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 .
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.
Claim(s) 3, 16, 24-25 and 27-28 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sahbaee Bagherzadeh et al. (US Pub No. 2019/0313990) in view of Zhao et al. (US Pub No. 2020/0012904) and Kalafut (US Pub No. 2010/0030073).
With regards to claim 3, Sahbaee Bagherzadeh et al. disclose a computer-implemented method, comprising:
training a neural network algorithm (i.e. “trained regressor (e.g. neural network) 220”) to estimate, based on a measurement indicative of a reference contrast medium bolus (i.e. “test bolus”, “bolus tracking curves 231”), pharmacokinetics of a contrast medium through a cardiovascular system of a patient, the training being dependent on at least one loss (paragraphs [0023]-[0026], referring to the use of a trained regressor (e.g., neural network ) 220 to determine the contrast agent application protocol (233), wherein the neural network (220) can be trained with contrast enhancement data (211) from the model (212) and evaluates the contrast enhancement data (211) defining the relationship between the injected contrast volume and its distribution at different locations as a function of time; further referring to the inputs to the neural network (220) being CA administration curve of test bolus data or bolus tracking data (231) and patient attribute, such as 2D or 3D image (232) from camera (101); paragraphs [0027]-[0029], referring to simulating flow of a contrast agent through a plurality of patient’s bodies using a simulator (212) and providing the contrast enhancement data (211) based on the simulating for use in the neural network; paragraphs [0043]-[0045], referring to the model (212) corresponding to a physiologically based pharmacokinetic (PBPK) model which can simulate contrast agent propagation in the human body, wherein the model can predict absorption, distribution, and metabolism of the contrast agent within the body, beginning with injection and wherein in Figure 3, each block represents an organ compartment and each ellipse represents a blood vessel compartment; paragraphs [0055]-[0057]; paragraph [0063], referring to incorporating rating data for continuously learning the best injection protocols to minimize dose and contrast volume, while maximizing image quality, wherein the learning machine of regressor (220) can learn from the “difference between the prediction results and real data”, wherein such a difference represents the “at least one loss” which the training is dependent upon; Figure 1, Figure 3, which depicts a model of the heart represented by three separate components (the left heart, right heart and heart muscle); Figure 4, which depicts the contrast enhancement data (211) generated by the multi-compartmented model (212); Figure 6; Table 1); and
configuring an angiographic imaging protocol (i.e. “contrast agent application protocol 233”) based on an estimate of the pharmacokinetics of the contrast medium obtained from the neural network algorithm (paragraph [0002], referring to dyhnamic contrast-enhanced (DCE) imaging providing dynamic information about the flow of an injected contrast agent through blood vessels to different tissues; paragraphs [0015]-[0016], referring to using machine learning to predict a scanning protocol for contrast enhancement imaging an internal organ on the subject based on the prediction, wherein the scanning can be computed tomography (CT) scanning; paragraphs [0023]-[0026], referring to the use of a trained regressor (e.g., neural network ) 220 to determine the contrast agent application protocol (233), paragraphs [0054]-[0057], referring to the outputs of the neural network (220) being used to determine the quantity of contrast agent, contrast agent administration protocol, and the scan delay (relative to the beginning of injection) and are used immediately for contrast imaging while the patient is still on the bed; Figures 1-3, 6; Table 1).
However, Sahbaee Bagherzadeh et al. do not specifically disclose that the “at least one loss” comprises multiple losses, wherein each loss among the multiple losses is based on a different function.
Further, though Sahbaee Bagherzadeh et al. do disclose that the at least one loss includes a loss determined based on a model of the cardiovascular system, the model being based on a response function of a vessel of the cardiovascular system with respect to a time-dependent inflow of the contrast medium (paragraphs [0033]-[0034], [0043]-[0044], [0048], [0063]; Figure 3; Table 1), Sahbaee Bagherzadeh et al. do not specifically disclose that the model is a linear time invariant model.
Zhao et al. disclose a system for training a convolutional neural network based on training data and a plurality of images, wherein a fourth loss function can be generated based on a first loss function, a second loss function and a third loss function, wherein the first, second and third loss functions are based on different functions (Abstract; paragraphs [0062]-[0067], referring to the first loss function being based on a plurality of masks, the second loss function being generated based on at least one image level label associated with the plurality of images and the third loss function being generated based on the bounding box, and therefore the loss functions are based on different functions; Figure 9). The fourth loss function is iteratively back propagate to tune parameters of the convolutional neural network based on the training data, which improves the performance of the neural network (paragraphs [0022], [0028], [0067]; Figure 9).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify the “at least one loss” of Sahbaee Bagherzadeh et al. to comprise multiple losses, wherein each loss among the multiple losses is based on a different function, as taught by Zhao et al., in order to improve the performance of the neural network (paragraphs [0022], [0028], [0067]).
However, though Sahbaee Bagherzadeh et al. do disclose that the at least one loss includes a loss determined based on a model of the cardiovascular system, the model being based on a response function of a vessel of the cardiovascular system with respect to a time-dependent inflow of the contrast medium (paragraphs [0033]-[0034], [0043]-[0044], [0048], [0063]; Figure 3; Table 1), the above combined references do not specifically disclose that the model is a linear time invariant model.
