DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
In response to amendments, filed March 12, 2026, claims 1, 12, and 16 have been amended. Claims 4, 6, and 18 have been cancelled. Claims 21-23 have been added. Claims 1-3, 5, 7-17, and 19-23 are pending.
Response to Arguments
Applicant’s arguments, see Remarks, filed March 12, 2026, with respect to the 35 USC 101 rejections have been fully considered and are persuasive in view of the amendments. The rejections under 35 USC 101 have been withdrawn.
Applicant's arguments filed March 12, 2026 have been fully considered but they are not persuasive. In response to applicant’s argument that there is no teaching, suggestion, or motivation to combine the references, the examiner recognizes that obviousness may be established by combining or modifying the teachings of the prior art to produce the claimed invention where there is some teaching, suggestion, or motivation to do so found either in the references themselves or in the knowledge generally available to one of ordinary skill in the art. See In re Fine, 837 F.2d 1071, 5 USPQ2d 1596 (Fed. Cir. 1988), In re Jones, 958 F.2d 347, 21 USPQ2d 1941 (Fed. Cir. 1992), and KSR International Co. v. Teleflex, Inc., 550 U.S. 398, 82 USPQ2d 1385 (2007). In this case, including mean arterial pressure (MAP) in the determination of ICP as disclosed in Mourad to enable accurate determination of ICP using non-invasively or minimally invasively data measured on an intermittent or on a substantially continuous basis (Mourad [0064]), which alone is an appropriate motivation to combine. Additionally, including MAP data per Mourad together with the optical data in Kyriacou would improve accuracy of the ICP determination as both methods (determination using optical data and MAP data, respectively) are successful/accurate and therefore, there not only would be a reasonable expectation of success but an improvement in accuracy as well.
In response to applicant's argument that the examiner's conclusion of obviousness is based upon improper hindsight reasoning, it must be recognized that any judgment on obviousness is in a sense necessarily a reconstruction based upon hindsight reasoning. But so long as it takes into account only knowledge which was within the level of ordinary skill at the time the claimed invention was made, and does not include knowledge gleaned only from the applicant's disclosure, such a reconstruction is proper. See In re McLaughlin, 443 F.2d 1392, 170 USPQ 209 (CCPA 1971). The claim language, in describing inputting optical data and MAP data into at least one trained machine learning model and outputting ICP based on the optical and MAP data, is broad and lacks specificity that would require disclosure of any particular non-linear relationship between ICP, MAP, and the shape features of the waveform data that applicant argues would require hindsight reasoning. Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Therefore, Kyriacou and Mourad have the necessary motivation to combine the respective teachings with an at least reasonable expectation of success, and it would have been obvious to one of ordinary skill in the art before the effective date of the claimed invention to have modified the method of Kyriacou to include MAP data in the determination of ICP as disclosed in Mourad.
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.
Claim(s) 1-3, 5, 7-17, and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kyriacou (US 20220409080 A1) in view of Mourad (US 20050015009 A1) and Shamim (US 20230001164 A1).
