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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 May 04, 2026 has been entered.
Claim Objections
Claim 1, 6 are objected to because of the following informalities: “posutre” should be “posture.” “wherein modelwherein” should be “wherein” Appropriate correction is required.
35 USC § 112(f) or 112 (pre-AIA ), Sixth Paragraph
CLAIM INTERPRETATION
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
Claim limitation:
“a generating part …;” recited in claims 8 and dependent claims thereafter has/have been interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because it uses/they use a generic placeholder “part” coupled with functional language “generates” without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
“a deducing part …” recited in claims 8, 12, and dependent claims thereafter has/have been interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because it uses/they use a generic placeholder “part” coupled with functional language “deduces” without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
“a selecting part …” recited in claims 10 and dependent claims thereafter has/have been interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because it uses/they use a generic placeholder “part” coupled with functional language “selects” without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
“a calculating part” recited in claims 12 and dependent claims thereafter has/have been interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because it uses/they use a generic placeholder “part” coupled with functional language “calculates” without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
“a reconfiguring part” recited in claims 15 and dependent claims thereafter has/have been interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because it uses/they use a generic placeholder “part” coupled with functional language “reconfigures” without reciting sufficient structure to achieve the function. Furthermore, the generic placeholder is not preceded by a structural modifier.
Since the claim limitation(s) invokes 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, claim(s) above has/have been interpreted to cover the corresponding structure described in the specification that achieves the claimed function, and equivalents thereof.
A review of the specification shows that the following appears to be the corresponding structure described in the specification for the 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph limitation:
For “a generating part …,” “a deducing part …,” “a selecting part …,” “a calculating part,” and “a reconfiguring part” recited in claims 8, 10, 12, 15 and dependent claims thereafter, the specification (0065, 0066, 0110, 0113) discloses the following:
[0065] The processing part 84 performs an image reconfiguring process and various other operations based on data of the patient 40 acquired by the gantry 2. The processing part 84 has one or more processors, and the one or more processors execute various processes described in the program stored in the storage part 83.
[0066] FIG. 4 is a diagram showing main functional blocks of the processing part 84. The processing part 84 has a generating part 841, a deducing part 842, a confirming part 843, and a reconfiguring part 844.
[0110] FIG. 13 is a diagram showing main functional blocks of the processing part 84 according to embodiment 2. The processing part 84 has a generating part 940, a deducing part 941, a calculating part 942, a confirming part 943, and a reconfiguring part 944.
[0113] The processing part 84 of the console 8 can read the program stored in the storage part 83 and execute the aforementioned operations (b1) to (b5).
Therefore, ““a generating part …,” “a deducing part …,” “a selecting part …,” “a calculating part,” and “a reconfiguring part” recited in claims 8, 10, 12, 15 and dependent claims thereafter have been interpreted as being any processors or any equivalent structures in light of the specification for the purpose of examination.
If applicant wishes to provide further explanation or dispute the examiner’s interpretation of the corresponding structure, applicant must identify the corresponding structure with reference to the specification by page and line number, and to the drawing, if any, by reference characters in response to this Office action.
If applicant does not intend to have the claim limitation(s) treated under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112 , sixth paragraph, applicant may amend the claim(s) so that it/they will clearly not invoke 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, or present a sufficient showing that the claim recites/recite sufficient structure, material, or acts for performing the claimed function to preclude application of 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
For more information, see MPEP § 2173 et seq. and Supplementary Examination Guidelines for Determining Compliance With 35 U.S.C. 112 and for Treatment of Related Issues in Patent Applications, 76 FR 7162, 7167 (Feb. 9, 2011).
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 1-15 and 17-19 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. Applicant has failed to show that Applicant possess any mathematical algorithm for “training the model.” A blank box of unknown information does not satisfy the requirement of the possession. Applicant has failed to identify the relevant descriptions in the Spec. Furthermore, the Examiner cannot find any Spec support for the limitation, “a camera positioned over the table” from the original Specification. It is unclear what structure would support such limitation. It is unclear if the camera 6 in fig. 2 is the support for the limitation “a camera position over the table.” In addition, the Examiner cannot find the support from Originally filed Spec on “wherein learning images and correct answer data corresponding to each normalized posutre are prepared separately … to generate posture-specific learned models,” “wherein the plurality of trained learned models are generated respectively by a plurality of neural networks using learning images and correct answer data prepared separately for different postures of the subject.”
