DETAILED ACTION
This office action is in response to the communication received on March 13, 2026 concerning application No. 18/311,338 filed on May 3, 2023.
Claims 1, 4-9, 11-12, and 14-19 are currently 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 .
Response to Arguments
Applicant's arguments filed 03/13/2026 regarding the 35 USC 112b rejection have been fully considered. The amendments to the claims have been entered and overcome the 35 USC 112b rejection of claim 1 previously set forth. The amendments do not overcome the rejections of claim 4 and 11 previously set forth, therefore those rejections stand.
Applicant's arguments filed 03/13/2026 regarding the 35 USC 112a rejection have been fully considered. The amendments to the claims have been entered and overcome the 35 USC 112a rejection of claims 1 and 11 previously set forth.
Applicant's arguments filed 03/13/2026 regarding the 35 USC 101 rejection have been fully considered, but they are not persuasive.
In response to applicant’s arguments regarding claim 1, the amendment to the claim does not require treatment to be performed on the patient. The claim recites treatment is performed only when the determined likelihood of future malignant edema occurring identifies the patient as at risk of future malignant edema. However, the claim does not perform the treatment when the patient is identified as being not at risk of future malignant edema. Therefore the particular treatment does not integrate the judicial exception into a practical application because the treatment is not required to be performed.
Regarding applicant’s arguments regarding claim 11, amending claim 11 to include “provide a recommendation to treat the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema” does not integrate the judicial exception into a practical application. A limitation cannot integrate the judicial exception into a practical application when a limitation itself is considered an abstract idea. Please see the rejection below for further details of why the limitation recited above is considered an abstract idea.
Lastly, in response to applicant’s arguments on pgs. 9-10 that the claims constitute a specific technological improvement in the field, examiner respectfully disagrees. The claims do not constitute a technological improvement in the field because the claims are currently drawn to gathering and analyzing information using conventional techniques known in the field to determine the likelihood of future malignant edema and utilizing conventional techniques in the field is not sufficient to show an improvement in the field.
Applicant's arguments filed 03/13/2026 regarding the prior art rejection have been fully considered but they are not persuasive. In response to the applicant’s arguments that the prior art fails to teach “determining a cerebrospinal fluid volume change (ΔCSF) and a CSF ratio from the CT scans, wherein determining the ΔCSF and the CSF ratio comprises extracting an image of the patient’s brain from the CT scans and identifying cerebrospinal fluid (CSF) in the extracted image of the patient’s brain in the CT scans, ΔCSF comprises a total CSF change between two temporally separated CT scans of the received CT scans, and the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in a contralateral hemisphere; executing at least one trained machine learning model to determine a likelihood of a future malignant edema occurring in the patient based at least in part on the ΔCSF and the CSF ratio; and treating the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema”, examiner respectfully disagrees. As set forth in the previous office action Foroushani teaches determining the ΔCSF and using a trained machine learning model to determine a likelihood of future malignant edema using the ΔCSF and Dhar 1 teaches determining a CSF ratio and using the CSF ratio to determine the likelihood of future malignant edema. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the determining step of Foroushani to include the CSF ratio of Dhar 1 in order to generate a method and system that relies on both the ΔCSF and CSF ratio to determine a likelihood of a future malignant edema occurring in the patient. Combining both the ΔCSF and CSF ratio would result in a more accurate determination of future malignant edema.
Further, Foroushani discloses on pg. 790, col. 2, para. 2, the final model is to, “be used to triage high-risk candidate for early aggressive medical interventions”, thereby determining which patients need to be treated. Pg. 785, Abstract, further discloses predicting which patients will develop malignant edema and performing decompressive hemicraniectomy on patients identified as having malignant edema. Therefore, Foroushani teaches “treating the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema”. For at least these reasons Foroushani in view of Dhar 1 teaches the argued limitation recited above.
For the same reasons as above, claims 4-9, 11-12, and 14-19 stand rejected in view of the prior art.
Claim Objections
Claims 4-5 and 7 are objected to because of the following informalities:
Claims 4-5 and 7, line 1, “the method of claim 2” should read “the method of claim 1”.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 4, 11-12, and 14-19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 4 recites the limitation "the volume of CSF in the contralateral hemisphere" in line 6. There is insufficient antecedent basis for this limitation in the claim. The claim does not previously recite determining a volume of CSF for the contralateral hemisphere.