Regarding claim 1, Kyriacou teaches a computer-implemented method (Claim 25, “A method of non-invasively predicting intracranial pressure in-vivo…”) comprising:
generating, with at least one processor (controller 11, processor 15), first waveform data using near-infrared spectroscopy (NIRS) to measure at least one light-based signal in each patient of a plurality of patients, wherein the first waveform data comprises a plurality of waveforms ([0034] “the sensor contains an array of four optical sources that are capable of emitted light of different wavelengths. In this arrangement the sources comprise light emitting diodes emitting light of four different wavelengths in the near infrared/infrared.” [0073] “The resultant data are then split into training and test data (at a discretionary ratio), following common Machine Learning techniques. The training data (usually 70-80% of total data) are used to train the prediction model, whereas the remaining data is held to later validate the performance of the model.” [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose.”), and wherein each waveform of the plurality of waveforms is associated with at least one blood attribute ([0077] “One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”);
training, with at least one processor, at least one machine learning model based on the first waveform data to produce at least one trained machine learning model, wherein the at least one trained machine learning model is configured to generate an output of intracranial pressure (ICP) based on one or more waveforms associated with the at least one blood attribute that is input to the at least one trained machine learning model ([0068] “In another envisaged arrangement depicted schematically in FIG. 7, an algorithm based on, for instance, multiple variable regression may be employed. Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor). The weights of the inputs of the model are adjusted during training to produce the lowest error (cost function) between the predicted and target ICP values.” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features). The prediction model will try to fit the response variable and predictors whilst minimising the error of prediction. In one illustrative implementation, the model is based on Support Vector Machines algorithms. However, another model among the many available, for instance Neural Network, could also be utilised.”);
generating, with at least one processor, second waveform data using NIRS to measure at least one light-based signal in a first patient, wherein the second waveform data comprises at least one waveform associated with the at least one blood attribute ([0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”); and
determining, with at least one processor, an estimated ICP in the first patient based on the at least one trained machine learning model ([0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”), wherein determining the estimated ICP in the first patient based on the at least one trained machine learning model comprises:
inputting at least a portion of the second waveform data to the at least one trained machine learning model; and generating an output from the at least one trained machine learning model comprising the estimated ICP based on at least one shape feature of the at least one waveform of the second waveform data ([0016] “the system further comprising a modelling module for establishing a model relating features of an optical signal to intracranial pressure, the optical signal being representative of a degree to which light input into a subject's skull is absorbed by the subject's brain; a feature extraction module operable to extract one or more signal features from an absorbance signal derived from the optical signals output by said detectors; and an intracranial pressure prediction module operable to input said signal features into said model and output an indication of intracranial pressure in accordance with said model”);
determining, with at least one processor (controller 11, processor 15).
However, Kyriacou fails to disclose using mean arterial pressure (MAP) to determine ICP.
Mourad teaches systems and methods for determining ICP based on parameters that can be measured using non-invasive or minimally invasive techniques. The combination of Kyriacou/Mourad discloses mean arterial pressure (MAP) data of the first patient, wherein determining the estimated ICP in the first patient based on the at least one trained machine learning model (Mourad: [0190] “taking measurements from a variety of patients under various circumstances, and the determination of displacement and ABP [arterial blood pressure] can then be used to calculate ICP, where ICP=F.sub.2(d)-MAP.” [0064] “A neural network, set up and trained;” Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”) further comprises:
inputting the MAP data to the at least one trained machine learning model; and generating the output from the at least one trained machine learning model comprising the estimated ICP based on the MAP data and the at least one shape feature of the at least one waveform of the second waveform data (Mourad: [0064] “A neural network, set up and trained as described below, was used to derive a non-linear relationship, which is further characterized below, and which provided accurate determinations of ICP in experimental protocols based on two variable parameters: V_mca measurements taken using TCD techniques; and ABP measured;” [0192] “This method uses a derived relationship between the amplitude of reflected acoustic signal(s) from CNS target tissue sites, ABP, and invasively monitored ICP to estimate ICP from non-invasively measured acoustic signals and ABP… Since the backscatter signal is related to the arterial pulse wave, a can be normalized to the MAP (as defined above), to produce a waveform .beta.. The relationship between this normalized waveform, .beta., and invasively measured ICP is then determined by taking simultaneous measurements of the backscatter signal, ABP, and ICP and solving for the equation.” Kyriacou: [0068] “Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor).” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).” [0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”).
Therefore, 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 method of Kyriacou to include mean arterial pressure (MAP) in the determination of ICP as disclosed in Mourad to accurately determine ICP using data measured non-invasively or minimally invasively on an intermittent or on a substantially continuous basis (Mourad [0064]).
While Kyriacou discloses in [0013] that in conventional approaches, “treatment is usually initiated if ICP increases >20 mm Hg,” the combination of Kyriacou/Mourad fails to particularly disclose performing treatment.
