of labelled 2D information to labelled 3D information about the 3D model based on the 2D point - 3D point correspondence (Ezhov, [0075], “a segmentation module 1008e (optionally, an instance segmentation module), operating over 2D panoramic image plane, provides accurate teeth segmentation masks and numbering to a corresponding 3D CBCT image. The initial step is to segment teeth instances on an automatically generated panoramic image. Using state-of-the-art 2D instance segmentation deep learning models (R-CNN detectors), 1) localize teeth in 2D bounding boxes; 2) assign a number to each detected tooth in accordance with the dental formula; and 3) provide accurate 2D masks to each detected tooth. To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator. With the assistance of the 3D panoramic ribbon, retrieve 3D bounding boxes, inferred from the 2D panoramic instances, defining correspondence between 2D and 3D bounding boxes coordinates. The 3D bounding boxes (as regions of interest) are further submitted to 3D segmentation module, to obtain accurate 3D teeth masks in the original CBCT fine scale.”), and training an ML model using the 3D information about the 3D model (Ezhov, [0075], “2D instance segmentation deep learning models (R-CNN detectors)”, “To train 2D instance segmentation module, utilize the mixture of the annotated OPT and panoramic images generated from CBCT, obtained as an output of our automatic panoramic generator.”).
Coulombe and Ezhov are both considered to be analogous to the claimed invention because they are in the same field of dental image analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Coulombe to incorporate the teachings of Ezhov obtaining labels of 3D information about the 3D model by: identifying 2D information and labelling the identified 2D information for each 2D slice, and converting the plurality of labelled 2D information to labelled 3D information about the 3D model based on the 2D point - 3D point correspondence, and training an ML model using the 3D information about the 3D model. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to obtain accurate 3D teeth masks in the original CBCT fine scale (Ezhov, [0075]).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Coulombe in view of Yang et al., (US 2025/0143852 A1, earliest filing date of foreign application is 11/08/2023), hereinafter referred to as Yang.
Claim 10
Coulombe discloses the method of claim 1 (Coulombe, [0002], “An object of the present invention is to provide a system and method for dental imaging and guided surgery technology using augmented intelligence in dental pattern recognition using specially formatted stacked data arrays”).
Coulombe does not explicitly disclose wherein the plurality of 2D information are converted to the 3D information by interpolating between adjacent 2D slices.
However, Yang teaches wherein the plurality of 2D information are converted to the 3D information by interpolating between adjacent 2D slices (Yang, [0145], “perform 3D transformation on multiple 2D oral images to obtain an initial 3D oral image. This can be achieved by using computer vision algorithms to locate and match feature points for each 2D oral image, and using 3D reconstruction methods in computer vision to restore depth information and shape from the images, transforming multiple 2D oral images into the initial 3D oral image; Resampling multiple voxels in the initial 3D oral image to the same voxel spacing to obtain a 3D oral image can be achieved by processing the initial 3D oral image and resampling its voxels to the same voxel spacing using interpolation or resampling algorithms, ensuring that the voxels in the 3D oral image have consistent spacing and resolution.”).
Coulombe and Yang are both considered to be analogous to the claimed invention because they are in the same field of dental image analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Coulombe to incorporate the teachings of Yang wherein the plurality of 2D information are converted to the 3D information by interpolating between adjacent 2D slices. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to ensure that the voxels in the 3D oral image have consistent spacing and resolution (Yang, [0145]).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DENISE G ALFONSO whose telephone number is (571)272-1360. The examiner can normally be reached Monday - Friday 7:30 - 5:30.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Amandeep Saini can be reached at (571)272-3382. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/DENISE G ALFONSO/Examiner, Art Unit 2662
/AMANDEEP SAINI/Supervisory Patent Examiner, Art Unit 2662
Prosecution Insights