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.
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-3, 16, 19 and 35 are rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. (US 20220108459 A1) in view of Sahbaee et al. (US 20210015438 A1).
Regarding claim 1, Lyu et al. teaches a system, comprising: at least one storage device including a set of instructions and at least one processor in communication with the at least one storage device (see para [0004]; “The system may include at least one storage device storing a set of instructions, and at least one processor configured to communicate with the at least one storage device”); wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including (see para [0022]; “When executing the executable instructions, the at least one processor may be configured to direct the system to perform one or more of the following operations”): obtaining medical images of the heart of a target subject (see para [0051]; “The imaging device 110 may be configured to acquire imaging data relating to at least part of a subject… the subject may include a specific portion, organ, and/or tissue of the patient. For example, the subject may include the head, the chest, the neck, the thorax, the heart”); the medical images being collected by performing a medical scan on the target subject when the target subject is injected with a first tracer (see para [0104]; “at least one sample first image generated based on a first tracer”, see also para [0102]; “the imaging device may scan the first sample subject to generate the first imaging data after the first tracer is injected into the first sample subject”); obtaining a target model that has been trained using training samples (see para [0004]; “The system may further generate a trained image processing model by training a preliminary model using the training samples”), each of which includes label information collected using a second tracer (see para [0016]; “generating the trained image processing model by training the second intermediate model using the at least one sample second image and the at least one reference second image”, see para [0005]; “obtaining one of the at least one reference second image generated based on second imaging data acquired by the imaging device according to the second tracer”, and para [0022]; “The trained image processing model may be provided by training a preliminary model using training samples. The training samples may include …at least one sample second image generated based on the second tracer different from the first tracer, and at least one reference second image each of which corresponds to one of the at least one sample second image and has a higher image quality than the corresponding sample second image” Note; the image higher-quality "ground truth" data implies label information). However, Lyu et al. does not teach and determining blood flow information and/or cardiac function information of the heart of the target subject based on the medical images and the target model.
In the same field of endeavor, Sahbaee et al. teaches and determining blood flow information and/or cardiac function information of the heart of the target subject based on the medical images and the target model (see claim 1; “the scanning providing coronary CT data representing contrast agent in a heart of the patient over time …estimating the myocardial perfusion by a machine-learned model having been trained by deep learning, the machine-learned model outputting the myocardial perfusion in response to input of the coronary CT data and the information”, see also para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. in order to provide perfusion imaging with less than the current, standard, or typical radiation dose (see claim 1).
Regarding claim 2, the rejection of claim 1 is incorporated herein.
Lyu et al. in the combination further teach wherein the first tracer includes fludeoxyglucose (FDG) or acetic acid; or the second tracer includes 13N-ammonia (NH3) or Rubidium 82 (Rb82) (see para [0101]; “sup.18F-FDG may be taken as an exemplary first tracer)”).
Regarding claim 3, the rejection of claim 1 is incorporated herein.
Sahbaee et al. in the combination further teach wherein when the target model is used to determine the blood flow information (see also para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”), the medical images include PET image frames that are consecutively acquired in the medical scan (see para [0030]; “scan data or medical images are obtained, such as magnetic resonance, x-ray, ultrasound, positron emission tomography (PET)”, see also para [0032]; “a sequence of frames of data or images are generated to show the perfusion or contrast agent at different times”).
Regarding claim 16, the scope of claim 16 is fully encompassed by the scope of claim 1,
accordingly, the rejection of claim 1 is fully applicable here (see also para [0051]; “the PET device may acquire imaging data of a subject after the subject is injected with a tracer” of Luy et al.).
Regarding claim 19, the rejection of claim 16 is incorporated herein.
