DETAILED ACTION
Claims 1-19 are pending.
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 Objections
Claims 10 and 17 are objected to because of the following informalities:
Line 3 of claim 10 recites “configured for” twice. This appears to be a typographical error and the claim limitation should read “a PET scanner configured for collecting a plurality…”
Lines 1-2 of claim 17 recite “wherein the ANN includes” twice. This appears to be a typographical error and the beginning of the claim should be “The system of claim 10 wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation…”
Appropriate correction is required.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1, 3-6, 10, 12-15, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Feng et al. (U.S. Publication No. US 2021/0150705 A1) ("Feng") in view of Zhao et al. (U.S. Publication No. US 2022/0198667 A1) ("Zhao") and further in view of Zhong et al. (NPL titled "Optimization of a Model Corrected Blood Input Function From Dynamic FDG-PET Images of Small Animal Heart In Vivo”) (“Zhong”).
Regarding claim 1, Feng discloses a method for automatically computing a blood input function (Fig. 1 and 5; [0043 and [0076], wherein plasma input function corresponds to blood input function) for dynamic positron emission tomography (PET) (Fig. 1; [0043-0044] [0077] and [0092]), the method comprising:
obtaining a plurality of dynamic PET image data sets comprising volumetric radioactive measurement data associated with an administered radioactive tracer ([0063-0064], wherein obtaining module obtains data for image; [0043] and [0067], wherein volume correction of the obtained data is performed thereby necessitating volumetric data; [0086] and [0088-0089], wherein the concentration of radioactivity (i.e. volumetric radioactive measurement data, since the calculation of concentration requires volume) of the tracer (i.e. radioactive tracer) is determined) present in a target site of a subject (Fig. 4A; [0043], [0046], [0063-0064], and [0088], wherein a plurality of PET images are obtained after a tracer is injected (i.e. administered) into the subject’s (i.e. a target site, where the subject may include a specific portion/organ of the patient) blood vessels) over multiple scanning intervals (Fig. 8; [0043], [0064], [0085], and [0118]);
Feng further teaches utilizing an artificial neural network (ANN) to model the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site (Fig. 9; [0064-0065], [0073], [0078], and [0119-0120], wherein a convolutional neural network (a convolutional neural network is a specific type of ANN) obtains a blood vessel model providing constraints to characteristics of the blood vessel for the second image and therefore the geometric info for the bold vessel is known). However, Feng fails to teach utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site, specifically (emphasis added). Zhao, on the other hand, teaches acquiring a segmented blood vessel from an image using an ANN. More specifically and as it relates to the applicant’s claims, Zhao discloses utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site ([0056-0058] and [0075], wherein an image segmentation unit uses an ANN to acquire a segmented blood vessel image from an original PET image).
Zhao is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to optimize the region of interest and remove unnecessary tissues or parts (Zhao, [0061-0062]).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site, as taught by Zhao, into the method, as taught by Feng, to obtain the invention as specified in claim 1.
Feng in view of Zhao further teaches automatically deriving , using the ANN, a blood input function (IDIF) based on radioactive tracer concentrations measured in one or more segments of the plurality of dynamic PET image data sets, the blood input function being a predictive model-corrected plasma input function that is also a time activity curve (Feng [0068], [0073], [0068], [0088] and [0108], wherein a plasma input function (IDIF) is determined based on corrected fourth images, which are generated by processing an image in the image sequence based on a blood vessel model ([0108]) (i.e. models of the plurality of dynamic PET image data sets, generated using the trained machine learning model (i.e. ANN, since a machine learning model is a convolutional neural network, which is a specific type of ANN), and describes the change of the concentration of radioactivity of the tracer in the plasma with time).
Although Feng directly uses partial volume correction to compute a blood input function ([0047], [0068], and [0088], wherein the plasma input function (i.e. IDIF) is based on partial volume corrected values), Feng and Zhao do not further teach computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF, specifically (emphasis added). Zhong, on the other hand, teaches using the time activity curve of the image-derived blood input function to determine a model corrected time activity curve. More specifically, and as it relates to the applicant’s claims, Zhong discloses computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF (Section I paragraph 1, wherein the IDIF is defined to be the same as the blood pool time activity curve (TAC) and therefore the model corrected time activity curve is the same as the MCIF; Section II-D paragraphs 1 and 3; Section III paragraph 2; Fig. 3(e), wherein the IDIF/ blood pool TAC is corrected due to spill-over radioactivity and partial volume effects, producing a model corrected TAC (i.e. MCIF)).