Shamim teaches a method for monitoring cerebrospinal fluid (CSF) in a ventriculo-peritoneal (VP) shunt implanted in a human patient and draining excess cerebrospinal fluid from the human patient's brain to prevent the intracranial pressure from getting too high. The combination of Kyriacou/Mourad/Shamim discloses and performing, with at least one processor, at least one treatment for the first patient based on the estimated ICP (Kyriacou: [0039] “provide warnings (for example, a visual and or sonic warning) in the event that nICP changes, for example exceeds a predetermined threshold.” Shamim: [0030] “The VP shunt 202 is a thin tube through which excess CSF is drained to prevent pressure from getting too high (e.g., higher than a threshold value) in the brain 106. The shunt 202 is placed under the skin.” [0031] “a controller that can control the activation or deactivation of the pump based on such determination of whether there is excess CSF in the brain 106.”).
Therefore, 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 combination of Kyriacou/Mourad to perform treatment based on the ICP as disclosed in Shamim to treat hydrocephalus and prevent pressure in the brain from getting too high (Shamim [0004, 0030]).
Regarding claim 2, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein generating the first waveform data (Kyriacou: [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose.”) further comprises:
removing, with at least one processor (Kyriacou: controller 11, processor 15), data outliers from the first waveform data using a preprocessing technique comprising at least one of the following: normalization (Kyriacou: [0070] “After filtering out noise, the pulsatile signals are then normalised by their DC baseline values (also filtered from the raw optical signals by low-pass filters). This normalisation process enables the ability to take into account of the proportion of total light absorbed, which may differ within patients, wavelengths or within measurements.”).
Regarding claim 3, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, further comprising: comparing, with at least one processor, the estimated ICP to at least one predetermined threshold ICP; and in response to the estimated ICP satisfying the at least one predetermined threshold ICP, generating, with at least one processor (Kyriacou: controller 11, processor 15), at least one alert to a computing device associated with a healthcare personnel providing care to the first patient (Kyriacou: [0039] “In an envisaged arrangement, the controller is coupled to a display for displaying a generated measure of nICP to clinicians, and may be configured to provide warnings (for example, a visual and or sonic warning) in the event that nICP changes, for example exceeds a predetermined threshold.”).
Regarding claim 5, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein the at least one machine learning model (Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”) comprises a random forest model (Shamim: [0036] “The computing server 304 can input the data received from the device 302 to a trained machine learning model to generate the insights.” [0038] “random forest model (e.g., model that involves creating multiple decision trees using bootstrapped datasets of the original data and randomly selecting a subset of variables at each step of the decision tree, where this model advantageously reduces the risk of error from an individual tree)”).
Regarding claim 7, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein the at least one shape feature of the at least one waveform comprises at least one of the following: area under the curve (AUC), x-coordinate of the center of mass (COMX), y-coordinate of the center of mass (COMy), peak height, peak width, peak location, or any combination thereof (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a).
Regarding claim 8, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 7, wherein generating the output comprising the estimated ICP further comprises: generating the output from the at least one trained machine learning model comprising the estimated ICP based on the at least one shape feature of the at least one waveform of the second waveform data (Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”), wherein the at least one shape feature comprises a plurality of different shape features (Kyriacou: [0016] “a feature extraction module operable to extract one or more signal features from an absorbance signal derived from the optical signals output by said detectors;” [0042] “Several features are extracted from each pulse waveform and averaged over a rolling-window of, in this particular implementation, approximately 15 seconds.” [043-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a).
Regarding claim 9, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein generating the first waveform data (Kyriacou: Figs. 2-4) further comprises:
generating, with at least one processor (Kyriacou: controller 11, processor 15), a subset of the plurality of waveforms for each patient of the plurality of patients using NIRS to measure a plurality of consecutive cardiac pulses (Kyriacou: [0071] “The feature extraction can be performed in a rolling window with no prefixed length. As changes in the intracranial pressure (absolute) values are monitored for rather longer period, this rolling window's length can be from few seconds up to 2 minutes. Once the features are extracted within the rolling window, considering that a rolling window will contain more than one pulse, the features extracted for each pulse can be averaged across the number of pulses (i.e. one feature value for each rolling window).”); and
determining, with at least one processor, an average cardiac waveform (ACPW) for said each patient based on the subset of the plurality of waveforms (Kyriacou: [0042] “Several features are extracted from each pulse waveform and averaged over a rolling-window” [0072] “After feature extraction and averaging within each rolling window, the extracted feature(s) are passed to a post-processing stage.” [0087] “Heart rate—obtained from the inter-beat intervals averaged over the sampling window.”).