Lyu et al. in the combination further teach wherein the determining at least two of blood flow information, cardiac function information, or myocardial metabolic information of the heart of the target subject based on the PET data of the heart of the target subject includes: generating medical images of the heart of the target subject based on the PET data of the heart of the target subject (see also para [0051]; “the subject may include the head, the chest, the neck, the thorax, the heart, …... In some embodiments, the imaging device 110 may include a positron emission tomography (PET) device”), obtaining a target model that has been trained using training samples, each of which includes label information collected using a second tracer (see para [0022]; “The trained image processing model may be provided by training a preliminary model using training samples. The training samples may include.. at least one sample second image generated based on the second tracer different from the first tracer”).
Sahbaee et al. in the combination further teach and determining the blood flow information and/or the cardiac function information of the heart of the target subject based on the medical images and the target model (see claim 1; “the scanning providing coronary CT data representing contrast agent in a heart of the patient over time …estimating the myocardial perfusion by a machine-learned model having been trained by deep learning, the machine-learned model outputting the myocardial perfusion in response to input of the coronary CT data and the information”, see also para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”).
Regarding claim 35, the scope of claim 35 is fully encompassed by the scope of claim 1, accordingly, the rejection of claim 1 is fully applicable here
Claims 4, 15 and rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Alpert et al. (US 20150230762 A1).
Regarding claim 4, the rejection of claim 3 is incorporated herein. The combination of Lyu et al. and Sahbaee et al. does not teach wherein a model input of the target model includes: segmentation images of the heart by segmenting the heart from the PET image frames; or time-activity curves of physical points at the heart.
In the same field of endeavor, Alpert et al. teach wherein a model input of the target model includes: segmentation images of the heart by segmenting the heart from the PET image frames; or time-activity curves of physical points at the heart (see para [0096]; “the ability to segment the reconstruction volume into different regions, those in which the kinetic model is valid”, see also para [0125]; “The rest and stress image volumes of the heart were segmented into 17 standard segments. Time-activity curves were calculated for each segment”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for performing a single-scan rest-stress cardiac measurement of Alpert et al. in order to achieve wide clinical utilization (see para [0096]).
Regarding claim 15, the rejection of claim 1 is incorporated herein.
Alpert et al. in the combination further teach wherein the operations further comprise: determining myocardial metabolic information of the heart by processing the medical images using a second pharmacokinetic model (see Abstract; “multi-compartment parametric model is a measure of myocardial blood flow for both a rest state and a stress state of the subject”, see also para [0017]; “the at least one kinetic parameter of blood flow comprises myocardial blood flow during both rest and stress states of the subject… the stress state is induced pharmacologically”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for performing a single-scan rest-stress cardiac measurement of Alpert et al. in order to achieve wide clinical utilization (see para [0017]).
Regarding claim 17, the rejection of claim 16 is incorporated herein.
Alpert et al. in the combination further teach wherein a first portion of the PET data for determining the blood flow information is collected in a first time period from when the tracer is injected (see para [0063]; “the part of the PET curve most highly dependent on blood flow was shown to be the initial period after IV injection of .sup.18F-tracers”, see also para [0159]; “For the first few minutes after injection, the distribution of .sup.18F-Flurpiridaz depends almost entirely on coronary blood flow”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for performing a single-scan rest-stress cardiac measurement of Alpert et al. in order to achieve wide clinical utilization (see para [0063]).
Claims 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Naghavi et al. (US 20240120095 A1).
Regarding claim 5, the rejection of claim 1 is incorporated herein. The combination of Lyu et al. and Sahbaee et al. does not teach wherein when the target model is used to determine the cardiac function information, the medical images include gated images generated based on a heartbeat signal of the heart collected in the medical scan.
