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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 05/13/2025 has been entered.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-11 and 18-20, 23, and 25 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Regarding claims 1, 18 and 25, the claim recites “generate a synthetic computed-tomography (CT) image the patient based on applying one or more learning algorithms to the surface image of the patient and anatomical structures generated from a collection of body scans, wherein the synthetic CT image comprises a plurality of voxels, each voxel providing an estimate of expected material density for the patient”. Regarding claim 25, the claim recites “generating a synthetic computed-tomography (CT) image of the patient based on the surface image of the patient and anatomical structures generated from a collection of body scans, wherein the synthetic CT image comprises a plurality of voxels, each voxel providing an estimate of expected material density for the patient”. However, the originally filed disclosure does not set forth how one would accomplish creating a synthetic CT of a patient with only a surface image and collected body scans (e.g. estimated density) where a surface image (which would have no information on organs and just surface information like clothing, skin, hair) and the estimated density/collection of scans (which appears from the specification just to be weighting volumes of organs that are unknown in the surface image) to show the interior tissues and organs of a body. The foregoing being true regardless of whether or not the surface image is processed by learning algorithm(s) which will be addressed separately below since it does not apply to all claims.
The inventor’s specification supplies two examples of equations to denote how a surface image becomes a synthetic CT. CTsynth = Wskin * Vskin + Wiungs * Vlungs + wpelvis * Vpelvis + Wskull * Vskull ... (t); which is said to be weights and “V” which appears to be masks; where it is unclear what the Vskin, Vlungs, Vpelvis and Vskull are representing from either the surface image or the estimated density and how they would be found from a surface image. The second equation is A * [Body Surface Volume] = [Volume with Anatomical Structure]. However, there is no explanation on what type of regression A is and simply states that the matrix of the body surface volume would be 0s and 1s which would imply only two image possibilities. The only other incite is through a reference to US 9525582 which uses a surface image to create an avatar where the avatar is still a mesh and not representing internal organs or tissues of a human. Furthermore, the drawings only show the image of the synthetic CT and no image of the surface image. Applicant’s own figures even show the synthetic CT has spine, heart, liver, stomach and kidneys, lungs and other internal tissues. And the question is how does one go from a surface image of skin to an internal image of organs.
PNG
media_image1.png
465
747
media_image1.png
Greyscale
Synthetic CT has a well-known definition in the art. The synthetic CT image providing an estimated measure of attenuation and providing a surface depth model of the patient could be found in a known synthetic CT. As these are not definitions just information provided within the synthetic image. As seen above, the synthetic CT of the originally filed disclosure does not appear to just show measure of attenuation or a surface depth model. Synthetic CT images are known to be made from MRI T1 and T2 images. When reviewing the specification, paragraph [0022] does point out a model also being used to help create this synthetic CT. However, further review does not give enough evidence or direction to understand how a model and surface image can be used to create a synthetic CT (or image of the interior of a subject based on intensity values, Hounsfield units that result in internal tissue). The specification rather just states the term “model” which is not enough information to show possession as this would be a specific algorithm or equations to achieve the outcome as claimed.
Applicant’s figures illustrate that the synthetic CT shows organ locations and internal tissue as seen in Fig. 5 reproduced below element 250.
PNG
media_image1.png
465
747
media_image1.png
Greyscale
.
The following are examples of a CT without contrast and a synthetic CT show how they appear to be very similar in nature and do show internal tissue and representations of organs.
Example of CT without contrast of lungs in a sagittal view*
PNG
media_image2.png
327
604
media_image2.png
Greyscale
* Bhalla AS, Das A, Naranje P, Irodi A, Raj V, Goyal A. Imaging protocols for CT chest: A recommendation. Indian J Radiol Imaging. 2019; 29(3):236-246. doi:10.4103/ijri.IJRI_34_19.
Example of synthetic CT from MRI* See below:
PNG
media_image3.png
598
355
media_image3.png
Greyscale
* Siemens, “MR-only RT planning for the brain and pelvis with Synthetic CT”.