Regarding claim 10, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 9, wherein the plurality of consecutive cardiac pulses numbers in a range of 60 to 120 consecutive cardiac pulses (Kyriacou: [0071] “The feature extraction can be performed in a rolling window with no prefixed length. As changes in the intracranial pressure (absolute) values are monitored for rather longer period, this rolling window's length can be from few seconds up to 2 minutes.” As normal resting heart rate should be between 60 to 100 beats per minute, the rolling window of a few second to 2 minutes would contain the range of 60-120 consecutive cardiac pulses).
Regarding claim 11, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein the at least one blood attribute comprises at least one of the following: change in oxygenated hemoglobin concentration (AHbO), change in total hemoglobin concentration (AHbT), or any combination thereof (Kyriacou: [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”).
Regarding claim 12, Kyriacou teaches a system comprising at least one processor programmed or configured ([0079] “The controller 11 further comprises read only memory (ROM) and/or random access memory (RAM) 19 that provides a processing environment in which the processor 15 can execute computer programs.”) to:
generate first waveform data using near-infrared spectroscopy (NIRS) to measure at least one light-based signal in each patient of a plurality of patients, wherein the first waveform data comprises a plurality of waveforms ([0034] “the sensor contains an array of four optical sources that are capable of emitted light of different wavelengths. In this arrangement the sources comprise light emitting diodes emitting light of four different wavelengths in the near infrared/infrared.” [0073] “The resultant data are then split into training and test data (at a discretionary ratio), following common Machine Learning techniques. The training data (usually 70-80% of total data) are used to train the prediction model, whereas the remaining data is held to later validate the performance of the model.” [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose.”), and wherein each waveform of the plurality of waveforms is associated with at least one blood attribute ([0077] “One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”);
train at least one machine learning model based on the first waveform data to produce at least one trained machine learning model, wherein the at least one trained machine learning model is configured to generate an output of intracranial pressure (ICP) based on one or more waveforms associated with the at least one blood attribute that is input to the at least one trained machine learning model ([0068] “In another envisaged arrangement depicted schematically in FIG. 7, an algorithm based on, for instance, multiple variable regression may be employed. Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor). The weights of the inputs of the model are adjusted during training to produce the lowest error (cost function) between the predicted and target ICP values.” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features). The prediction model will try to fit the response variable and predictors whilst minimising the error of prediction. In one illustrative implementation, the model is based on Support Vector Machines algorithms. However, another model among the many available, for instance Neural Network, could also be utilised.”);
generate second waveform data using NIRS to measure at least one light-based signal in a first patient, wherein the second waveform data comprises at least one waveform associated with the at least one blood attribute ([0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”);
determine an estimated ICP in the first patient based on the at least one trained machine learning model, wherein, while determining the estimated ICP in the first patient based on the at least one trained machine learning model ([0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”), the at least one processor is further programmed or configured to:
input at least a portion of the second waveform data to the at least one trained machine learning model; and generate an output from the at least one trained machine learning model comprising the estimated ICP based on at least one shape feature of the at least one waveform of the second waveform data ([0016] “the system further comprising a modelling module for establishing a model relating features of an optical signal to intracranial pressure, the optical signal being representative of a degree to which light input into a subject's skull is absorbed by the subject's brain; a feature extraction module operable to extract one or more signal features from an absorbance signal derived from the optical signals output by said detectors; and an intracranial pressure prediction module operable to input said signal features into said model and output an indication of intracranial pressure in accordance with said model”). However, Kyriacou fails to disclose using mean arterial pressure (MAP) to determine ICP.