In the same field of endeavor, Naghavi et al. teach wherein when the target model is used to determine the cardiac function information, the medical images include gated images generated based on a heartbeat signal of the heart collected in the medical scan (see para [0027]; “an artificial intelligence system (AIS) configured to estimate cardiac chambers volumes based on non-contrast ECG-gated CT scan images of the heart…. and determining an EF measurement based, at least in part, on the first and second measurements”, see also para [0089]; “At steps 325 and 330, measurements for end-diastole left ventricle (LV) volume (“first measurement”) and end-systole LV volume (“second measurement”) are received. In one embodiment, said measurements are based, at least in part, on the end-diastole and end-systole images (335). At step 340, an EF measurement is determined based, at least in part, on the first and second measurements”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and methods for determining an individual's risk of an adverse health outcome particularly a near-term cardiovascular event of Naghavi et al. in order to improve predictive risk assessment of adverse health outcomes in health care settings (see para [0027]).
Regarding claim 6, the rejection of claim 5 is incorporated herein.
Naghavi et al. in the combination further teach wherein a model input of the target model includes segmentation images of the heart by segmenting the heart from the gated images (see para [0062]; “Two DL models were developed for cardiac and full chest CT scan images, respectively. The models were trained to detect trabecular bone and disks with a training data set of 132 cardiac and 37 lung CT images. For ground truth, 225 cardiac and lung CT images for whole spine and disk locations were manually segmented”).
Claims 7-9 are rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Wanget al. NPL “Generation of synthetic PET images of synaptic densityand amyloid from 18 F- FDG images using deep learning”.
Regarding claim 7, the rejection of claim 1 is incorporated herein.
Lyu et al. in the combination further teach wherein each training sample of the target model further includes a training input and is determined by: obtaining first sample medical images and second sample medical images of a sample subject (see para [0111]; “The at least one sample first image and the at least one sample second image in the training samples may serve as inputs of the training process”), the first sample medical images being collected by performing a first sample medical scan on the sample subject when the sample subject is injected with the first tracer (see para [0004]; “The system may obtain training samples. The training samples may include at least one sample first image generated based on a first tracer”), and the second sample medical images being collected by performing a second sample medical scan on the sample subject when the sample subject is injected with the second tracer (see para [0116]; “the trained image processing model trained using the training samples may correspond to the second tracer”). However, the combination of Lyu et al. and Sahbaee et al. does not teach determining the label information based on the second sample medical images; and determining the training input based on the first sample medical images.
In the same field of endeavor, Wang et al. teach determining the label information based on the second sample medical images (see page 5118, left col. 1st para; “A fully convolutional layer was used to map the feature map to the label at the final stage”); and determining the training input based on the first sample medical images (see page 5118, 2.3; “where18 F- FDG SUVR images were used as the input data for Network #1”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and advanced deep learning methods using the U- Netmodel to predict 11C- UCB-J PET images of synaptic vesicle protein 2A (SV2A),a surrogate of synaptic density, from 18 F- FDG PET data of Wang et al. in order to assess the feasibility of generating synthetic PET im-ages of less-available tracers from the PET image of another common tracer, in particular 18 F- FDG (see page 5118, left col. 1st para).
Regarding claim 8, the rejection of claim 1 is incorporated herein.
Lyu et al. in the combination further teach determining a model input based on the pseudo image of each medical image (see para [0116]; “the trained image processing model trained using the training samples may correspond to the second tracer. That is, the trained image processing model may be used to process images generated based on the second tracer”).
Sahbaee et al. in the combination further teach wherein the determining blood flow information and/or cardiac function information of the heart of the target subject based on the medical images and the target model (see claim 1; “the scanning providing coronary CT data representing contrast agent in a heart of the patient over time …estimating the myocardial perfusion by a machine-learned model having been trained by deep learning, the machine-learned model outputting the myocardial perfusion in response to input of the coronary CT data and the information”, see also para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”), determining the blood flow information and/or the cardiac function information by inputting the model input into the target model (see claim 1; “estimating the myocardial perfusion by a machine-learned model having been trained by deep learning, the machine-learned model outputting the myocardial perfusion in response to input of the coronary CT data and the information”, see also para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”).