Claim 11 recites the limitation "the CSF in a stroke affected hemisphere" in lines 15-16. There is insufficient antecedent basis for this limitation in the claim. The claim does not previously recite determining CSF in a stroke affected hemisphere. The claim only recites segmented CSF in the CT scans which is not the same as the CSF in the stroke affected hemisphere.
Claim 11 recites the limitation "the CSF in contralateral a hemisphere" in lines 16-17. There is insufficient antecedent basis for this limitation in the claim. The claim does not previously recite determining a CSF in a contralateral hemisphere. The claim only recites segmented CSF in the CT scans which is not the same as the CSF in the stroke affected hemisphere.
Claims dependent upon the rejected claims above, but not directly addressed, are also rejected because they inherit the indefiniteness of the claim(s) they respectively depend upon.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 4-9, 11-12, and 14-19 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea. Claim 1 recites “executing at least one trained machine learning model to determine a likelihood of future malignant edema occurring in the patient based at least in part on the ΔCSF and the CSF ratio”.
The limitation of executing at least one trained machine learning model to determine the likelihood of a future malignant edema occurring, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “machine learning model” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “machine learning model” language, “determine” in the context of this claim encompasses the user observing the ΔCSF and the CSF ratio and predicting whether malignant edema is likely to occur or not. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. The step of “determining a cerebrospinal fluid volume change (ΔCSF) and a CSF ratio of from the CT scans, wherein determining the ΔCSF and the CSF ratio comprises extracting an image of the patient’s brain from the CT scans and identifying cerebrospinal fluid (CSF) in the extracted image of the patient’s brain in the CT scans, ΔCSF comprises a total CSF change between two temporally separated CT scans of the received CT scans, and the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in a contralateral hemisphere”, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, “determining” encompasses a user viewing two different CT scans and either mentally or by hand determining a change in volume of CSF between the two scans a user either mentally or by hand identifying the CSF in each of the stroke affected hemisphere and the contralateral hemisphere and dividing the total CSF in the stoke affected hemisphere by the CSF in the contralateral hemisphere. Lastly, the step of “treating the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema”, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, a user either mentally or by hand identifies the patient as being at risk of future malignant edema or not being at risk of future malignant edema and further decides whether or not to treat the patient based on the identification. The claim as written does not currently require the treatment to actually be performed on the patient.
The judicial exception is not integrated into a practical application. The additional step of claim 1 is a receiving step. The receiving of CT scans of the patient after occurrence of a stroke amounts to data gathering recited at a high level of generality which is required to obtain the input data for the determining step. Accordingly, the additional element does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The receiving of CT scans of a patient after a stroke has been determined to be well-understood, routine, and conventional activity in the field, as recognized by multiple references within the field. Foroushani et al. (“Quantitative Serial CT Imaging-Derived Features Improve Prediction of Malignant Cerebral Edema after Ischemic Stroke”, pg. 786 discloses a CT scan was performed on the patients after experiencing a stroke at both baseline and 24-h after onset, the images where then input into an algorithm for analysis), (Dhar et al., “CSF Volumetric Analysis for Quantification of Cerebral Edema After Hemispheric Infarction”, pg. 3 discloses the CT scan was performed on the patient at baseline and follow-up following a stroke and were further input into a software for image processing and analysis), and (Dhar et al., “Application of Machine Learning to Automated Analysis of Cerebral Edema in Large Cohorts of Ischemic Stroke Patients”, Pgs. 2-3, Materials and Methods discloses CT imaging was performed post stroke and the scans were further input for processing). For these reasons, the additional step does not result in the claim, as a whole, amounting to significantly more than the judicial exception.
Claim 11 recites “input the determined ΔCSF and the CSF ratio into at least one trained machine learning model to determine a likelihood of a future malignant edema occurring in the patient”.
The limitation of input the determined ΔCSF and the CSF ratio into at least one trained machine learning model to determine a likelihood of a future malignant edema, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components. That is, other than reciting “machine learning model” nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the “machine learning model” language, “determining” in the context of this claim encompasses the user observing the ΔCSF and the CSF ratio and predicting whether malignant edema is likely to occur or not. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea. Additionally, the step of “segment the CT scans into cerebrospinal fluid (CSF)” is a process that under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, “segment” encompasses a user viewing a CT scan and mentally or by hand locating the areas within the scan that contain CSF. The step of “determine, based on the segmented CSF, a CSF volume change as a change in total CSF between two temporally separated CT scans” is a process that under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, “determine” encompasses a user viewing two different CT scans and either mentally or by hand determining a change in volume of CSF between the two scans. The step of “determine, based on the segmented CSF, the CSF in a stroke affected hemisphere of the patient’s brain and the CSF in a contralateral hemisphere of the patient’s brain; determine a CSF ratio as a ratio of the CSF in the stroke affected hemisphere divided by the CSF in the contralateral hemisphere” is a process that under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, “determine” in the context of the claim encompasses a user either mentally or by hand identifying the CSF in each of the stroke affected hemisphere and the contralateral hemisphere and dividing the total CSF in the stoke affected hemisphere by the CSF in the contralateral hemisphere. Lastly, the step of “provide a recommendation to treat the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema” is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind. For example, the user can mentally or by hand compare the likelihood of edema to potential treatment options and choose the most appropriate treatment option.