Zhong is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to accurately and precisely determine the downstream rate of FDG influx (Zhong, Section IV paragraph 2).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF, as taught by Zhong, into the method, as taught by Feng and Zhao, to obtain the invention as specified in claim 1.
Claim 19 has limitations that are substantially similar to claim 1. Therefore, the rejection applied to claim 1, please see above, also applies equally to claim 19. Furthermore, Feng teaches one or more non-transitory computer readable media ([0038]) having stored thereon executable instructions ([0038]) that when executed by a processor of a computer cause the computer to perform steps ([0038]).
Regarding claim 3, Feng, Zhao, and Zhong teach the method of claim 1.
Feng additionally teaches wherein deriving the IDIF includes continuously collecting the plurality of volumetric radioactive measurements at the multiple scanning intervals over a predefined time period ([0068], [0085], and [0087-0088], wherein obtaining the one or more fourth images associated with the blood vessel, which determines the plasma input function (i.e. deriving the IDIF), are determined by scanning the subject in each of the multiple consecutive time periods (i.e. continuously collecting at the multiple scanning intervals over a predefined time period), and partial volume correction is performed on each of the fourth images ([0043] and [0067], wherein volume correction of the obtained data is performed thereby necessitating volumetric radioactive measurements; [0086] and [0088-0089], wherein the concentration of radioactivity (i.e. volumetric radioactive measurement data, since the calculation of concentration requires volume) of the tracer is determined).
Regarding claim 4, Feng, Zhao, and Zhong teach the method of claim 1. Feng additionally discloses wherein prior to the obtaining step, a subject is injected with the radioactive tracer ([0043] and [0085]).
Regarding claim 5, Feng, Zhao, and Zhong teach the method of claim 1.
Feng fails to teach wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET. Zhong, on the other hand, teaches performing dynamic 18F-FDG PET. More specifically, and as it relates to the applicant’s claims, Zhong discloses wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET (Abstract; Section II-A paragraph 1, wherein dynamic 18F-FDG PET corresponds to dFDG-PET).
Zhong is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to simultaneously correct for the spill-over radioactivity and partial volume effects and generate kinetic rate constants without any invasive blood sampling (Zhong, Section IV paragraph 1).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET, as taught by Zhong, into the method, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 5.
Regarding claim 6, Feng, Zhao, and Zhong teach the method of claim 1. Feng additionally discloses wherein the one or more blood vessels includes one or more carotid arteries ([0047], [0092], and [0118-0119]).
Regarding claim 10, Feng discloses a system for performing dynamic positron emission tomography (PET) (Fig. 1; [0043]), the system comprising:
a PET scanner (Fig. 1 element 110; [0043] and [0046]) configured for collecting a plurality of dynamic PET image data sets comprising volumetric radioactive measurement data associated with an administered radioactive tracer ([0063-0064], wherein obtaining module obtains data for image; [0043] and [0067], wherein volume correction of the obtained data is performed thereby necessitating volumetric data; [0086] and [0088-0089], wherein the concentration of radioactivity (i.e. volumetric radioactive measurement data, since the calculation of concentration requires volume) of the tracer (i.e. radioactive tracer) is determined) present in a target site of a subject (Fig. 4A; [0043], [0046], [0063-0064], and [0088], wherein a plurality of PET images are obtained after a tracer is injected (i.e. administered) into the subject’s (i.e. a target site, where the subject may include a specific portion/organ of the patient) blood vessels) over multiple scanning intervals (Fig. 8; [0043], [0064], [0085], and [0118]);
a computer system (Fig. 1 element 120; [0049]) comprising:
at least one processor (Fig. 2 element 210; [0055]);
a memory element (Fig. 1 element 130, Fig. 2 element 220; [0050] and [0058]);
and an image analyzer (Fig. 4A; [0044] and [0063-0064], wherein the imaging system includes modules/components for performing imaging and/or other related analysis) stored in the memory element (Fig. 2 element 210, Fig. 4A element 440; [0069]) and when executed by the at least one processor (Fig. 2 element 210; [0063]) is configured for:
obtaining the plurality of dynamic PET image data sets (Fig. 4A element 410; [0043]) from the PET scanner (Fig. 1 element 110; [0043] and [0046]).