Wang et al. in the combination further teach comprises: for each of the medical images, generating a pseudo image corresponding to the second tracer by processing the medical image using an image transformation model (see page 5117, left col. 2nd para; “we explored a similar network training strategy to generate synthetic 11C- PiB static images from 18 F- FDG images”, see also page 5116, 1st para; “It is feasible to use 3D U- Net-based methods to generate syn-thetic 11C- UCB-J PET images from 18 F- FDG images with reasonable prediction accuracy”), the image transformation model being a trained machine learning model (see page 5115, methods; “We implemented advanced deep learning methods using the U- Netmodel to predict 11C- UCB-J PET images of synaptic vesicle protein 2A (SV2A),a surrogate of synaptic density, from 18 F- FDG PET data…. For 11C- UCB-J image prediction, four networkmodels were trained and tested”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and advanced deep learning methods using the U- Netmodel to predict 11C- UCB-J PET images of synaptic vesicle protein 2A (SV2A),a surrogate of synaptic density, from 18 F- FDG PET data of Wang et al. in order to assess the feasibility of generating synthetic PET im-ages of less-available tracers from the PET image of another common tracer, in particular 18 F- FDG (see page 5118, left col. 1st para).
Regarding claim 9, the rejection of claim 8 is incorporated herein.
Lyu et al. in the combination further teach wherein each training sample of the target model further includes a training input and is determined by: obtaining second sample medical images of a sample subject, the second sample medical images being collected by performing a second sample medical scan on the sample subject when the sample subject is injected with the second tracer (see para [0116]; “the trained image processing model trained using the training samples may correspond to the second tracer. That is, the trained image processing model may be used to process images generated based on the second tracer”); and determining the training input and the label information based on the second sample medical images (see para [0047]; “the plurality of reference second images may serve as reference outputs of the training process”, see also para [0122]; “The training samples may include…. at least one sample second image generated based on a second tracer different from the first tracer, and at least one reference second image each of which corresponds to one of the at least one sample second images and has a higher image quality than the corresponding sample second image”).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claims 1, and 3 above, and further in view of Alpert et al. and Bond et al. (US 20140226880 A1)
Regarding claim 10, the rejection of claim 3 is incorporated herein.
Sahbaee et al. in the combination further teach wherein the determining blood flow information and/or cardiac function information of the heart of the target subject based on the medical images and the target model comprises: determining a second portion of the blood flow information by processing the second subset using the target model (see claim 1; “estimating the myocardial perfusion by a machine-learned model having been trained by deep learning”, see also para [0007]; “the myocardial perfusion is estimated as a perfusion quantity, such as a peak, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”). However, the combination of Lyu et al. and Sahbaee et al. does not teach dividing the PET image frames into a first subset corresponding to a first stage of the medical scan and a second subset corresponding to a second stage of the medical scan, the first tracer being more concentrated in the blood pool of the heart than the myocardium of the heart in the first stage, the first tracer being more concentrated in the myocardium than the blood pool in the second stage; determining a first portion of the blood flow information by processing the first subset using at least one pharmacokinetic model;
In the same field of endeavor, Bond et al. teach dividing the PET image frames into a first subset corresponding to a first stage of the medical scan and a second subset corresponding to a second stage of the medical scan (see para [0004]; “An effect of this is that early images show the location of blood, carrying the tracer. Later images show the position of muscle tissue such as myocardium, once tracer has been absorbed into the muscle tissue…… The heart may accordingly be imaged in two distinct phases--early frames, which represent the location of volumes of blood within the heart, and late frames, which represent the location of myocardium”), the first tracer being more concentrated in the blood pool of the heart than the myocardium of the heart in the first stage (see para [0004]; “early images show the location of blood, carrying the tracer”, see also para [0033]; “the selected frame will have the highest intensity of tracer within the blood”), the first tracer being more concentrated in the myocardium than the blood pool in the second stage (see para [0004]; “Later images show the position of muscle tissue such as myocardium, once tracer has been absorbed into the muscle tissue”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for estimating an orientation of a cardiac long axis from corresponding early frame and late frame images of Bond et al. in order to assist in reorientation of a late frame myocardium image (see para [0004]). However, the combination of Lyu et al. Sahbaee et al. and Bond et al. does not teach determining a first portion of the blood flow information by processing the first subset using at least one pharmacokinetic model.