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-15 and 17-19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In re claims 1, 6, 17, it is unclear what the model is in “training a model.” Applicant has failed to provide any mathematical algorithm for the neural network’s learning model. A blank box of unknown information has failed to explaining how this model actually works. Applicant has failed to identify the relevant descriptions in the Spec.
In response to Applicant’s argument that “one of skill in the art would understand there are multiple types of neural network that may be used to train the model, and that the novelty is in the training inputs, and using outputs to determine patient body weight is specific patient position during an imaging scan,” the examiner disagrees. If there is a novelty in training, then a simple input without more, and then simple output with more is NOT novelty. Putting an egg into a black box and claim that the output is an orange is not novelty without explain how. That is magic. Applicant needs to explain how that black box change egg into orange for the purpose of patent applications. The Examiner has indicated that if Applicant is claiming that everyone knows what the black box is, i.e. “one of skill in the art would understand there are multiple types of neural network that may be used to train the model,” then See 101 rejection below. High generality and simple input data with a direct output is NOT novelty. They are data collection to be used because the processing work is done by the black box/machine learning.
In re claims 2-5, for example, claim 2 is referring to “the learned model generating method according to claim 1,” however, claim 1 now is “a method of training a model that outputs … wherein training the model comprises;” it is unclear if claims 2-5 are still based on the claim 1. Again “learned model” is NOT mentioned in claim 1. Applicant still have not resolved this.
In re claim 6 and 17, “the identified posture of the subject” lacks antecedent basis.
Now if the training a model is a model that’s well-known by one of ordinary skill in the art as what Applicant claims. See Arg. Page 7, last para.
Then
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-15, 17-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Statutory Category: YES - The claim recites a method (claim 1) or a medical device (claim 6) and, therefore, is a device.
Step 2A, Prong 1, Judicial Exception: YES - The claim recites the limitation of providing learning images, the providing a weight from training of these learned images, and calculating BMI (note that doctor today already uses a BMI chart with weight and height chart table). This limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of elements such as generic computer components, or additional element of table and camera. That is, other than reciting elements, nothing in the claim element precludes the step from practically being performed in the mind. For example, but for these element language, the claim encompasses a user look at that these picture or patients on these beds many times, and then determined acceptable weight, height and BMI of the patient in his/her mind. The mere nominal recitation of a generic elements does not take the claim limitation out of the mental processes grouping. Thus, the claim recites a mental process.
Step 2A, Prong 2, Integrated into Practical Application: No - The claim recites additional elements: images of different posture of patient on a table, number of images. The storing step is recited at a high level of generality amounts to mere data gathering, which is a form of insignificant extra-solution activity. The processing circuitry that performs the comparison step is also recited at a high level of generality, and merely automates the comparison step. Each of the additional limitations is no more than mere instructions to apply the exception using a generic computer component, or elements such as generic camera and a table.
The combination of these additional elements is no more than mere instructions to apply the exception using a generic computer component, or elements such as generic camera and a table. Accordingly, even in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claim is directed to the abstract idea.
Step 2B, Inventive Concept: No - As discussed with respect to Step 2A Prong Two, the additional elements in the claim amount to no more than mere instructions to apply the exception using a generic computer component, or elements such as generic camera and a table. The same analysis applies here in 2B, i.e., mere instructions to apply an exception on a generic computer cannot integrate a judicial exception into a practical application at Step 2A or provide an inventive concept in Step 2B. Under the 2019 PEG, a conclusion that an additional element is insignificant extra-solution activity in Step 2A should be re-evaluated in Step 2B. Here, images of different posture of patient on a table, number of images were considered to be extra-solution activity in Step 2A, and thus it is re-evaluated in Step 2B to determine if it is more than what is well-understood, routine, conventional activity in the field. The background of the example does not provide any indication that the processing circuitry and storage is anything other than a generic, off-the-shelf computer component, and the Symantec, TLI, and OIP Techs. court decisions cited in MPEP 2106.05(d)(II) indicate that mere collection or receipt of data over a network is a well‐understood, routine, and conventional function when it is claimed in a merely generic manner (as it is here). Accordingly, a conclusion that the collecting and comparing step is well-understood, routine, conventional activity is supported under Berkheimer Option 2. For these reasons, there is no inventive concept in the claim, and thus it is ineligible.