Feng further teaches utilizing an artificial neural network (ANN) to model the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site (Fig. 9; [0064-0065], [0073], [0078], and [0119-0120], wherein a convolutional neural network (a convolutional neural network is a specific type of ANN) obtains a blood vessel model providing constraints to characteristics of the blood vessel for the second image and therefore the geometric info for the bold vessel is known). However, Feng fails to teach utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site, specifically (emphasis added). Zhao, on the other hand, teaches acquiring a segmented blood vessel from an image using an ANN. More specifically and as it relates to the applicant’s claims, Zhao discloses utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site ([0056-0058] and [0075], wherein an image segmentation unit uses an ANN to acquire a segmented blood vessel image from an original PET image).
Zhao is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to optimize the region of interest and remove unnecessary tissues or parts (Zhao, [0061-0062]).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate utilizing an artificial neural network (ANN) to segment the plurality of dynamic PET image data sets displaying one or more blood vessels in the target site, as taught by Zhao, into the system, as taught by Feng, to obtain the invention as specified in claim 10.
Feng in view of Zhao further teaches automatically deriving , using the ANN, a blood input function (IDIF) based on radioactive tracer concentrations measured in one or more segments of the plurality of dynamic PET image data sets, the blood input function being a predictive model-corrected plasma input function that is also a time activity curve (Feng [0068], [0073], [0068], [0088] and [0108], wherein a plasma input function (IDIF) is determined based on corrected fourth images, which are generated by processing an image in the image sequence based on a blood vessel model ([0108]) (i.e. models of the plurality of dynamic PET image data sets, generated using the trained machine learning model (i.e. ANN, since a machine learning model is a convolutional neural network, which is a specific type of ANN), and describes the change of the concentration of radioactivity of the tracer in the plasma with time).
Although Feng directly uses partial volume correction to compute a blood input function ([0047], [0068], and [0088], wherein the plasma input function (i.e. IDIF) is based on partial volume corrected values), Feng and Zhao do not further teach computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF, specifically (emphasis added). Zhong, on the other hand, teaches using the time activity curve of the image-derived blood input function to determine a model corrected time activity curve. More specifically, and as it relates to the applicant’s claims, Zhong discloses computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF (Section I paragraph 1, wherein the IDIF is defined to be the same as the blood pool time activity curve (TAC) and therefore the model corrected time activity curve is the same as the MCIF; Section II-D paragraphs 1 and 3; Section III paragraph 2; Fig. 3(e), wherein the IDIF/ blood pool TAC is corrected due to spill-over radioactivity and partial volume effects, producing a model corrected TAC (i.e. MCIF)).
Zhong is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to accurately and precisely determine the downstream rate of FDG influx (Zhong, Section IV paragraph 2).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate computing a predictive model-corrected blood input function (MCIF) using time activity curve input associated with the automatically derived IDIF, as taught by Zhong, into the system, as taught by Feng and Zhao, to obtain the invention as specified in claim 10.
Regarding claim 12, Feng, Zhao, and Zhong teach the system of claim 10.
Feng additionally teaches wherein deriving the IDIF includes continuously collecting the plurality of volumetric radioactive measurements at the multiple scanning intervals over a predefined time period ([0068], [0085], and [0087-0088], wherein obtaining the one or more fourth images associated with the blood vessel, which determines the plasma input function (i.e. deriving the IDIF), are determined by scanning the subject in each of the multiple consecutive time periods (i.e. continuously collecting at the multiple scanning intervals over a predefined time period), and partial volume correction is performed on each of the fourth images ([0043] and [0067], wherein volume correction of the obtained data is performed thereby necessitating volumetric radioactive measurements; [0086] and [0088-0089], wherein the concentration of radioactivity (i.e. volumetric radioactive measurement data, since the calculation of concentration requires volume) of the tracer is determined).
Regarding claim 13, Feng, Zhao, and Zhong teach the system of claim 10. Feng additionally discloses wherein prior to the obtaining step, a subject is injected with the radioactive tracer ([0043] and [0085]).