The judicial exception is not integrated into a practical application. the additional step of claim 11 is a receiving step. The receiving of CT scans of the patient after occurrence of a stroke amounts to data gathering recited at a high level of generality which is required to obtain the input data for the determining step. Additionally, the additional elements include an input, a processor for performing the determining steps, and a memory including instructions to program the processor. The input is used for data gathering and is recited at a high-level of generality which is required to obtain input data for the determining steps. The processor and memory are recited at a high level of generality (i.e., as a generic processor and memory performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Accordingly, these additionally elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. The receiving of CT scans of a patient after a stroke via an input has been determined to be well-understood, routine, and conventional activity in the field, as recognized by multiple references within the field. Foroushani et al. (“Quantitative Serial CT Imaging-Derived Features Improve Prediction of Malignant Cerebral Edema after Ischemic Stroke”, pg. 786 discloses a CT scan was performed on the patients after experiencing a stroke at both baseline and 24-h after onset, the images where then input into an algorithm for analysis), (Dhar et al., “CSF Volumetric Analysis for Quantification of Cerebral Edema After Hemispheric Infarction”, pg. 3 discloses the CT scan was performed on the patient at baseline and follow-up following a stroke and were further input into a software for image processing and analysis), and (Dhar et al., “Application of Machine Learning to Automated Analysis of Cerebral Edema in Large Cohorts of Ischemic Stroke Patients”, Pgs. 2-3, Materials and Methods discloses CT imaging was performed post stroke and the scans were further input for processing). The additional elements of having a memory for storing instructions for a processor to perform the determining steps of claim 11 amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using a generic computer component cannot provide an inventive step. For these reasons, the additional steps do not result in the claim, as a whole, amounting to significantly more than the judicial exception.
Claim 12 recites the additional abstract idea of identifying CSF in the image of the patient’s brain in the CT scan which can be done mentally or by hand. The claim does not recite any additional elements beyond these abstract ideas.
Claims 4 and 14 further limit the determining of a CSF metric step by merely specifying how the user is to determine the CSF metric. The claim does not recite additional elements.
Claims 5 and 15 recite the additional abstract ideas of extracting an image of a patient’s brain of the CT scan and identifying CSF within the extracted image which can be done mentally or by hand by viewing the images, recognizing the brain within the image and visually identifying the CSF within the images.
Claims 6 and 16 further limit the determining of a CSF metric step by merely specifying how the user is to determine the CSF metric. The claim does not recite additional elements.
Claims 7 and 17 recite the additional element of a deep learning algorithm for performing the abstract idea of claim 12. The deep learning algorithm is recited at a high level of generality (i.e., as a generic deep learning algorithm for performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Accordingly, these additionally elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The recitation of a deep learning algorithm for performing the abstract idea amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using a generic computer component cannot provide an inventive step. For these reasons, the additional steps do not result in the claim, as a whole, amounting to significantly more than the judicial exception.
Claims 8-9 and 18-19 recite the additional element of a prediction algorithm which comprises a recurrent neural network for performing the abstract idea of claim 11. The prediction algorithm is recited at a high level of generality (i.e., as a generic deep learning algorithm for performing a generic computer function) such that it amounts to no more than mere instructions to apply the exception using a generic computer component. Accordingly, these additionally elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The recitation of a prediction algorithm for performing the abstract idea amounts to no more than mere instructions to apply the exception using generic computer components. Mere instructions to apply an exception using a generic computer component cannot provide an inventive step. For these reasons, the additional steps do not result in the claim, as a whole, amounting to significantly more than the judicial exception.
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:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 4-6, 8, 11-12, 14-16, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable by Mohammadian Foroushani et al. (“Quantitative Serial CT Imaging-Derived Features Improve Prediction of Malignant Cerebral Edema after Ischemic Stroke”, As cited in Applicant’s 05/03/2023 IDS, hereinafter Foroushani) in view of Dhar et al. (“CSF Volumetric Analysis for Quantification of Cerebral Edema After Hemispheric Infarction”, as cited in Applicant’s 05/03/2023 IDS, Hereinafter Dhar 1).