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.
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.
Claim(s) 1-11 and 17-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sa et al. (US 2020/0271507, hereinafter Sa ‘507 in view of Tadayon et al. (US 2014/0079297, hereinafter Tadayon ‘297) and further in view of Tamersoy et al. (US 2020/0297237, hereinafter Tamerosy ‘237).
In re claim 1, Sa ‘507 teaches a method of training a model that outputs a body weight of an imaging subject based on an input image of the imaging subject lying on a table of a medical device, wherein training the model comprises:
providing a plurality of learning images generated based on a plurality of camera images of a human lying on a table of a medical device (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070),
wherein the plurality of camera images are obtained from a camera positioned over the table of the medical device (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070); and
providing a plurality of correct answer data corresponding to the plurality of learning images, where each of the plurality of correct answer data represents a body weight of a human included in a corresponding learning image (0052, 0055-0059); and
Sa ‘507 teaches the surface data is normalized prior to input. The surface data is rescaled, resized, warped, or shifted (e.g., interpolation). See 0038.
Sa ‘507 merely fails to teach to normalize orientation and posture of the subject.
Tadayon ‘297 teaches wherein the camera images are pre-processed to normalize orientation (0481) and posture of the subject (0481, note that, head or face of a human to be in the right angle or direction is a posture of the head or face).
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition.
Sa ‘507 and Tadavon ‘297 fail to teach training a plurality of models, each of the plurality of models corresponding to a different normalized posture of the subject, wherein each model is trained using learning images and correct answer data associated with the corresponding posture.
Tamerosy ‘237 teaches training a plurality of models, each of the plurality of models corresponding to a different
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition, and to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose.
Although Sa ‘507, Tadayon ‘297 and Tamerosy ‘237 fails to explicitly teach “wherein learning images and correct answer data corresponding to each normalized posture are prepared separately, and …. to generate posture-specific learned models,” it would have been obvious to generate posture-specific learned models. It would have been obvious from the teachings of Tamerosy ‘237 that each pose data with respect to weight can be trained to obtain that specific pose-weight model. The AI/machine learning model is doing all the work, and the input is merely user preference; and the training of Tamerosy ‘237 can be done as well. Furthermore, Tamerosy ‘237 actually teaches:
[0052] The body pose network 36 is designed or configured to output a class membership for pose, such as one of four classes (e.g., head-first supine, feet-first supine, head-first prone, and feet-first prone). A dense net, sequential layer structure (e.g., feed forward without skip connections), convolutional neural network, or another network 37 is provided. In the example of FIG. 3, two stages of a convolution layer with 3×3 kernels, spatial dropout layer, and max pooling layer feed to two stages of dense layer and spatial drop out layer.
[0053] Additional, different, or fewer networks 30, 32, 34, 36 may be provided in the multi-task progressive neural network. FIG. 3 shows four sub-parts where each sub-part handles a different patient modeling task. Other workflows may use two, three, five, or more patient modeling tasks and corresponding networks. The multi-task progressive neural network is designed or configured based on the number of tasks with each component network architecture being designed or configured based on the task. In FIG. 3, each of the columns, when examined separately, forms a Convolutional Neural Network (CNN).
[0054] The machine-learned models of the progressive multi-task network estimates the characteristics. The progressive multi-task network is a machine-learned network for performing multiple tasks. Each machine-learned model includes learned features, such as learned convolution kernels, for the task.
Hence, it would have been obvious that one of ordinary skill in the art can use what Tamerosy ‘237 teaches to arrive the feature of “wherein learning images and correct answer data corresponding to each normalized posture are prepared separately, and …. to generate posture-specific learned models” in order to obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
Hence, It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition, and to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose and obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
In re claim 2, Sa ‘507 teaches wherein the plurality of learning images includes an image of a human lying on a table in a prescribed posture (fig. 4).