In the same field of endeavor, Alpert et al. teach determining a first portion of the blood flow information by processing the first subset using at least one pharmacokinetic model (see para [0062]; “the part of the PET curve most highly dependent on blood flow was shown to be the initial period after IV injection”, see also para [0097]; “the early frames are acquired at higher temporal frequency, when the concentrations are changing rapidly…. parametric cardiac imaging is performed on reconstructed dynamic PET frames using voxel-by-voxel (or ROI) kinetic analysis”, and para [0126]; “A region of interest was drawn over the myocardium and used to calculate the time-activity curve which was then fitted to the standard four parameter, two-compartment model”, and para [0140]; “the endpoints for analysis are PET measurements of local MBF obtained from the first 6, 10 and 60 minutes of PET scan. The 6 and 10 minute endpoints can use a simplified kinetic”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for estimating an orientation of a cardiac long axis from corresponding early frame and late frame images of Bond et al. and method for performing a single-scan rest-stress cardiac measurement of Alpert et al. in order to achieve wide clinical utilization (see para [0062]).
Claim 11 is rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. and Sahbaee et al. in view of Naghavi et al. as applied in claims 1, and 5 above, and further in view of Bond et al.
Regarding claim 11, the rejection of claim 5 is incorporated herein.
Naghavi et al. in the combination further teach the determining blood flow information and/or cardiac function information of the heart of the target subject based on the medical images and the target model comprises: for each of one or more candidate gated images of the first group of gated images, determining a left ventricular volume of the target subject corresponding to the candidate gated image (see para [0027]; “an artificial intelligence system (AIS) configured to estimate cardiac chambers volumes based on non-contrast ECG-gated CT scan images of the hear;t providing a first image comprising a non-contrast, ECG-gated CT scan image of the heart acquired during end-diastolic period; providing a second image comprising a non-contrast, ECG-gated CT scan image of the heart acquired during end-systolic period; wherein the first and second measurements are based, at least in part, on the first and second images; and calculating the difference in left ventricular volume between the first and second images; and determining an EF measurement based, at least in part, on the first and second measurements”); determining a first portion of the cardiac function information based on the one or more left ventricular volumes corresponding to one or more candidate gated images; and determining a second portion of the cardiac function information by processing the second group of gated images using the target model (see para [0028]; “the method can further include receiving from the AIS a third measurement comprising a measurement of LV wall volume and total heart volume to calculate total heart volume changes between end-diastolic period and end-systolic period”, see also para [0082]; “Retrospective ECG gating is generally used to measure cardiac function. Because images are acquired throughout the cardiac cycle, volume measurements of the right and left ventricles can be obtained in the end-systole phase (ES) and end-diastole phase (ED), allowing the calculation of stroke volume, EF, and cardiac output”).
Bond et al. in the combination further teach wherein the gated images include a first group of gated images corresponding to a first stage of the medical scan and a second group of gated images corresponding to a second stage of the medical scan (see para [0004]; “An effect of this is that early images show the location of blood, carrying the tracer. Later images show the position of muscle tissue such as myocardium, once tracer has been absorbed into the muscle tissue…… The heart may accordingly be imaged in two distinct phases--early frames, which represent the location of volumes of blood within the heart, and late frames, which represent the location of myocardium”), the first tracer being more concentrated in the blood pool of the heart than the myocardium of the heart in the first stage (see para [0004]; “early images show the location of blood, carrying the tracer”, see also para [0033]; “the selected frame will have the highest intensity of tracer within the blood”), the first tracer being more concentrated in the myocardium than the blood pool in the second stage (see para [0004]; “Later images show the position of muscle tissue such as myocardium, once tracer has been absorbed into the muscle tissue”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and methods for determining an individual's risk of an adverse health outcome particularly a near-term cardiovascular event of Naghavi et al. and a method for estimating an orientation of a cardiac long axis from corresponding early frame and late frame images of Bond et al. in order to assist in reorientation of a late frame myocardium image (see para [0004]).