Regarding claim 1, Foroushani teaches a method (pg. 786 discloses a Methods section that the “Imaging Analysis” Subsection discloses using MATLAB to perform the method) comprising:
receiving computed tomography (CT) scans of a patient after occurrence of a stroke in the patient (pg. 786, “Study Participants and Clinical Data” discloses Patients diagnosed with ischemic stroke…were enrolled” and “Imaging Analysis” discloses “Both baseline and 24-h CT imaging were processed”, therefore the CT images obtained were after occurrence of a stroke), the CT scans including a brain of the patient (pg. 786, “Imaging Analysis” discloses segmenting the brain compartments in the CT images, meaning the CT scans include a brain of the patient);
determining a cerebrospinal fluid (CSF) volume change (ΔCSF) from the CT scans (pg. 785, “Abstract” discloses “we extracted quantitative imaging features from baseline and follow-up CTs, including CSF volume”), wherein determining the ΔCSF comprises extracting an image of the patient’s brain from the CT scans and identifying CSF in the extracted image of the patient’s brain in the CT scans (pg. 786, “Imaging Analysis” discloses “extraction of the intracranial supratentorial space using k-means clustering for skull removal…these cranial regions then underwent automated segmentation into CSF and brain compartments”), ΔCSF comprises a total CSF change between two temporally separated CT scans of the received CT scans (pg. 785, “Abstract” discloses “we extracted quantitative imaging features from baseline and follow-up CTs, including CSF volume”. pg. 786, “Imaging Analysis” further discloses “percent change in CSF volume from baseline to 24-h was calculated as ΔCSF”. Therefore the change in CSF volume includes at total CSF change between two temporally separated CT scans); and
executing at least one trained machine learning model to determine a likelihood of a future malignant edema occurring in the patient based at least in part on the ΔCSF (pg. 786, col. 1, para. 2, “a multivariable machine-learning model for predicting malignant cerebral edema using a combination of clinical and quantitative imaging variable extracted from CTs at baseline and 24-h”, the determined ΔCSF is considered the quantitative imaging variable extracted used by the machine learning model); and
treating the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema (pg. 790, col. 2, para. 2, discloses the final model is to, “be used to triage high-risk candidate for early aggressive medical interventions”, thereby determining which patients need to be treated. Pg. 785, Abstract discloses predicting which patients will develop malignant edema and performing decompressive hemicraniectomy on patients identified as having malignant edema).
Foroushani does not specifically teach determining a CSF ratio from the CT scans, wherein determining the CSF ratio comprises extracting an image of the patient’s brain from the CT scans and identifying CSF in the extracted image of the patient’s brain in the CT scans, the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in a contralateral hemisphere and using the CSF ratio to determine a likelihood of a future malignant edema.
However,
Dhar 1 in a similar field of analyzing cerebral edema disclosed determining a CSF ratio (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”. IL (ipsilateral) and CL (contralateral)), wherein determining the CSF ratio comprises extracting an image of the patient’s brain from the CT scans and identifying CSF in the extracted image of the patient’s brain in the CT scans (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses outlining the CSF within the obtained CT scan), the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified” and “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”) and using the CSF ratio to determine a likelihood of a future malignant edema (pg. 8, “conclusion” discloses using the CSF values to predict malignant edema).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the determining step disclosed by Foroushani to have determined a CSF ratio and use the CSF ratio to determine a likelihood of a future malignant edema in order to provide additional metrics that can assist in predicting malignant edema, thereby making the determination more accurate, as recognized by Dhar 1 (pg. 2, para. 4-pg. 3, para. 1).
Regarding claim 4, Foroushani in view of Dhar 1 teaches the method of claim 1, as set forth above. Dhar 1 further teaches determining the CSF ratio comprises determining a midline of the extracted image of the patient's brain that defines the stroke affected hemisphere and the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Supratentorial CSF spaces [sulci and ventricles ipsilateral (IL) and contralateral (CL) to stroke, third ventricle] and basal cisterns were outlined on each slice”, by outlining the ipsilateral (affected hemisphere) and contralateral hemisphere, a midline is determined between the two hemispheres), determining a volume of CSF in the stroke affected hemisphere and the volume of CSF in the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified”), and calculating the CSF ratio as the volume of CSF in the stroke affected hemisphere divided by the volume of CSF in the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”).