The combination of Kyriacou/Mourad discloses determine mean arterial pressure (MAP) data of the first patient, wherein, while determining the estimated ICP in the first patient based on the at least one trained machine learning model (Mourad: [0190] “taking measurements from a variety of patients under various circumstances, and the determination of displacement and ABP [arterial blood pressure] can then be used to calculate ICP, where ICP=F.sub.2(d)-MAP.” [0064] “A neural network, set up and trained;” Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”), the at least one processor (Kyriacou: controller 11, processor 15) is programmed or configured to:
input the MAP data to the at least one trained machine learning model; and generate the output from the at least one trained machine learning model comprising the estimated ICP based on the MAP data and the at least one shape feature of the at least one waveform of the second waveform data (Mourad: [0064] “A neural network, set up and trained as described below, was used to derive a non-linear relationship, which is further characterized below, and which provided accurate determinations of ICP in experimental protocols based on two variable parameters: V_mca measurements taken using TCD techniques; and ABP measured;” [0192] “This method uses a derived relationship between the amplitude of reflected acoustic signal(s) from CNS target tissue sites, ABP, and invasively monitored ICP to estimate ICP from non-invasively measured acoustic signals and ABP… Since the backscatter signal is related to the arterial pulse wave, a can be normalized to the MAP (as defined above), to produce a waveform .beta.. The relationship between this normalized waveform, .beta., and invasively measured ICP is then determined by taking simultaneous measurements of the backscatter signal, ABP, and ICP and solving for the equation.” Kyriacou: [0068] “Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor).” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).” [0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”).
Therefore, 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 method of Kyriacou to include mean arterial pressure (MAP) in the determination of ICP as disclosed in Mourad to accurately determine ICP using data measured non-invasively or minimally invasively on an intermittent or on a substantially continuous basis (Mourad [0064]).
While Kyriacou discloses in [0013] that in conventional approaches, “treatment is usually initiated if ICP increases >20 mm Hg,” the combination of Kyriacou/Mourad fails to particularly disclose performing treatment.
The combination of Kyriacou/Mourad/Shamim discloses and perform at least one treatment for the first patient based on the estimated ICP (Kyriacou: [0039] “provide warnings (for example, a visual and or sonic warning) in the event that nICP changes, for example exceeds a predetermined threshold.” Shamim: [0030] “The VP shunt 202 is a thin tube through which excess CSF is drained to prevent pressure from getting too high (e.g., higher than a threshold value) in the brain 106. The shunt 202 is placed under the skin.” [0031] “a controller that can control the activation or deactivation of the pump based on such determination of whether there is excess CSF in the brain 106.”).
Therefore, 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 combination of Kyriacou/Mourad to perform treatment based on the ICP as disclosed in Shamim to treat hydrocephalus and prevent pressure in the brain from getting too high (Shamim [0004, 0030]).
Regarding claim 13, the combination of Kyriacou/Mourad/Shamim discloses the system of claim 12, wherein the at least one shape feature of the at least one waveform comprises at least one of the following: area under the curve (AUC), x-coordinate of the center of mass (COMX), y-coordinate of the center of mass (COMy), peak height, peak width, peak location, or any combination thereof (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a).
Regarding claim 14, the combination of Kyriacou/Mourad/Shamim discloses the system of claim 13, wherein, while generating the output comprising the estimated ICP, the at least one processor is programmed or configured to: generate the output from the at least one trained machine learning model comprising the estimated ICP based on the at least one shape feature of the at least one waveform of the second waveform data (Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”), wherein the at least one shape feature comprises a plurality of different shape features (Kyriacou: [0016] “a feature extraction module operable to extract one or more signal features from an absorbance signal derived from the optical signals output by said detectors;” [0042] “Several features are extracted from each pulse waveform and averaged over a rolling-window of, in this particular implementation, approximately 15 seconds.” [043-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a).
Regarding claim 15, the combination of Kyriacou/Mourad/Shamim discloses the system of claim 12, wherein the at least one blood attribute comprises at least one of the following: change in oxygenated hemoglobin concentration (AHbO), change in total hemoglobin concentration (AHbT), or any combination thereof (Kyriacou: [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”).
