DETAILED ACTION
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 .
Response to Arguments
Applicant's arguments filed 06/04/2026 have been fully considered but they are not persuasive.
Rejections under 35 USC 101 have been withdrawn.
Applicant argues, see Applicant’s arguments pages 8-9, that the motivation for combining Hirakawa and Nicolaes is insufficient. Examiner disagrees with Applicant’s assertion that “identifying a fracture in at least one of the VOIs” is the motivation for combining the references; this is the limitation from Nicolaes being combined with Hirakawa. This motivation for combining the references, as stated in the rejection of claim 1, is that this step “provides a functional use to the steps of obtaining and segmenting images of the spine”. The steps taught by Hirakawa amount to imaging and segmenting a spine. By themselves, these steps provide very little practicality; however, these steps are crucial when performed prior to a medical procedure involving the spine to determine the location and severity of an injury and the type and extent of intervention. As such, Examiner asserts that one having ordinary skill in the art would understand that the steps of Hirakawa can be used for the specific application of identifying a spinal fracture, as is taught by Nicolaes. Although Examiner upholds the motivation statement for combining Hirakawa and Nicolaes as set forth in the previous Office Action, the motivation statement has been updated for the sake of clarity of record and compact prosecution.
Applicant argues, see Applicant’s arguments pages 9-11, that the combination of Hirakawa and Nicolaes is improper. As is stated above, the steps taught by Hirakawa lack practical application in the medical field. However, one having ordinary skill in the art would understand that a segmented image of a patient’s spine is useful and necessary to be obtained when locating the presence and locations of spinal fractures. Whether Hirakawa further teaches that the spine image is used in a longitudinal study is irrelevant; only the claimed steps of the instant application are being cited as being read upon by Hirakawa, and these steps form the basis of the combination with Nicolaes. Examiner upholds the combination of Hirakawa and Nicolaes.
Claim Rejections - 35 USC § 101
Rejections under 35 USC 101 have been withdrawn in response to Applicant’s arguments filed 06/04/2026.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claims 1-15 are rejected under 35 U.S.C. 103 as being unpatentable over Hirakawa (US 2020/0058098) in view of Nicolaes (US 2020/0364856).
Regarding claim 1, Hirakawa teaches a computer-implemented method for detecting spinal fractures, comprising:
receiving three-dimensional image data including at least a portion of a spine of a subject ([0037]);
identifying the spine of the subject in the three-dimensional image data ([0037]);
defining a spline (spline interpolation, [0060]) approximating a local curvature along the spine of the subject ([0060]); and
defining multiple volumes of interest (VOIs) (vertebra regions VR, [0058]), each VOI containing at least a portion of a vertebra of the spine of the subject ([0058] & Figure 6), wherein each VOI is defined relative to an adjacent segment of the spline ([0060]).
However, Hirakawa fails to disclose identifying a fracture in at least one of the VOIs.
Nicolaes teaches identifying a fracture ([0040] & [0059]) in at least one of the VOIs (vertebra voxel, [0059]).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa to include identifying a fracture in at least one of the VOIs, as taught by Nicolaes. This provides a functional use of determining the presence, location, and severity of a spinal fracture in order to inform the type and extent of a medical intervention to the steps of obtaining and segmenting images of the spine.
Regarding claim 2, Hirakawa in view of Nicolaes teach the method of claim 1, and Nicolaes further teaches that the spine is identified by a convolutional neural network (CNN) ([0051]) trained for segmenting the spine ([0051] & [0059]).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa such that the spine is identified by a convolutional neural network (CNN) trained for segmenting the spine. Convolutional neural networks are well-known in the art to be applied to imaging-related problems; per [0051] of Nicolaes, “CNNs have been applied successfully to classification, object detection and segmentation tasks”.
Regarding claim 3, Hirakawa in view of Nicolaes teach the method of claim 1, and Hirakawa further teaches that for each VOI a corresponding center point (middle point P3, [0059], Figure 7) is sampled at a location defined relative to the spline ([0060]), wherein the center points of the VOIs are sampled at regular intervals along the spline ([0059]).
Because each vertebra lies on the spline, the center points are sampled at a location defined relative to the spline. Additionally, sampling the center point at each vertebra constitutes a regular interval.
Regarding claim 4, Hirakawa in view of Nicolaes teach the method of claim 3, and Hirakawa further teaches that center points for adjacent VOIs are located so as to generate overlapping VOIs ([0059] & Figure 7).
As is detailed in [0059] and illustrated in Figure 7, the setting of the middle point P3 in each vertebra region VR is based on intersection points P1 and P2 in the intervertebral discs D both above and below each vertebral body C. Thus, each intervertebral disc belongs in the vertebra regions of the vertebra both above and below it, and each vertebra region overlaps in the discs.
Regarding claim 5, Hirakawa in view of Nicolaes teach the method of claim 3, and Hirakawa further teaches that each VOI is formed about the center point ([0059] & Figure 7) and is oriented based on a tangent of the spline adjacent the corresponding center point ([0060]).
Regarding claim 6, Hirakawa in view of Nicolaes teach the method of claim 5, and Hirakawa further teaches that after defining each VOI, each VOI is extracted ([0062]) and resampled from the three-dimensional image data to a target resolution ([0050]-[0051]).
Regarding claim 7, Hirakawa in view of Nicolaes teach the method of claim 1, and Nicolaes further teaches that the fracture is identified by applying a convolutional neural network (CNN) ([0051]) to each VOI ([0059]).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa such that the fracture is identified by applying a convolutional neural network (CNN) to each VOI, as taught by Nicolaes. Convolutional neural networks are well-known in the art to be applied to imaging-related problems; per [0051] of Nicolaes, “CNNs have been applied successfully to classification, object detection and segmentation tasks”.