Regarding claim 5, Foroushani in view of Dhar 1 teaches the method of claim 1, as set forth above. Foroushani teaches the received CT scans comprise a first CT scan and a second CT scan, the first CT scan being acquired at an earlier time than the second CT scan (pg. 786, “Imaging Analysis” discloses “both baseline and 24-h CT imaging were processed”, the baseline CT scan was acquired earlier than the 24-h CT scan which was acquired at follow-up), extracting an image of the patient's brain from the CT scans comprises extracting a first image of the patient's brain from the first CT scan and extracting a second image of the patient's brain from the second CT scan (pg. 786, “Imaging Analysis” discloses “both baseline and 24-h CT imaging were processed…this included first extraction of the intracranial supratentorial space…for skull removal”, therefore both the baseline and 24-h CT scans had images of the patient’s brain extracted), and identifying CSF in the extracted image of the patient's brain in the CT scans includes identifying a first volume of CSF from the first image of the patient's brain and a second volume of CSF from the second image of the patient's brain (pg. 786, “Imaging Analysis” discloses “these cranial regions then underwent automated segmentation into CSF and brain compartments”, the segmentation of the CSF in the baseline scan represents the first volume of the CSF and segmentation in the 24-H CT scan represents the second volume of CSF).
Regarding claim 6, Foroushani in view of Dhar 1 teaches the method of claim 5, as set forth above. Foroushani further teaches determining ΔCSF comprises subtracting the first volume of CSF from the second volume of CSF (pg. 786, “Imaging Analysis” discloses “percent change in CSF volume from baseline to 24-h was calculated as ΔCSF”. The formula for percent change is (X1-X2)/X1, therefore to calculate ΔCSF the first volume is subtracted from the second volume).
Regarding claim 8, Foroushani in view of Dhar 1teaches the method of claim 1, as set forth above. Foroushani further teaches determining a likelihood of a future malignant edema occurring in the patient comprises inputting the ΔCSF into a prediction algorithm (pg. 786, col. 1, para. 2, “a multivariable machine-learning model for predicting malignant cerebral edema using a combination of clinical and quantitative imaging variable extracted from CTs at baseline and 24-h”, the machine-learning model is considered the predictive algorithm and the determined ΔCSF is considered the quantitative imaging variable extracted used by the machine learning model).
Dhar 1 further teaches using the CSF ratio to determine a likelihood of a future malignant edema (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified” and “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”. pg. 8, “conclusion” discloses using the CSF values which comprise the ratio of the volumes to predict malignant edema).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the determining step disclosed by Foroushani in view of Dhar 1 to use the CSF ratio to determine a likelihood of future malignant edema in order to provide additional metrics that can assist in predicting malignant edema, thereby making the determination more accurate, as recognized by Dhar 1 (pg. 2, para. 4-pg. 3, para. 1).
Regarding claim 11, Foroushani teaches a system for determining a likelihood of future malignant edema occurring in a patient (pg. 786, col. 1, para. 2, “a multivariable machine-learning model for predicting malignant cerebral edema using a combination of clinical and quantitative imaging variable extracted from CTs at baseline and 24-h”. pg. 786 discloses a Methods section that the “Imaging Analysis” Subsection discloses used MATLAB, which is a computer program. The computer that performs the machine-learning model and imaging analysis is considered the system), the system comprising:
an input (pg. 786, “Methods” discloses obtaining imaging of the patients which are CT scans, the device used to obtain the CT scans is considered the input);
a processor coupled to the input (pgs. 786-787, “Imaging Analysis”, “machine-learning model building” and “model performance” disclose the scans were processed using MATLAB which is a computer program and a machine learning model is built, the computer that runs the machine learning model and MATLAB is considered the processor); and
a memory coupled to the processor (pgs. 786-787, “Imaging Analysis”, “machine-learning model building” and “model performance” disclose using MATLAB and testing the machine learning model, the part of the system where the MATLAB algorithms and machine learning model are stored is considered the memory. The algorithms and model are considered the instructions), the memory including instructions that program the processor to:
receive, through the input, computed tomography (CT) scans of a patient after occurrence of a stroke in the patient (pg. 786, “Study Participants and Clinical Data” discloses Patients diagnosed with ischemic stroke…were enrolled” and “Imaging Analysis” discloses “Both baseline and 24-h CT imaging were processed”, therefore the CT images obtained were after occurrence of a stroke), the CT scans including a brain of the patient (pg. 786, “Imaging Analysis” discloses segmenting the brain compartments in the CT images, meaning the CT scans include a brain of the patient);