Regarding claim 16, Kyriacou teaches a computer program product comprising at least one non-transitory computer-readable medium comprising one or more instructions that, when executed by at least one processor ([0079] “The controller 11 further comprises read only memory (ROM) and/or random access memory (RAM) 19 that provides a processing environment in which the processor 15 can execute computer programs. The controller also includes a data store 21 for the storage of computer programs for execution by the processor.”), cause the at least one processor to:
generate first waveform data using near-infrared spectroscopy (NIRS) to measure at least one light-based signal in each patient of a plurality of patients, wherein the first waveform data comprises a plurality of waveforms ([0034] “the sensor contains an array of four optical sources that are capable of emitted light of different wavelengths. In this arrangement the sources comprise light emitting diodes emitting light of four different wavelengths in the near infrared/infrared.” [0073] “The resultant data are then split into training and test data (at a discretionary ratio), following common Machine Learning techniques. The training data (usually 70-80% of total data) are used to train the prediction model, whereas the remaining data is held to later validate the performance of the model.” [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose.”), and wherein each waveform of the plurality of waveforms is associated with at least one blood attribute ([0077] “One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”);
train at least one machine learning model based on the first waveform data to produce at least one trained machine learning model, wherein the at least one trained machine learning model is configured to generate an output of intracranial pressure (ICP) based on one or more waveforms associated with the at least one blood attribute that is input to the at least one trained machine learning model ([0068] “In another envisaged arrangement depicted schematically in FIG. 7, an algorithm based on, for instance, multiple variable regression may be employed. Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor). The weights of the inputs of the model are adjusted during training to produce the lowest error (cost function) between the predicted and target ICP values.” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features). The prediction model will try to fit the response variable and predictors whilst minimising the error of prediction. In one illustrative implementation, the model is based on Support Vector Machines algorithms. However, another model among the many available, for instance Neural Network, could also be utilised.”);
generate second waveform data using NIRS to measure at least one light-based signal in a first patient, wherein the second waveform data comprises at least one waveform associated with the at least one blood attribute ([0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”); and
determine an estimated ICP in the first patient based on the at least one trained machine learning model, wherein the one or more instructions that cause the at least one processor to determine the estimated ICP in the first patient based on the at least one trained machine learning model ([0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”) cause the at least one processor to:
input at least a portion of the second waveform data to the at least one trained machine learning model; and generate an output from the at least one trained machine learning model comprising the estimated ICP based on at least one shape feature of the at least one waveform of the second waveform data ([0016] “the system further comprising a modelling module for establishing a model relating features of an optical signal to intracranial pressure, the optical signal being representative of a degree to which light input into a subject's skull is absorbed by the subject's brain; a feature extraction module operable to extract one or more signal features from an absorbance signal derived from the optical signals output by said detectors; and an intracranial pressure prediction module operable to input said signal features into said model and output an indication of intracranial pressure in accordance with said model”). However, Kyriacou fails to disclose using mean arterial pressure (MAP) to determine ICP.
The combination of Kyriacou/Mourad discloses determine mean arterial pressure (MAP) data of the first patient, wherein the one or more instructions that cause the at least one processor to determine the estimated ICP in the first patient based on the at least one trained machine learning model (Mourad: [0190] “taking measurements from a variety of patients under various circumstances, and the determination of displacement and ABP [arterial blood pressure] can then be used to calculate ICP, where ICP=F.sub.2(d)-MAP.” [0064] “A neural network, set up and trained;” Kyriacou: [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).”), cause the at least one processor (Kyriacou: controller 11, processor 15) to:
input the MAP data to the at least one trained machine learning model; and generate the output from the at least one trained machine learning model comprising the estimated ICP based on the MAP data and the at least one shape feature of the at least one waveform of the second waveform data Mourad: [0064] “A neural network, set up and trained as described below, was used to derive a non-linear relationship, which is further characterized below, and which provided accurate determinations of ICP in experimental protocols based on two variable parameters: V_mca measurements taken using TCD techniques; and ABP measured;” [0192] “This method uses a derived relationship between the amplitude of reflected acoustic signal(s) from CNS target tissue sites, ABP, and invasively monitored ICP to estimate ICP from non-invasively measured acoustic signals and ABP… Since the backscatter signal is related to the arterial pulse wave, a can be normalized to the MAP (as defined above), to produce a waveform .beta.. The relationship between this normalized waveform, .beta., and invasively measured ICP is then determined by taking simultaneous measurements of the backscatter signal, ABP, and ICP and solving for the equation.” Kyriacou: [0068] “Features from the acquired waveforms (such as one or more of the aforementioned features) are input into a model (prediction function) trained on data obtained in a clinical trial and labelled with target ICP values obtained using a gold standard method (such as an invasive ICP monitor).” [0069] “FIG. 8 is a schematic representation of an illustrative process for establishing the aforementioned model.” [0074] “The model is then trained/built using supervised Machine Learning, where the target (known) intracranial pressure is used to construct a model that will predict the response variable (ICP) based on the input predictors (features).” [0076] “Once the prediction model has been finalised and validated, it can be implemented into a ‘live’ process implementing an algorithm/software that predicts intracranial pressure measurements in real-time, as shown in FIG. 9.”).