In re claims 3-5, Sa ‘507 teaches the camera captures the outer surface with the patient in a particular position, such as capturing a front facing surface as the patient lies in a bed or on a table for treatment or imaging (0033). Sa ‘507 fails to teach patient in a particular position/posture.
Tamerosy ‘237 teaches an image of the human lying on a table in a different posture from the prescribed posture of claim 3 (0006, 0010, 0011, 0012, 0025, 0032, 0042, 0049, 0052, 0066, 0068, 0069, 0081, 0093, 0094); specifically, teaches: [0052] The body pose network 36 is designed or configured to output a class membership for pose, such as one of four classes (e.g., head-first supine, feet-first supine, head-first prone, and feet-first prone) of which reads on claim 4: wherein the plurality of learning images includes at least two of: a first learning image of the human lying in a supine position; a second learning image of the human lying in a prone position; a third learning image of the human lying in a left lateral decubitus position; and a fourth learning image of the human lying in a right lateral decubitus position, and read on claim 5: wherein the plurality of learning images include an image of the human lying on a table in a head- first condition and an image of the human lying on a table in a feet-first condition. See also para 0011, 0042, etc.
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 and Tadayon ‘297 to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose.
In re claims 6 and 17, Sa ‘507 teaches a medical device comprising: a table; a camera positioned over the table, the camera to obtain a camera image of an imaging subject positioned on the table (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070); and
a processing device that executes instructions stored on a memory of the processing device (fig. 7, memory 74),
wherein the instructions include:
Sa ‘507 teaches the surface data is normalized prior to input. The surface data is rescaled, resized, warped, or shifted (e.g., interpolation). See 0038.
Sa ‘507 fails to teach to pre-processing the camera image to normalize orientation and posture of the subject: selecting, based on the posture of the subject, a trained learned model from a plurality of trained learned models, each trained corresponding to a different posture of the subject; and
Tadayon ‘297 teaches pre-processing the camera image to normalize orientation (0481) and posture of the subject (0481, note that, head or face of a human to be in the right angle or direction is a posture of the head or face).
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition.
Tamerosy ‘237 teaches selecting, based on the posture of the subject, a trained learned model from a plurality of trained learned models, each trained corresponding to a different posture of the subject (0006, 0010-0011, 0032, 0042, 0066, 0081, note that pre-processing and normalization are taught by Tadayon ‘297), wherein each model is trained using learning images and correct answer data (0040, 0057) associated with the corresponding posture (0053, 0054, 0057, 0074).
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition, and to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose.
Although Sa ‘507, Tadayon ‘297 and Tamerosy ‘237 fails to explicitly teach “wherein the plurality of trained learned images are generated respectively by a plurality of neural network using learning images and correct answer data prepared separately for different postures of the subject, and …. Wherein the trained learned model corresponds to the identified posture of the imaging subject,” it would have been obvious to generate posture-specific learned models. It would have been obvious from the teachings of Tamerosy ‘237 that each pose data with respect to weight can be trained to obtain that specific pose-weight model. The AI/machine learning model is doing all the work, and the input is merely user preference; and the training of Tamerosy ‘237 can be done as well. Furthermore, Tamerosy ‘237 actually teaches:
[0052] The body pose network 36 is designed or configured to output a class membership for pose, such as one of four classes (e.g., head-first supine, feet-first supine, head-first prone, and feet-first prone). A dense net, sequential layer structure (e.g., feed forward without skip connections), convolutional neural network, or another network 37 is provided. In the example of FIG. 3, two stages of a convolution layer with 3×3 kernels, spatial dropout layer, and max pooling layer feed to two stages of dense layer and spatial drop out layer.
[0053] Additional, different, or fewer networks 30, 32, 34, 36 may be provided in the multi-task progressive neural network. FIG. 3 shows four sub-parts where each sub-part handles a different patient modeling task. Other workflows may use two, three, five, or more patient modeling tasks and corresponding networks. The multi-task progressive neural network is designed or configured based on the number of tasks with each component network architecture being designed or configured based on the task. In FIG. 3, each of the columns, when examined separately, forms a Convolutional Neural Network (CNN).