segment the CT scans into cerebrospinal fluid (CSF) (pg. 786, Imaging Analysis, discloses the cranial regions of the CT images underwent automated segmentation into CSF and brain compartments);
determine, based on the segmented CSF, a CSF volume change (ΔCSF) as a change in total CSF between two temporally separated CT scans of the received CT scans (pg. 785, “Abstract” discloses “we extracted quantitative imaging features from baseline and follow-up CTs, including CSF volume”. pg. 786, “Imaging Analysis” further discloses “percent change in CSF volume from baseline to 24-h was calculated as ΔCSF”. Therefore the change in CSF volume includes at total CSF change between two temporally separated CT scans);
input the determined ΔCSF into at least one trained machine learning model to determine a likelihood of a future malignant edema occurring in the patient (pg. 786, col. 1, para. 2, “a multivariable machine-learning model for predicting malignant cerebral edema using a combination of clinical and quantitative imaging variable extracted from CTs at baseline and 24-h”, the determined ΔCSF is considered the quantitative imaging variable extracted and input into the machine learning model); and
provide a recommendation to treat the patient using at least one of decompressive hemicraniectomy or osmotic therapy when the determined likelihood of future malignant edema occurring identifies the patient as at risk of a future malignant edema (pg. 790, col. 2, para. 2, discloses the final model is to, “be used to triage high-risk candidate for early aggressive medical interventions”, thereby providing a treatment that should be performed. Pg. 785, Abstract, discloses patients predicted to develop malignant edema are treated with decompressive hemicraniectomy).
Foroushani does not specifically teach determine, based on the segmented CSF, the CSF in a stroke affected hemisphere of the patient’s brain and the CSF in a contralateral hemisphere of the patient’s brain; and determining a CSF ratio from the CT scans, wherein the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in contralateral hemisphere and using the CSF ratio to determine a likelihood of a future malignant edema.
However,
Dhar 1 in a similar field of analyzing cerebral edema disclosed determine, based on the segmented CSF (pg. 2, Introduction Para. 3, discloses the CSF is segmented from CT data), the CSF in a stroke affected hemisphere of the patient’s brain and the CSF in a contralateral hemisphere of the patient’s brain (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified”, where the compartments include contralateral hemisphere and stroke affected hemisphere); and
determining a CSF ratio (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”. IL (ipsilateral) and CL (contralateral)), wherein the CSF ratio comprises a ratio of the CSF in a stroke affected hemisphere divided by the CSF in contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified” and “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”) and using the CSF ratio to determine a likelihood of a future malignant edema (pg. 8, “conclusion” discloses using the CSF values to predict malignant edema).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the determining step disclosed by Foroushani to have determined a CSF ratio and use the CSF ratio to determine a likelihood of a future malignant edema in order to provide additional metrics that can assist in predicting malignant edema, thereby making the determination more accurate, as recognized by Dhar 1 (pg. 2, para. 4-pg. 3, para. 1).
Regarding claim 12, Foroushani in view of Dhar 1 teaches the system of claim 11, as set forth above. Foroushani teaches the instructions program the processor to extract an image of the patient’s brain from the CT scans and identify CSF in the extracted image of the patient’s brain in the CT scans (pg. 786, “Imaging Analysis” discloses “extraction of the intracranial supratentorial space using k-means clustering for skull removal…these cranial regions then underwent automated segmentation into CSF and brain compartments”).
Regarding claim 14, Foroushani in view of Dhar 1 teaches the system of claim 12, as set forth above. Dhar 1 further teaches the instructions program the processor to determine a CSF ratio by determining a midline of the extracted image of the patient's brain that defines the stroke affected hemisphere and the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Supratentorial CSF spaces [sulci and ventricles ipsilateral (IL) and contralateral (CL) to stroke, third ventricle] and basal cisterns were outlined on each slice”, by outlining the ipsilateral (affected hemisphere) and contralateral hemisphere, a midline is determined between the two hemispheres), determine a volume of CSF in the stroke affected hemisphere and a volume of CSF in the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified”), and calculate the CSF ratio as the volume of CSF in the stroke affected hemisphere divided by the volume of CSF in the contralateral hemisphere (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”).