Claims 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Cadieu et al. (US 20190104949 A1).
Regarding claim 12, the rejection of claim 1 is incorporated herein.
Sahbaee et al. in the combination further teachwherein the label information includes label blood flow information, the blood flow information is determined based on the medical images and the target model (see para [0044]; “Machine learning uses training data of many hundreds or thousands of samples of inputs with labeled ground truth output or outputs for each sample”, see also para [0046]; “The ground truth provided for each sample in the training data include a perfusion parametric image, color-map image, quantitative values such as peak value, time to peak, cerbal blood flow (CBF), and/or cerbal blood volume (CBV)”, and para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”). However, the combination of Lyu et al. and Sahbaee et al. does not teach, and the cardiac function information is determined based on the medical images using another target model that has been trained using training samples, each of which including label cardiac function information collected using the second tracer.
In the same field of endeavor, Cadieu et al. teaches and the cardiac function information is determined based on the medical images using another target model that has been trained using training samples, each of which including label cardiac function information collected using the second tracer (see para [0005]; “a method for artificially intelligent ejection fraction determination includes training a neural network with different sets of cardiac imaging data acquired of a ventricle for different hearts and a known ejection fraction for each of the sets”, see also para [0013]; “The sets are then supplied as training input to a neural network with respect to the associated ejection fractions so as to train the neural network”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for artificially intelligent ejection fraction determination of Cadieu et al. in order to display in a display of the computer an ejection fraction determination output by the neural network without tracing the ventricle border of the heart (see para [0005]).
Regarding claim 13, the rejection of claim 1 is incorporated herein.
Sahbaee et al. in the combination further teach and the blood flow information is determined based on the medical images using another target model that has been trained using training samples, each of which includes label blood flow information collected using the second tracer (see para [0044]; “Machine learning uses training data of many hundreds or thousands of samples of inputs with labeled ground truth output or outputs for each sample”, see also para [0046]; “The ground truth provided for each sample in the training data include a perfusion parametric image, color-map image, quantitative values such as peak value, time to peak, cerbal blood flow (CBF), and/or cerbal blood volume (CBV)”, and para [0054]; “The machine-learned model may estimate the myocardial perfusion as one or more quantities. For example, a peak value, time-to-peak, blood flow, blood volume, mean transit time, or combination thereof”).
Cadieu et al. in the combination further teach wherein the label information includes label cardiac function information, the cardiac function information is determined based on the medical images and the target model (see para [0005]; “a method for artificially intelligent ejection fraction determination includes training a neural network with different sets of cardiac imaging data acquired of a ventricle for different hearts and a known ejection fraction for each of the sets”, see also para [0013]; “The sets are then supplied as training input to a neural network with respect to the associated ejection fractions so as to train the neural network”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and a method for artificially intelligent ejection fraction determination of Cadieu et al. in order to display in a display of the computer an ejection fraction determination output by the neural network without tracing the ventricle border of the heart (see para [0005]).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Hsiao et al. (US 20230147286 A1).
Regarding claim 14, the rejection of claim 1 is incorporated herein. The combination of Lyu et al. and Sahbaee et al. does not teach wherein the label information includes label blood flow information and label cardiac function information, the target model is a multitask model, and both the blood flow information and the cardiac function information are determined based on the medical images and the multitask model.