Regarding claim 14, Feng, Zhao, and Zhong teach the system of claim 10.
Feng fails to teach wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET. Zhong, on the other hand, teaches performing dynamic 18F-FDG PET. More specifically, and as it relates to the applicant’s claims, Zhong discloses wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET (Abstract; Section II-A paragraph 1, wherein dynamic 18F-FDG PET corresponds to dFDG-PET).
Zhong is combinable with Feng because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to simultaneously correct for the spill-over radioactivity and partial volume effects and generate kinetic rate constants without any invasive blood sampling (Zhong, Section IV paragraph 1).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the dynamic PET includes dynamic fluoro-2-deoxy-D-glucose (dFDG)-PET, as taught by Zhong, into the system, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 14.
Regarding claim 15, Feng, Zhao, and Zhong teach the system of claim 10. Feng additionally discloses wherein the one or more blood vessels includes one or more carotid arteries ([0047], [0092], and [0118-0119]).
Claims 2 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over Feng et al. (U.S. Publication No. US 2021/0150705 A1) ("Feng") in view of Zhao et al. (U.S. Publication No. US 2022/0198667 A1) ("Zhao") and Zhong et al. (NPL titled "Optimization of a Model Corrected Blood Input Function From Dynamic FDG-PET Images of Small Animal Heart In Vivo”) (“Zhong”), and further in view of Gallezot et al. (NPL titled “Parametric Imaging With PET and SPECT”) (“Gallezot”).
Regarding claim 2, Feng, Zhao, and Zhong teach the method of claim 1.
Although Feng teaches wherein the IDIF is used to generate objective parametric PET maps of the target site (Abstract; [0043] and [0067], wherein the plasma input function (i.e. IDIF) is used to determine more accurate PET parametric images (i.e. maps)), Feng fails to teach wherein the MCIF is used to generate objective parametric PET maps of the target site, specifically (emphasis added). Gallezot, on the other hand, teaches applying corrections to the IDIF to create the most quantitatively accurate PET parametric maps for the physiological targets. More specifically, and as it relates to the applicant’s claims, Gallezot discloses wherein the MCIF (Section III-3 paragraph 3 and section III-5 paragraph 1, wherein an IDIF is determined and correction is applied (i.e. MCIF)) is used to generate objective parametric PET maps of the target site (Section IV paragraph 1, wherein applying all appropriate corrections to the original dynamic imaging data (i.e. MCIF) generates parametric PET maps that are most quantitatively accurate for the physiological targets (i.e. objective parametric PET maps of the target site).
Gallezot is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to more precisely and accurately map the pharmacokinetics (Gallezot, section I-A paragraph 1, section II-A2 paragraph 1).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the MCIF is used to generate objective parametric PET maps of the target site, as taught by Gallezot, into the method, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 2.
Regarding claim 11, Feng, Zhao, and Zhong teach the system of claim 10.
Although Feng teaches wherein the IDIF is used to generate objective parametric PET maps of the target site (Abstract; [0043] and [0067], wherein the plasma input function (i.e. IDIF) is used to determine more accurate PET parametric images (i.e. maps)), Feng fails to teach wherein the MCIF is used to generate objective parametric PET maps of the target site, specifically (emphasis added). Gallezot, on the other hand, teaches applying corrections to the IDIF to create the most quantitatively accurate PET parametric maps for the physiological targets. More specifically, and as it relates to the applicant’s claims, Gallezot discloses wherein the MCIF (Section III-3 paragraph 3 and section III-5 paragraph 1, wherein an IDIF is determined and correction is applied (i.e. MCIF)) is used to generate objective parametric PET maps of the target site (Section IV paragraph 1, wherein applying all appropriate corrections to the original dynamic imaging data (i.e. MCIF) generates parametric PET maps that are most quantitatively accurate for the physiological targets (i.e. objective parametric PET maps of the target site).
Gallezot is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to more precisely and accurately map the pharmacokinetics (Gallezot, section I-A paragraph 1, section II-A2 paragraph 1).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the MCIF is used to generate objective parametric PET maps of the target site, as taught by Gallezot, into the system, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 11.