Regarding claim 15, Foroushani in view of Dhar 1 teaches the system of claim 12, as set forth above. Foroushani teaches the received CT scans comprise a first CT scan and a second CT scan, the first CT scan being acquired at an earlier time than the second CT scan (pg. 786, “Imaging Analysis” discloses “both baseline and 24-h CT imaging were processed”, the baseline CT scan was acquired earlier than the 24-h CT scan which was acquired at follow-up),
extract an image of the patient's brain from the CT scans by extracting a first image of the patient's brain from the first CT scan and extracting a second image of the patient's brain from the second CT scan (pg. 786, “Imaging Analysis” discloses “both baseline and 24-h CT imaging were processed…this included first extraction of the intracranial supratentorial space…for skull removal”, therefore both the baseline and 24-h CT scans had images of the patient’s brain extracted), and
identify CSF in the extracted image of the patient's brain in the CT scans by identifying a first volume of CSF from the first image of the patient's brain and a second volume of CSF from the second image of the patient's brain (pg. 786, “Imaging Analysis” discloses “these cranial regions then underwent automated segmentation into CSF and brain compartments”, the segmentation of the CSF in the baseline scan represents the first volume of the CSF and segmentation in the 24-H CT scan represents the second volume of CSF).
Regarding claim 16, Foroushani in view of Dhar 1 teaches the system of claim 15, as set forth above. Foroushani further teaches the instructions program the processor to determine ΔCSF by subtracting the first volume of CSF from the second volume of CSF (pg. 786, “Imaging Analysis” discloses “percent change in CSF volume from baseline to 24-h was calculated as ΔCSF”. The formula for percent change is (X1-X2)/X1, therefore to calculate ΔCSF the first volume is subtracted from the second volume).
Regarding claim 18, Foroushani in view of Dhar 1teaches the system of claim 11, as set forth above. Foroushani further teaches the instructions program the processor to determine a likelihood of a future malignant edema occurring in the patient by processing the ΔCSF with a prediction algorithm stored in the memory (pg. 786, col. 1, para. 2, “a multivariable machine-learning model for predicting malignant cerebral edema using a combination of clinical and quantitative imaging variable extracted from CTs at baseline and 24-h”, the machine-learning model is considered the predictive algorithm and the determined ΔCSF is considered the quantitative imaging variable extracted used by the machine learning model).
Dhar 1 further teaches using the CSF ratio to determine a likelihood of a future malignant edema (Pg. 3, “Imaging Processing and Volumetric Analysis” discloses “the volumes of each compartment and total volume of CSF were then quantified” and “Hemispheric and Sulcal Symmetry was calculated as the ratio of IL vs. CL CSF volumes”. pg. 8, “conclusion” discloses using the CSF values which comprise the ratio of the volumes to predict malignant edema).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the determining step disclosed by Foroushani in view of Dhar 1 to use the CSF ratio to determine a likelihood of future malignant edema in order to provide additional metrics that can assist in predicting malignant edema, thereby making the determination more accurate, as recognized by Dhar 1 (pg. 2, para. 4-pg. 3, para. 1).
Claim(s) 7 and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Foroushani in view of Dhar 1 as applied to claims 1 and 12 above, and further in view of Dhar et al (“Application of Machine Learning to Automated Analysis of Cerebral Edema in Large Cohorts of Ischemic Stroke Patients”, as cited in Applicant’s 05/03/2023 IDS, hereinafter Dhar 2).
Regarding claim 7, Foroushani in view of Dhar 1 teaches the method of claim 1, as set forth above Foroushani in view of Dhar 1 does not specifically teach identifying CSF in the extracted image of the patient's brain comprises processing the extracted image of the patient's brain with a deep learning algorithm trained to perform CSF segmentation.
However,
Dhar 2 in a similar field of analyzing cerebral edema discloses identifying CSF in the extracted image of the patient's brain comprises processing the extracted image of the patient's brain with a deep learning algorithm trained to perform CSF segmentation (pg. 1, Abstract discloses “a machine learning algorithm capable of segmenting and measuring CSF volume from serial CT scans of stroke patients”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the CSF identification disclosed by Foroushani in view of Dhar 1 to have identified CSF in the extracted image of the patient's brain by processing the extracted image of the patient's brain with a deep learning algorithm trained to perform CSF segmentation in order to more rapidly and reliably measure CSF volume on the CT scans, as recognized by Dhar 2 (pg. 2, col. 2, para. 2).
Regarding claim 17, Foroushani in view of Dhar 1 teaches the system of claim 12, as set forth above. Foroushani in view of Dhar 1 does not specifically teach the instructions program the processor to identify CSF in the extracted image of the patient's brain by processing the extracted image of the patient's brain with a deep learning algorithm stored in the memory trained to perform CSF segmentation.