Therefore, 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 method of Kyriacou to include mean arterial pressure (MAP) in the determination of ICP as disclosed in Mourad to accurately determine ICP using data measured non-invasively or minimally invasively on an intermittent or on a substantially continuous basis (Mourad [0064]).
While Kyriacou discloses in [0013] that in conventional approaches, “treatment is usually initiated if ICP increases >20 mm Hg,” the combination of Kyriacou/Mourad fails to particularly disclose performing treatment.
The combination of Kyriacou/Mourad/Shamim discloses and perform at least one treatment for the first patient based on the estimated ICP (Kyriacou: [0039] “provide warnings (for example, a visual and or sonic warning) in the event that nICP changes, for example exceeds a predetermined threshold.” Shamim: [0030] “The VP shunt 202 is a thin tube through which excess CSF is drained to prevent pressure from getting too high (e.g., higher than a threshold value) in the brain 106. The shunt 202 is placed under the skin.” [0031] “a controller that can control the activation or deactivation of the pump based on such determination of whether there is excess CSF in the brain 106.”).
Therefore, 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 combination of Kyriacou/Mourad to perform treatment based on the ICP as disclosed in Shamim to treat hydrocephalus and prevent pressure in the brain from getting too high (Shamim [0004, 0030]).
Regarding claim 17, the combination of Kyriacou/Mourad/Shamim discloses the computer program product of claim 16, wherein the one or more instructions further cause the at least one processor to: compare the estimated ICP to at least one predetermined threshold ICP; and in response to the estimated ICP satisfying the at least one predetermined threshold ICP, generate at least one alert to a computing device associated with a healthcare personnel providing care to the first patient (Kyriacou: [0039] “In an envisaged arrangement, the controller is coupled to a display for displaying a generated measure of nICP to clinicians, and may be configured to provide warnings (for example, a visual and or sonic warning) in the event that nICP changes, for example exceeds a predetermined threshold.”).
Regarding claim 19, the combination of Kyriacou/Mourad/Shamim discloses the computer program product of claim 16, wherein the at least one shape feature of the at least one waveform comprises at least one of the following: area under the curve (AUC), x-coordinate of the center of mass (COMX), y-coordinate of the center of mass (COMy), peak height, peak width, peak location, or any combination thereof (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a).
Regarding claim 20, the combination of Kyriacou/Mourad/Shamim discloses the computer program product of claim 16, wherein the at least one blood attribute comprises at least one of the following: change in oxygenated hemoglobin concentration (AHbO), change in total hemoglobin concentration (AHbT), or any combination thereof (Kyriacou: [0077] “Different optical signals can be used to predict the intracranial pressure. The optical probe enables the collection of light at multiple (in this instance, four) different wavelengths, hence different options are available for this purpose. One illustrative option is to use a single wavelength at the isobestic point to extract the relevant features for intracranial pressure prediction. The advantage of using the isobestic wavelength is its independence from blood oxygenation. Another option is to obtain a pulsatile signal from the pulsatile component of the oxygenated haemoglobin (HbO2). By using two of the wavelengths, the oxygenated and deoxygenated haemoglobin signals can be spectroscopically separated. From this, the HbO2 can be used to extract the relevant intracranial pressure features.”).
Claim(s) 21-23 is/are rejected under 35 U.S.C. 103 as being unpatentable over Kyriacou (US 20220409080 A1) in view of Mourad (US 20050015009 A1) and Shamim (US 20230001164 A1), and in further view of Eide (US 20030100845 A1).