In the same field of endeavor, Hsiao et al. teach wherein the label information includes label blood flow information and label cardiac function information (see para [0073]; “each prong's output inference is denoted as ŷ with its corresponding ground truth as y”, see also para [0058]; “regression of cardiac function parameters 410 such as stroke volume and ejection fraction”, and para [0069]; “serve as CNN blood flow synthesis ground truth”), the target model is a multitask model (see para [0022]; “multi-task hybrid CNN referred to as “Triton-Net”.. FIG. 5 shows preliminary results of automated synthesis of blood flow fields (stroke volume and ejection fraction) produced by the inventive Triton-Net-Flow hybrid architecture”), and both the blood flow information and the cardiac function information are determined based on the medical images and the multitask model (see Fig. 4, claim 4; “regression of values comprises synthesis of blood flow or myocardial velocity field”, and para [0010]; “automated measurement from biomedical images, explainable AI, quantification of cardiac function from medical images, visualization of blood flow fields from CT and more”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and method for analysis of time series images from an image source employs a 3D-UNet convolutional neural network (CNN) of Hsiao et al. in order to develop CNN architectures that overcome technical variations, uncommon diseases, and technological change (see para [0073]).
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over Lyu et al. in view of Sahbaee et al. as applied in claim 1 above, and further in view of Wang (US 20200367846 A1) and Sorensen (US 20100168554 A1).
Regarding claim 18, the rejection of claim 16 is incorporated herein. The combination of Lyu et al. and Sahbaee et al. does not teach wherein a first time period for collecting a first portion of the PET data or a second time period for collecting a second portion of the PET data are smaller than a third time period for collecting a third portion of the PET data, the first portion of the PET data being used to determine the blood flow information, the second portion of the PET data being used to determine the cardiac function information, the third portion of the PET data being used to determine the myocardial metabolic information.
In the same field of endeavor, Wang teach wherein a first time period for collecting a first portion of the PET data or a second time period for collecting a second portion of the PET data are smaller than a third time period for collecting a third portion of the PET data (see para [0030]; “PET imaging may be used to derive blood flow from the early phase (e.g., first 2 minutes) of FDG uptake…. assess glucose metabolism using PET imaging approximately forty-five minutes after administration”), the first portion of the PET data being used to determine the blood flow information (see para [0030]; “PET imaging may be used to derive blood flow from the early phase”), the third portion of the PET data being used to determine the myocardial metabolic information (see para [0069]; “The HTR early-dynamic FDG-PET method of blood flow imaging has many clinical applications. As an example, myocardial flow may be imaged in the context of myocardial viability in ischemic cardiomyopathy”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and methods are disclosed for quantifying blood flow using time-varying kinetic modeling of high temporal-resolution dynamic positron emission tomography (PET) data of Wang in order to reduce resource-intensive with long imaging time and significant cost (see para [0030]). However, the combination of Lyu et al. Sahbaee et al. and Wang does not teach the second portion of the PET data being used to determine the cardiac function information.
In the same field of endeavor, Sorensen teach the second portion of the PET data being used to determine the cardiac function information (see para [0068]; “retrieving a serial image sequence and a scanning time of the contrast media as the contrast media passes thru the heart”, see also para [0054]; “the scanning time needed to measure the MPTT is about 90 seconds”). Accordingly, it would have been obvious to one of ordinary skill in the art at before the effective filling date of the invention to modify method for image reconstruction of Lyu et al. in view of a method generates perfusion examination information from CT scans of the patient of Sahbaee et al. and methods are disclosed for quantifying blood flow using time-varying kinetic modeling of high temporal-resolution dynamic positron emission tomography (PET) data of Wang and a method suitable to generate a central circulatory turnover (CCT) index for an evaluation of cardiac function of a patient of Sorensen in order to reduce resource-intensive with long imaging time and significant cost (see para [0068]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to WINTA GEBRESLASSIE whose telephone number is (571)272-3475. The examiner can normally be reached Monday-Friday9:00-5:00.
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, Andrew Bee can be reached at 571-270-5180. 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.
/WINTA GEBRESLASSIE/Examiner, Art Unit 2677