However,
Dhar 2 in a similar field of analyzing cerebral edema discloses identifying CSF in the extracted image of the patient's brain by processing the extracted image of the patient's brain with a deep learning algorithm trained to perform CSF segmentation (pg. 1, Abstract discloses “a machine learning algorithm capable of segmenting and measuring CSF volume from serial CT scans of stroke patients”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the CSF identification disclosed by Foroushani in view of Dhar 1 to have identified CSF in the extracted image of the patient's brain by processing the extracted image of the patient's brain with a deep learning algorithm trained to perform CSF segmentation in order to more rapidly and reliably measure CSF volume on the CT scans, as recognized by Dhar 2 (pg. 2, col. 2, para. 2).
Claim(s) 9 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Foroushani in view of Dhar 1 as applied to claims 8 and 19 above, and further in view of Lee et al. (US 20230411018, hereinafter Lee).
Regarding claim 9, Foroushani in view of Dhar 1 teaches the method of claim 8, as set forth above. As discussed above Foroushani in view of Dhar 1 teaches it would have been obvious to input the ΔCSF and the CSF ratio into a prediction algorithm.
Foroushani in view of Dhar 1 does not specifically teach inputting the ΔCSF and the CSF ratio into a recurrent neural network employing a long short-term memory (LSTM) architecture.
However,
Lee in a similar field of predicting the occurrence of disease discloses inputting at least one metric into a prediction algorithm comprises inputting the at least one metric into a recurrent neural network employing a long short-term memory (LSTM) architecture ([0071] discloses “a recurrent neural network (RNN)” for example “a long short-term memory (LTSM) network has been proposed” for prediction. [0116] discloses “a processor calculates disease prediction information by using a long short-term memory (LSTM) based on health data and comparison information”, the health data and comparison information is considered the metric input to obtain the prediction).
The invention, specifically the prediction algorithm of Foroushani in view of Dhar 1 can be considered a “base” device (method) upon which the claimed invention can be seen as an “improvement”. The invention of Lee discussed above can be considered a “comparable” device such as a neural network that has been improved in the same way as the claimed invention by having the prediction algorithm comprise a recurrent neural network employing a long short-term memory (LSTM) architecture. One of ordinary skill in the art would have applied the known improvement of having the prediction algorithm comprise a recurrent neural network employing a long short-term memory (LSTM) architecture to the “base” device because the improvements allow for a result that would have been predictable to one of ordinary skill in the art. For example, a more robust neural network that would more accurately predict the occurrence of disease. The motivation to apply the known improvement above to the method of Foroushani in view of Dhar 1 would be to allow for the use of the known technique recited above to improve similar devices in the same way.
Regarding claim 19, Foroushani in view of Dhar 1 teaches the system of claim 18, as set forth above. Foroushani in view of Dhar 1 does not specifically teach the prediction algorithm comprises a recurrent neural network employing a long short-term memory (LSTM) architecture.
However,
Lee in a similar field of predicting the occurrence of disease discloses a prediction algorithm comprising a recurrent neural network employing a long short-term memory (LSTM) architecture ([0071] discloses “a recurrent neural network (RNN)” for example “a long short-term memory (LTSM) network has been proposed” for prediction. [0116] discloses “a processor calculates disease prediction information by using a long short-term memory (LSTM) based on health data and comparison information”, the health data and comparison information is considered the metric input to obtain the prediction).
The invention, specifically the prediction algorithm of Foroushani in view of Dhar 1 can be considered a “base” device upon which the claimed invention can be seen as an “improvement”. The invention of Lee discussed above can be considered a “comparable” device such as a neural network that has been improved in the same way as the claimed invention by having the prediction algorithm comprise a recurrent neural network employing a long short-term memory (LSTM) architecture. One of ordinary skill in the art would have applied the known improvement of having the prediction algorithm comprise a recurrent neural network employing a long short-term memory (LSTM) architecture to the “base” device because the improvements allow for a result that would have been predictable to one of ordinary skill in the art. For example, a more robust neural network that would more accurately predict the occurrence of disease. The motivation to apply the known improvement above to the system of Foroushani in view of Dhar 1 would be to allow for the use of the known technique recited above to improve similar devices in the same way.
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 ANDREW BEGEMAN whose telephone number is (571)272-4744. The examiner can normally be reached Monday-Thursday 8:30-5:00.
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, Keith Raymond can be reached at 5712701790. 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.
/ANDREW W BEGEMAN/Examiner, Art Unit 3798
Prosecution Insights