Kalafut discloses a method and/or device for the prediction of a likely contrast medium behavior for a contrast medium-assisted examination of an object under investigation, wherein, given that model/patient transfer function is identified, one may attempt to solve for an input signal that will produce a desired output (that is a desired level of contrast enhancement in an anatomical region of interest), wherein the model can assume linearity and time-invariance (Abstract; paragraphs 0002], [0027], [0098], [0154]). A method of delivering a contrast enhancing fluid to a patient using an injector system can comprise determining an injection procedure input using at least one patient transfer function to determine an injection procedure input, wherein the patient transfer function can be based on mathematical models to control injection of fluid into the patient to create a patient response can be based at least in part on the basis of a mathematical model, wherein the mathematical model can assumes linearity and time invariance (Abstract; paragraphs [0021]-[0023]; [0027], [0044], claim 19). Assuming a Linear Time Invariant system, the input-output relationship of a discrete-time (sampled) system is used (see Eq. 15, wherein in the equation 15, the H(z) term can be computed by a-priori modeling of the patient/drug system, computed by system identification techniques operating on data collected during a brief inquiry of the system with a small injection of pharmaceutical, or computed with a combination of both approaches (Abstract; paragraphs [0012], [0154]).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to substitute the model of Sahbaee Bagherzadeh et al. with a linear time invariant model that is based on a response function of a vessel of the cardiovascular system with respect to a time-dependent inflow of the contrast medium, as taught by Kalafut, as the substitution of one known model for another yields predictable results (i.e. providing an estimate of the pharmacokinetics of a contrast medium) to one of ordinary skill in the art. One of ordinary skill in the art would have been able to carry out such a substitution and the results are reasonably predictable.
With regards to claim 16, Sahbaee Bagherzadeh et al. disclose a computing device comprising at least one processor (111) and a memory (112, 114), the at least one processor being configured to load program code from the memory and to execute the program code, wherein the at least one processor, upon executing the program code, is configured to cause the computing device to perform the method of claim 3 [see rejection of claim 3] (paragraphs [0017], [0066], referring to the methods and system being embedded in the form of computer-implemented processes and apparatus for practicing those processes, wherein when the computer program code is loaded into and executed by a computer, the computer becomes an apparatus for practicing the method; Figure 1).
With regards to claim 24, Sahbaee Bagherzadeh et al. disclose that the measurement includes time series data that includes samples that are irregularly spaced in the time domain (paragraphs [0022], [0052]-[0053], [0060]-[0061], referring to the test bolus enhancement curve; paragraph [0025], referring to the contrast enhancement data (211) defining the relationship between the injected contrast volume and its distribution at different locations as a function of time; paragraph [0051]; Figures 4, 6), wherein the neural network algorithm includes a Neural Ordinary Differential Equations algorithm (paragraphs [0046]-[0047], referring to the flows of contrast agent into and out of each organ and changes in concentration within the organ can be expressed by a respective differential equation, wherein, for example, the differential equations for each organ can indicate that blood flow rate out of the organ equals blood flow rate into the organ, which corresponds to differential equations dependent on only a single independent variable (i.e. and therefore, by definition, corresponding to “ordinary differential equations”).
With regards to claim 25, Zhao et al. disclose that the method further comprises weighting the multiple losses (paragraph [0066], referring to the weights being applied to the first, second and third losses; Figure 9).
With regards to claim 27, Sahbaee Bagherzadeh et al. disclose that the another loss among the multiple losses is based on a deviation of an estimated contrast medium enhancement curve (paragraphs [0043]-[0053], referring to the contrast enhancement curves and referring to the differential equations, which express the flows of contrast agent into and out of each organ and changes in concentration in organ and are ultimately used to derive the contrast enhancement curves, being used to indicate the difference/loss between the input CA concentration and the output CA concentration; Figure 4)
With regards to claim 28, Sahbaee Bagherzadeh et al. disclose that another loss among the multiple losses is based on a deviation from an analytical model (paragraphs [0043]-[0053], referring to the use of compartment models (i.e. analytical model) being used to determine the difference/loss between input and output CA concentrations).
Claim(s) 26 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sahbaee Bagherzadeh et al. in view of Zhao et al. and Kalafut, as applied to claim 3 above, and further in view of Hong et al. (US Pub No. 2020/0089653) .
With regards to claim 26, as discussed above, the above combined references meet the limitations of claim 3. However, they do not specifically disclose that the weighting is based on uncertainty-weighting.
Hong et al. disclose the use of an ensemble model that combines result values of a plurality of models to output a final result value, thereby more accurately predicting a biometric state of the user (paragraphs [0032]-[0036]). An uncertainty determination model may obtain weight values based on an uncertainty of a result value and the final output value of the ensemble model may be obtained by using the result values with the weight values applied thereto, thereby obtaining more accurate results (paragraphs [0036], [0047]-[0049], [0093], [0097]-[0098], [0100]).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to have the weighting of the above combined references be based on uncertainty-weighting, as taught by Hong et al., in order to obtain more accurate results (paragraphs [0036], [0047]-[0049]).
Response to Arguments
Applicant’s arguments with respect to claim(s) 3, 16 and 24-28 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Zhao has been introduced to teach that the training is dependent on multiple losses, wherein each loss is based on a different function, etc..
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 KATHERINE L FERNANDEZ whose telephone number is (571)272-1957. The examiner can normally be reached Monday-Friday 9:00 AM - 5:30 PM (ET).
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, Pascal Bui-Pho can be reached at (571) 272-2714. 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.
/KATHERINE L FERNANDEZ/ Primary Examiner, Art Unit 3798