[0054] The machine-learned models of the progressive multi-task network estimates the characteristics. The progressive multi-task network is a machine-learned network for performing multiple tasks. Each machine-learned model includes learned features, such as learned convolution kernels, for the task.
Hence, it would have been obvious that one of ordinary skill in the art can use what Tamerosy ‘237 teaches to arrive the feature of “wherein the plurality of trained learned images are generated respectively by a plurality of neural network using learning images and correct answer data prepared separately for different postures of the subject, and …. Wherein the trained learned model corresponds to the identified posture of the imaging subject” in order to obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
Hence, It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition, and to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose and obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
In re claim 7, Sa ‘507 teaches wherein the trained learned model outputs the body weight of the imaging subject when an input image generated based on the camera image is input (fig. 1 and 2, 0052, 0055-0059).
In re claim 8, Sa ‘507 teaches a generating part that generates the input image based on the camera image; and a deducing part that deduces the body weight of the imaging subject by inputting the input image into the trained learned model (fig. 1 and 2, 0052, 0055-0059).
In re claim 9, Sa ‘507 teaches wherein the trained learned model is generated by a neural network executing learning using: a plurality of learning images generated based on a plurality of camera images of a human positioned on a table of a medical device (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070); and a plurality of correct answer data corresponding to the plurality of learning images, where each of the plurality of correct answer data represents a body weight of a human included in a corresponding learning image (0049, 0050, 0058, 0059).
In re claim 10, Sa ‘507 teaches wherein the processing device includes: a selecting part that selects a learned model used for deducing the body weight of the imaging subject from the plurality of learned models corresponding to a plurality of possible postures of the imaging subject during imaging (0040-0044), wherein the deducing part deduces the body weight of the imaging subject using the selected learned model (0040-0044).
In re claim 11, Sa ‘507 teaches wherein the processing device includes a confirming part for confirming to an operator whether or not to update a deduced body weight (0062, 0063, 0077, 0081, 0084).
In re claim 18, note that lateral decubitus position refers to a body orientation where an individual lies on their side.
Tamerosy ‘237 teaches wherein the different postures of the subject includes prone, supine, left lateral decubitus, and right lateral decubitus (0006, 0042).
In re claim 19, Tadayon ‘297 teaches identifying the posture of the subject based on the image (0328, 0347, 0350).
Claim(s) 12-15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sa ‘507, Tadayon ‘297 and Tamerosy ‘237 in view of De Brouwer et al. (US 2018/0289334, hereinafter De Brouwer ‘334).
In re claim 12, Sa ‘507 teaches comprising: a deducing part that deduces the height of the imaging subject, containing a learned model that outputs the body height of the imaging subject when an input image generated based on the camera image is input (0072); but fails to teach a calculating part that calculates the body weight of the imaging subject based on the height and BMI of the imaging subject.
De Brouwer ‘334 teaches calculates the body weight of the imaging subject based on the height and BMI of the imaging subject (0028-0029, note that the equation to calculate BMI can be easily adjust to Mass = BMI * height).
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 and Tadayon ‘297 to include the features of De Brouwer ‘334 in order to use additional weight related information as input to allow the learning model to learn more information in order to arrive at better answers.
In re claim 13, Sa ‘507 teaches wherein the learned model is generated by a neural network executing learning using: a plurality of learning images generated based on a plurality of camera images of a human lying on a table of a medical device (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070); and a plurality of correct answer data corresponding to the plurality of learning images, where each of the plurality of correct answer data represents a height (0072, note that weight and/or heights can be both obtained) of a human included in a corresponding learning image (0049, 0050, 0058, 0059, hence, it is obvious that the fig. 5 of the volume vs weight can be adjust to volume vs height per teaching of para 0072).
In re claim 14, Sa ‘507 teaches further comprising a generating part that generates the input image based on the camera image (fig. 7, sensor 77, patient 78, table 79; 0032, 0035, 0068-0070).
In re claim 15, De Brouwer ‘334 teaches comprising: a reconfiguring part that reconfigures a scout image obtained by scout scanning the imaging subject, wherein the calculating part calculates the BMI based on the scout image (0027-0029).