Claims 7, 9, 16, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Feng et al. (U.S. Publication No. US 2021/0150705 A1) ("Feng") in view of Zhao et al. (U.S. Publication No. US 2022/0198667 A1) ("Zhao") and Zhong et al. (NPL titled "Optimization of a Model Corrected Blood Input Function From Dynamic FDG-PET Images of Small Animal Heart In Vivo”) (“Zhong”), and further in view of Caleo et al. (U.S. Publication No. US 2007/0218084 A1) (“Caleo”).
Regarding claim 7, Feng, Zhao, and Zhong teach the method of claim 1.
Feng, Zhao, and Zhong fail to teach wherein the target site is a human brain. Caleo, on the other hand, discloses wherein the target site is a human brain ([0021], [0023], [0034], [0113]).
Caleo is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to allow for a precise determination of the epileptogenic focus by providing information about structural abnormalities and the underlying aetiology of seizures in the brain (Caleo, [0014]).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the target site is a human brain, as taught by Caleo, into the method, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 7.
Regarding claim 9, Feng, Zhao, and Zhong teach the method of claim 1.
Feng, Zhao, and Zhong fail to teach automatically identifying one or more seizure foci for human dynamic FDG brain PET. Caleo, on the other hand, discloses automatically identifying one or more seizure foci for human dynamic FDG brain PET ([0113]).
Caleo is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to more accurately and effectively locate the area of seizure onset for treatment or surgery (Caleo, [0120]).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate automatically identifying one or more seizure foci for human dynamic FDG brain PET, as taught by Caleo, into the method, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 9.
Regarding claim 16, Feng, Zhao, and Zhong teach the system of claim 10.
Feng, Zhao, and Zhong fail to teach wherein the target site is a human brain. Caleo, on the other hand, discloses wherein the target site is a human brain ([0021], [0023], [0034], [0113]).
Caleo is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to allow for a precise determination of the epileptogenic focus by providing information about structural abnormalities and the underlying aetiology of seizures in the brain (Caleo, [0014]).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the target site is a human brain, as taught by Caleo, into the system, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 16.
Regarding claim 18, Feng, Zhao, and Zhong teach the system of claim 10.
Feng, Zhao, and Zhong fail to teach automatically identifying one or more seizure foci for human dynamic FDG brain PET. Caleo, on the other hand, discloses automatically identifying one or more seizure foci for human dynamic FDG brain PET ([0113]).
Caleo is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to more accurately and effectively locate the area of seizure onset for treatment or surgery (Caleo, [0120]).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate automatically identifying one or more seizure foci for human dynamic FDG brain PET, as taught by Caleo, into the system, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 18.
Claims 8 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Feng et al. (U.S. Publication No. US 2021/0150705 A1) ("Feng") in view of Zhao et al. (U.S. Publication No. US 2022/0198667 A1) ("Zhao") and Zhong et al. (NPL titled "Optimization of a Model Corrected Blood Input Function From Dynamic FDG-PET Images of Small Animal Heart In Vivo”) (“Zhong”), and further in view of Xie et al. (U.S. Publication No. US 2022/0230310 A1) (“Xie”) and Thara et al. (NPL titled “Electroencephalogram for epileptic seizure detection using stacked bidirectional LSTM_GAP neural network”) (“Thara”).
Regarding claim 8, Feng, Zhao, and Zhong teach the method of claim 1.
Although Feng teaches an ANN (Feng, [0073]) and Zhao teaches an ANN performs segmentation (Zhao [0056-0058]), Feng, Zhao, and Zhong fail to teach wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation, specifically (emphasis added). Xie, on the other hand, teaches a 3D convolutional neural network that can perform end-to-end segmentation. More specifically, and as it relates to the applicant’s claims, Xie discloses wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation ([0003 and 0088], wherein a convolutional neural network is a specific type of ANN).
Xie is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to improve the practicality of obtaining size and volumetric data for quantitative protocols (Xie, [0046]).