The specification also points to US Patent 9524582 as how a synthetic image is created using a model as well as a surface image. This Patent, however, is silent on getting estimated attenuation and a surface depth model. This patent takes a surface image to create a surface mesh to help plan a medical image. The patent uses a point cloud to then develop an avatar mesh (see Fig. 7 and Fig. 1); however, all the information appears to be developing information on a hollow (3D) mesh. The patent does not state how the model and surface image can be used to create a synthetic CT image that also provides an estimated measure of attenuation and provides a surface depth model. The article (Shu-Hui Hsu et al., “Investigation of a method for generating synthetic CT models from MRI scans of the head and neck for radiation therapy” 2013 Phys. Med. Biol. 58 8419) shows how complicated an algorithm and equations are to develop a synthetic CT even from MRI. See below reproduced figures and paragraphs from Shu-Hui Hsu et al.
PNG
media_image4.png
205
604
media_image4.png
Greyscale
PNG
media_image5.png
223
564
media_image5.png
Greyscale
PNG
media_image6.png
368
626
media_image6.png
Greyscale
The detail for which is presented in the equations and explications on how an MR can be taken to CT illustrates how detailed information on creating a synthetic CT is presented and how detailed one would need to be in order to show possession at the time of filing.
Siemen’s own white paper illustrates the complexity for which Synthetic CT images are produced and complexity of the algorithm to even go from an MRI which shows internal tissue to synthetic CT.
PNG
media_image7.png
300
698
media_image7.png
Greyscale
Because the originally filed disclosure is seen as broad, while the intent of the invention specific, the claims/application are not seen as adhering to the written description requirement. The question still remains on how one would go from a surface image to creating or generating a synthetic CT. There is not enough direction, skill in the art, nor predictability in the art for one to make or use the invention as it is unknown how a surface image can be used to create a synthetic CT image.
In closing, the specification appears to be lacking in description for the equations and explanations needed to create a synthetic CT in a way to show that the applicant possessed the claimed invention at the time of filing. No paragraphs in the specification state how this was accomplished. The specification [0022]-[0025] appear to be putting forth the reasoning that a surface image can be used to create a synthetic CT. However, no details in those paragraphs show how this is accomplished. The two equations are examples. Also, the equations state elements but do not define how the elements were created. For example, the regression matrix A is not described to be anything but a regression matrix; begging the question of what regression was used to accomplish said conversion. Also, it is unclear how the surface image volume would ever have a 1 value in it to represent internal tissue when all the image contains is information of outside surfaces and thus would only be a matrix of zeros. The other equations states that weights (which can be approximations of houndsfield units) can be a representation of brightness and can be added up after being multiplied with some type of vector. It is not described or explained how the V values were obtained in order to use them with the weighted values. There is not enough explanation in the specification to show that applicant had possession of the synthetic CT from the surface image of a camera at the time of filing.
Claims 2-11 and 19-21 and 23 are each similarly affected, at least by virtue of dependency.
Additionally, in the foregoing the examiner mentioned in the foregoing rejection that the use of “one or more learning algorithms” would be addressed separately because it did not apply to all claims. However, for claims 1 and 18 the claim requires that surface image data is applied to “one or more learning algorithms” which raises a second separate issue of lacking adequate written description under this title. Specifically, the applicant additionally does not disclose these learning algorithms and therefore there is a profound prima facie case that there is inadequate written disclosure to allow one of ordinary skill in the art to make or use an invention commensurate with the scope of the claims.