It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 and Tadayon ‘297 to include the features of De Brouwer ‘334 in order to use additional weight related information as input to allow the learning model to learn more information in order to arrive at better answers.
Response to Arguments
Applicant's arguments filed on May 04, 2026 have been fully considered but they are not persuasive.
In response to Applicant’s argument that “one of skill in the art would understand there are multiple types of neural network that may be used to train the model, and that the novelty is in the training inputs, and using outputs to determine patient body weight is specific patient position during an imaging scan,” the examiner disagrees. If there is a novelty in training, then a simple input without more, and then simple output with more is NOT novelty. Putting an egg into a black box and claim that the output is an orange is not novelty without explain how. That is magic. Applicant need to explain how that black box change egg into orange for the purpose of patent applications. The Examiner has indicated that if Applicant is claiming that everyone knows what the black box is, i.e. “one of skill in the art would understand there are multiple types of neural network that may be used to train the model,” then See 101 rejection. High generality and simple input data with a direct output is NOT novelty. They are data collection to be used because the processing work is done by the black box/machine learning. Furthermore, when Applicant argues that black box is known in the art, then what exactly is new? It seems like that black box or a neural network/machine learning model does everything. It’s like Applicant argues that if Applicant put a person with 100 lb lying face up in a black box, the black box is going to say that person is 100 lb.
In response to Applicant’s argument that:
the claims do not recite a mental process. The claims include limitations including: acquiring images obtained from a camera, pre-processing the images to normalize orientation and posture of a subject, training machine-learning models using labeled training data, and determining a body weight of the subject based on the camera image. These steps require physical image acquisition hardware, computer-based image processing, and machine-learning model execution. Training and using learned models with posture-normalized image inputs cannot be practically or meaningfully performed in the human mind.
The Examiner disagrees.
The Examiner clearly indicated that “the claim recites the limitation of providing learning images, the providing a weight from training of these learned images, and calculating BMI (note that doctor today already uses a BMI chart with weight and height chart table). This limitation, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of elements such as generic computer components, or additional element of table and camera.” These physical hardwares are generic, and have been identified that none of these are being integrated under Step 2A prong 2. Applicant also acknowledges that the machine learning is generic that one of ordinary skill in the art would know. See. Arg. Page 9, last para. Hence, the input data gathering on different posture images are only data gathering that can be viewed by a user and be remembered. A user by seeing tons of patient with a weight measurement can provide a good guess on the next patient’s weight. Applicant also admit that “a person may be able to roughly determine difference between a 100 lb patient and a 200 lb patient.” See Arg. Page 11, last para. The claims not provide any specificity of how precise the weight determination has be to.
In response to Applicant’s argument that:
they recite a specific technical workflow that uses a camera integrated with a medical imaging device, processes posture-dependent patient images, applies posture-specific learned models, and generates patient body-weight information used by the imaging system. This workflow improves the operation of a medical imaging system by enabling patient-specific parameter determination (e.g., body weight) directly from camera images obtained while the patient is already positioned on the imaging table.
The Examiner disagrees. Again camera is used for data gathering, table is merely where the data gather take places. The prior arts above have shown that this workflow is known. Applicant hence is not improving the workflow. Furthermore, Applicant has not improved the actual machine learning algorithm because Applicant claims that one of ordinary skill in the art would know how to use the algorithm. Hence, Applicant has not improve on anything.
In response to Applicant’s argument that “the imaging system can reliably obtain current patient data without requiring manual weighing or reliance on outdated records, thereby improving imaging setup, dose control, and clinical workflow efficiency,” the Examiner disagrees. Applicant failed to provide any factual evidence to show this. Furthermore, Applicant’s argument is similar to a person argues that because a calculator can calculate 3.14 x 100 and output correct answer that is better than a person can, therefore calculation of number is not an abstract idea because it improves the accuracy of the calculation. Just like the generic calculator, Applicant did not improve the machine learning algorithm. Applicant did not use any specific machine learning algorithm because Applicant fails to show this specific machine learning algorithm. Hence, anyone can use a generic machine learning algorithm, and with the teaching of the prior arts about pose and weight monitoring currently, Applicant has failed to show any improvement.