Although Feng teaches an LSTM to create an MCIF (Feng, [0047], [0068], [0073], and [0088], wherein a machine learning model (i.e. LSTM) plasma input function based on partial volume corrected values corresponds to a predictive model-corrected blood input function (MCIF)) and Zhong teaches determining the MCIF as output directly from the IDIF (Zhong, Section II-D paragraphs 1 and 3; Section III paragraph 2; Fig. 3(e), wherein the IDIF/ blood pool TAC is corrected due to spill-over radioactivity and partial volume effects, producing a model corrected TAC (i.e. MCIF)), Feng, Zhao, and Zhong fail to teach a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer to predict the MCIF as output directly from the IDIF, specifically (emphasis added). Thara, on the other hand, teaches an LSTM trained over a time distributed dense layer to predict the output of each sample based on the output of the previous sample. More specifically, as it relates to the applicant’s claims, Thara discloses a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer (Section 2.4 paragraphs 1-2) to predict the MCIF as output directly from the IDIF (Section 2.4 paragraph 3, wherein the output of a sample (in this case, the sample output will be the MCIF determined from the IDIF as taught by Zhong) is predicted based on the output of the previous sample (in this case, the previous sample output will be the IDIF taught by Zhong)).
Thara is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to incorporate a more powerful and suitable model for time series data by remembering the computed information for a long duration (Thara, Section 2.4 paragraph 3).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation, as taught by Xie, and a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer to predict the MCIF as output directly from the IDIF into the method, as taught by Thara, into the method, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 8.
Regarding claim 17, Feng, Zhao, and Zhong teach the system of claim 10.
Although Feng teaches an ANN (Feng, [0073]) and Zhao teaches wherein the ANN performs segmentation (Zhao [0056-0058]), Feng, Zhao, and Zhong fail to teach wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation, specifically (emphasis added). Xie, on the other hand, teaches a 3D convolutional neural network that can perform end-to-end segmentation. More specifically, and as it relates to the applicant’s claims, Xie discloses wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation ([0003 and 0088], wherein a convolutional neural network is a specific type of ANN).
Xie is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to improve the practicality of obtaining size and volumetric data for quantitative protocols (Xie, [0046]).
Although Feng teaches an LSTM to create an MCIF (Feng, ([0047], [0068], [0073], and [0088], wherein a machine learning model (i.e. LSTM) plasma input function based on partial volume corrected values corresponds to a predictive model-corrected blood input function (MCIF)) and Zhong teaches determining the MCIF as output directly from the IDIF (Zhong, Section II-D paragraphs 1 and 3; Section III paragraph 2; Fig. 3(e), wherein the IDIF/ blood pool TAC is corrected due to spill-over radioactivity and partial volume effects, producing a model corrected TAC (i.e. MCIF)), Feng, Zhao, and Zhong fail to teach a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer to predict the MCIF as output directly from the IDIF, specifically (emphasis added). Thara, on the other hand, teaches an LSTM trained over a time distributed dense layer to predict the output of each sample based on the output of the previous sample. More specifically, as it relates to the applicant’s claims, Thara discloses a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer (Section 2.4 paragraphs 1-2) to predict the MCIF as output directly from the IDIF (Section 2.4 paragraph 3, wherein the output of a sample (in this case, the sample output will be the MCIF determined from the IDIF as taught by Zhong) is predicted based on the output of the previous sample (in this case, the previous sample output will be the IDIF taught by Zhong)).
Thara is combinable with Feng, Zhao, and Zhong because they are from the same art of image processing.
The suggestion/motivation for doing so would have been to incorporate a more powerful and suitable model for time series data by remembering the computed information for a long duration (Thara, Section 2.4 paragraph 3).
Therefore, it would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate wherein the ANN includes an end-to-end 3D convolutional neural network for segmentation, as taught by Xie, and a long-short-term memory (LSTM) network architecture that is trained over a time-distributed dense layer to predict the MCIF as output directly from the IDIF into the method, as taught by Thara, into the system, as taught by Feng, Zhao, and Zhong, to obtain the invention as specified in claim 17.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Georgi et al. (U.S. Publication No. US 20130261440 A1) teaches utilizing FDG-PET for generating a blood input function to produce an arterial input function as a function of the adjusted TAC after performing partial volume and spillover corrections.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RACHEL Y DANG whose telephone number is (571)438-9519. The examiner can normally be reached Monday - Thursday: 7am - 4:30pm.
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, John Villecco can be reached at (571) 272-7319. 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.
/RACHEL Y DANG/Examiner, Art Unit 2661
/XUEMEI G CHEN/Primary Examiner, Art Unit 2661
Prosecution Insights