In more detail “learning” and its derivatives appear only twice in the applicant’s specification. First at [0022] which states: “In some embodiments, the synthetic CT image 250 is generated by employing a model based approach to fit a depth image data of a person. The model based approach can include applying one or more learning algorithms including a statistical correlation model between the detailed patient surface geometry (from the surface image) and the anatomical structures (generated from a collection of real and/or synthetic body scans).” This is the applicant’s only relevant disclosure and even then, the statement only provides the information that “a statistical correlation model” (a model which is itself undisclosed as there are no details as to what this model is or how it is formed) can be included in the learning algorithm (e.g. the algorithm can comprise, but clearly does not consist of this model, as correlation is generally not considered learning) without providing anything else for the reader to go on. This alone is not adequate written description to allow one of ordinary skill in the art to discern what the applicant had in their possession at the time of filing; and moreover, it certainly does not provide enough written description to allow one of ordinary skill in the art prior to the date of invention to make or use an invention commensurate with the scope of the claims. The second mention of learning is not relevant to the claim limitation as it relates to the sanity checking of/validation of an unclaimed mask generation step. Not only does this not address the claimed steps but it occurs after the step and does not form anything and therefore does not even attempt to describe how a learning algorithm and a collection of scans could be used in the actual formation of the SCT image. Regardless and for the sake of thoroughness, the examiner notes that [0026] does mention learning and states: “In some embodiments, a non-linear projection manifold learning technique can be applied to ensure the generated masks are physically plausible.” Notably this again does not give enough information on the learning technique to describe what sort of algorism the applicant had in their possession at the time of filing nor to allow one of ordinary skill in the art to make or use an invention commensurate with the claims even if it was relevant to the SCT formation which it appears not to be. There are no other mentions of related terminology that addressed the claimed scope (e.g. “training” is not present, “trained” only occurs in [0031] and then only to describe the human reviewer, etc.) in the specification itself. Likewise, the examiner has already addressed the incorporated documents in the foregoing rejection since none of the incorporated document adequately describe forming a synthetic CT image in the first place, but for the sake of thoroughness again notes that the incorporated documents do not disclose the claimed subject matter. For example, U.S. Pat. No. 9,524,582 is incorporated by the applicant and describes using learning to determine poses and model the deformation of the human body as a mesh. However, the document does not ever mention synthetic CT images nor does it even mention determine tissue density (much less Hounsfield units/attenuation maps) of the patient’s internal tissues. As such and for the foregoing reasons, neither the specification nor incorporated disclosure contains any adequate disclosure of the learning algorithm(s).
In closing, it can be seen that the first rejection addresses all claims and clarifies that the generation of a synthetic CT from a surface image and collection of body scans, regardless of any other factors, is not adequately described in the specification, and the second rejection addresses how applying a learning algorithm in particular to the surface image and collection of body scans is additionally and further not adequately described.
Claims 2-11, 19-20, and 23 are each similarly affected, at least by virtue of dependency.
Response to Arguments
Applicant’s arguments, see pages 10-12, filed 05/13/2025, with respect to the 103(a) rejections have been fully considered and are persuasive. The associated rejections of the previous office action have been withdrawn.
Applicant's arguments filed 05/13/2025 with respect to the 112(a) rejection have been fully considered but they are not persuasive as follows:
As an initial matter, the applicant opines that the exact claim wording was not cited by the examiner in the rejection as the citation in the rejection was taken from prior wording that did not contain learning algorithms. In this instance the examiner notes that the assertion is correct, though the examiner notes that the rejection clarifies that the generation of an SCT image (through any means, learning algorithm included) is not disclosed and that the body of the rejection did address learning algorithms anyways which renders this point moot/a formality. However and regardless, the examiner has updated the wording of the rejection as the applicant’s point is fundamentally correct even if it would not cause the rejection to be deficient because it does not address the substance or merit of the rejection. Additionally, the examiner notes that the applicant’s latest amendment has now separated out plural independent claims with different scopes wherein some use and some do not use learning algorithms. As such the examiner has taken the opportunity to raise a second and separate grounds of rejection under 112(a) for the lack of adequate disclosure of forming a SCT image using a learning algorithm in particular in addition to the lack of adequate disclosure for forming a SCT image by any means that remains of record.
With that addressed, the remainder of the applicant’s arguments only raises points already covered in the examiner’s prior response issued 03/17/2025 which is incorporated by reference to clearly and fully rebut the points raised by the applicant.
The applicant then concludes by opining that because all examined claims are allowable the remaining claims should be rejoined and allowed. In this instance the examiner was not convinced that examined claims are allowable and therefore is similarly not convinced that the application is due for a rejoinder at this juncture.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael S Kellogg whose telephone number is (571)270-7278. The examiner can normally be reached M-F 9am-1pm.
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 (571)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.
/MICHAEL S KELLOGG/ Examiner, Art Unit 3798
/KEITH M RAYMOND/ Supervisory Patent Examiner, Art Unit 3798
Prosecution Insights