In response to Applicant’s argument that:
the claims do not broadly recite "determine body weight using AI." Instead, they require posture-normalized camera images of a subject lying on a medical device table, learned models trained on posture-specific training data … These limitations define how the result is achieved in a technically constrained manner, not merely what result is achieved. The claims therefore recite a technological solution to the technological problem of accurate, automated acquisition of patient body-weight information in a medical imaging environment.
the Examiner disagrees. The Examiner has showed that normalization of data is conventional and being taught. The Examiner has shown that using camera to obtain a posture of a patient lying on a medical device table is known and being taught. The Examiner has shown that a generical machine learning can be trained to determine posture-weight relationship.
Furthermore, Applicant is claiming what result is achieved. Applicant claims the input data, i.e. images taken by a camera, and then using machine learning for training, to obtain the body weight.
The claim does not recite a technological solution to a technological problem of accurate, automated acquisition of patient body-weight information in a medical imaging environment because as prior arts have shown that body weight info can be obtained from imaging of a pose. And Applicant has failed to provide any factual evidence to support claim of better accuracy than others using the same generic neural network.
In response to Applicant’s argument that:
Sa, Tadayon, and Tamerosy fail to teach or suggest seelcting, based on the posture of the subject, a trained learned model from a plurality of trained learned models, each trained corresponding to a different posture of the subject, wherein the plurality of trained learned models are generated respectively by a plurality of neural networks using learning images and correct answer data prepared separately for different postures of the subject, and determining a body weight of the imaging subject based on the camera image of the imaging subject positioned on the table of the medical device using the selected trained learned model, wherein the selected trained learned model corresponds to the identified posture of the subject.
The Examiner disagrees.
As explained above, it would have been obvious from the teachings of Tamerosy ‘237 that each pose data with respect to weight can be trained to obtain that specific pose-weight model. The AI/machine learning model is doing all the work, and the input is merely user preference; and the training of Tamerosy ‘237 can be done as well. Furthermore, Tamerosy ‘237 actually teaches:
[0052] The body pose network 36 is designed or configured to output a class membership for pose, such as one of four classes (e.g., head-first supine, feet-first supine, head-first prone, and feet-first prone). A dense net, sequential layer structure (e.g., feed forward without skip connections), convolutional neural network, or another network 37 is provided. In the example of FIG. 3, two stages of a convolution layer with 3×3 kernels, spatial dropout layer, and max pooling layer feed to two stages of dense layer and spatial drop out layer.
[0053] Additional, different, or fewer networks 30, 32, 34, 36 may be provided in the multi-task progressive neural network. FIG. 3 shows four sub-parts where each sub-part handles a different patient modeling task. Other workflows may use two, three, five, or more patient modeling tasks and corresponding networks. The multi-task progressive neural network is designed or configured based on the number of tasks with each component network architecture being designed or configured based on the task. In FIG. 3, each of the columns, when examined separately, forms a Convolutional Neural Network (CNN).
[0054] The machine-learned models of the progressive multi-task network estimates the characteristics. The progressive multi-task network is a machine-learned network for performing multiple tasks. Each machine-learned model includes learned features, such as learned convolution kernels, for the task.
Hence, it would have been obvious that one of ordinary skill in the art can use what Tamerosy ‘237 teaches to arrive the feature of “wherein the plurality of trained learned images are generated respectively by a plurality of neural network using learning images and correct answer data prepared separately for different postures of the subject, and …. Wherein the trained learned model corresponds to the identified posture of the imaging subject” in order to obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
Hence, It would have been prima facie obvious to one of ordinary skills in the art at the time of invention to modify the method/device of Sa ‘507 to include the features of Tadayon ‘297 in order to increases the accuracy of the recognition, and to include the features of Tamerosy ‘237 in order to better control and operate the medical device based on the changes in patient pose and obtain a pose-specific modeling to simplify the finding a weight within the specific pose.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BO JOSEPH PENG whose telephone number is (571)270-1792. The examiner can normally be reached Monday thru Friday: 8:00 AM-5:00 PM EST.
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, ANNE M KOZAK can be reached at (571) 270-0552. 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.
/BO JOSEPH PENG/Primary Examiner, Art Unit 3797