Regarding claim 8, Hirakawa in view of Nicolaes teach the method of claim 1, and Nicolaes further teaches that the output of the CNN, when applied to each VOI, is a probability map (classification layer, [0059]) identifying likely fractures within the corresponding VOI ([0059]-[0060]).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa such that the output of the CNN, when applied to each VOI, is a probability map identifying likely fractures within the corresponding VOI, as taught by Nicolaes. This allows an operator to assess both the likelihood that a fracture is present in a patient, but also the location of an expected fracture.
Regarding claim 9, Hirakawa in view of Nicolaes teach the method of claim 8, and Hirakawa further teaches that the VOIs are defined such that adjacent VOIs overlap ([0059] & Figure 7).
Nicolaes further teaches that all predictions corresponding to a particular location are aggregated into a final probability map ([0076] & [0105]; 1102, Figure 11; Figure 14).
Regarding the limitation that multiple predictions are generated for the overlapping voxels, [0079] of Nicolaes states that a centroid is located for each vertebra and the fracture probability is considered for each vertebra. When combining the teachings of Nicolaes with those of Hirakawa, which teach segmenting each vertebra region with overlapping components (the intervertebral discs), the overlapping voxels would be considered for multiple vertebrae when assessing each vertebra’s probability for a fracture.
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa to include generating multiple predictions for at least some equivalent voxels occurring in the multiple VOIs, wherein all predictions corresponding to a particular location are aggregated into a final probability map, as taught by Nicolaes. This increases the accuracy of fracture determination, and the final probability map allows the operator to visually understand where a potential fracture may be located on the spine.
Regarding claim 10, Hirakawa in view of Nicolaes teach the method of claim 8, and Nicolaes further teaches generating a final probability map from the probability maps associated with individual VOIs ([0076] & [0105]; 1102, Figure 11; Figure 14), the final probability map comprising a collation of the VOI probability maps into a coherent representation of the three-dimensional image data (Figure 14).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa to include generating a final probability map from the probability maps associated with individual VOIs, the final probability map comprising a collation of the VOI probability maps into a coherent representation of the three-dimensional image data, as taught by Nicolaes. Providing a visual representation of the fracture probability determination allows the operator to conceptually understand each patient’s condition.
Regarding claim 11, Hirakawa in view of Nicolaes teach the method of claim 10, and Nicolaes further teaches generating binary predictions based on the final probability map ([0059]; Figure 14) and filtering fracture candidates based on a relationship between a candidate location and the spine of the subject.
Paragraph [0059] teaches that each voxel is classified into one of three class probabilities: background, normal vertebra, or fractured vertebra. Because the background voxels do not correlate to a vertebra, there are two classes a vertebra voxel can be classified as. Thus, this is a binary prediction. Nevertheless, Figure 14 shows four classes: normal, mild, moderate, severe. Although there are four classes, they can be grouped as a binary prediction if so desired, in which “normal” constitutes a normal vertebra, and “mild”, “moderate”, and “severe” constitute a fractured vertebra. Regarding the filtering of the fracture candidates, the operator can mentally separate the normal vertebrae from the fractured vertebrae.
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa to include generating binary predictions based on the final probability map and filtering fracture candidates based on a relationship between a candidate location and the spine of the subject, as taught by Nicolaes. A binary prediction of normal and fractured vertebra allows the operator to understand the necessity of intervention, as a fractured vertebra requires intervention regardless of the level of severity.
Regarding claim 12, Hirakawa in view of Nicolaes teach the method of claim 1, and Hirakawa further teaches that the spline approximates a centerline (center line CL2, [0059], Figure 7) of a spinal canal for the spine ([0059]-[0060] & Figure 7).
Regarding claim 13, Hirakawa in view of Nicolaes teach the method of claim 1, and Hirakawa further teaches that a size for a first VOI is selected based on an adjacent first location along the spine, and wherein a size for a second VOI is selected based on an adjacent second location along the spine ([0059]).
Because each vertebra region VR is associated with an individual vertebra, and each vertebra has a different size, the sizes of the VOIs are based on the size of the vertebrae, which is a function of where along the spine they are located.
Regarding claim 14, Hirakawa in view of Nicolaes teach the method of claim 1, and Nicolaes further teaches locating fractures identified in a representation of the three-dimensional image data (Figure 14) and displaying identified fractures in the context of the spine of the subject (Figure 14).
It would have been obvious to one having ordinary skill in the art prior to the effective filing date of the claimed invention to have modified the method of Hirakawa to include locating fractures identified in a representation of the three-dimensional image data and displaying identified fractures in the context of the spine of the subject, as taught by Nicolaes. This allows the operator to observe the results of the fracture determination performed by the CNN.
Claim 15 is rejected for similar reasons to claim 1. Hirakawa further teaches a memory (memory 12, [0040]) that stores a plurality of instructions (image processing program, [0042]); and a processor (CPU 11, [0040]) that couples to the memory ([0042]-[0043]) and is configured to execute the plurality of instructions ([0043]).
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 ADAM KOLKIN whose telephone number is (571)272-5480. The examiner can normally be reached Monday-Friday 1:00PM-10:00PM EDT.
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, Keith Raymond can be reached at (572)-270-1790. 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.
/ADAM D. KOLKIN/Examiner, Art Unit 3798
/KEITH RAYMOND/Supervisory Patent Examiner, Art Unit 3798
Prosecution Insights