Regarding claim 21, the combination of Kyriacou/Mourad/Shamim discloses the method of claim 1, wherein the at least one waveform is turned into a polygon and the at least one shape feature of the at least one waveform (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a). However, the combination of Kyriacou/Mourad/Shamim fails to disclose center of mass.
Eide teaches a system and method for digital sampling, quantitative analysis and presentation of pressures in a body cavity. Eide discloses at least one shape feature of the at least one waveform comprises a x- coordinate of the center of mass (COMx) and/or a y-coordinate of the center of mass (COMy) ([0161 “The matrix 53 or histogram 66 may be subject to further mathematical analysis to determine the centroid (or centroidial axis) or centre of mass of the single wave combinations of latency and amplitude.” [0163] “Instead of presenting the histograms, the balanced position of single wave combinations of latency/amplitude such as centroidial axis or centre of mass may be computed and displayed.”).
Therefore, 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 combination of Kyriacou/Mourad/Shamim to include center of mass as at least one shape feature of the at least one waveform as disclosed in Eide to provide a balanced point incorporating combinations of both latency and amplitude, rather than a conventional strategy of computing the average of pressure signals (Eide [0163]).
Regarding claim 22, the combination of Kyriacou/Mourad/Shamim discloses the system of claim 12, wherein the at least one waveform is turned into a polygon and the at least one shape feature of the at least one waveform (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a). However, the combination of Kyriacou/Mourad/Shamim fails to disclose center of mass.
Eide discloses at least one shape feature of the at least one waveform comprises a x- coordinate of the center of mass (COMx) and/or a y-coordinate of the center of mass (COMy) ([0161 “The matrix 53 or histogram 66 may be subject to further mathematical analysis to determine the centroid (or centroidial axis) or centre of mass of the single wave combinations of latency and amplitude.” [0163] “Instead of presenting the histograms, the balanced position of single wave combinations of latency/amplitude such as centroidial axis or centre of mass may be computed and displayed.”).
Therefore, 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 combination of Kyriacou/Mourad/Shamim to include center of mass as at least one shape feature of the at least one waveform as disclosed in Eide to provide a balanced point incorporating combinations of both latency and amplitude, rather than a conventional strategy of computing the average of pressure signals (Eide [0163]).
Regarding claim 22, the combination of Kyriacou/Mourad/Shamim discloses the computer program product of claim 16, wherein the at least one waveform is turned into a polygon and the at least one shape feature of the at least one waveform (Kyriacou: [0042-066] discuss features including the following and more: Pulse Amplitude, Upstroke Gradient, Rise Time, Early Diastolic Decay Rate, Pulse width, Late Diastolic Area Under Curve, Total Backscattered Intensity, Decay Time, Area Under the Aurce, AUC Ratio, Second derivative Pulse ratio; Table 1; Figs. 3, 4, and 6a). However, the combination of Kyriacou/Mourad/Shamim fails to disclose center of mass.
Eide discloses at least one shape feature of the at least one waveform comprises a x- coordinate of the center of mass (COMx) and/or a y-coordinate of the center of mass (COMy) ([0161 “The matrix 53 or histogram 66 may be subject to further mathematical analysis to determine the centroid (or centroidial axis) or centre of mass of the single wave combinations of latency and amplitude.” [0163] “Instead of presenting the histograms, the balanced position of single wave combinations of latency/amplitude such as centroidial axis or centre of mass may be computed and displayed.”).
Therefore, 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 combination of Kyriacou/Mourad/Shamim to include center of mass as at least one shape feature of the at least one waveform as disclosed in Eide to provide a balanced point incorporating combinations of both latency and amplitude, rather than a conventional strategy of computing the average of pressure signals (Eide [0163]).
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 MOLLY HALPRIN whose telephone number is (703)756-1520. The examiner can normally be reached 12PM-8PM ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Robert (Tse) Chen can be reached at (571) 272-3672. 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.
/M.H./Examiner, Art Unit 3791
/DEVIN B HENSON/Primary Examiner, Art Unit 3791
Prosecution Insights