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 Arguments
Applicant’s arguments, filed 11/26/2025, with respect to the objection of claim 9 have been fully considered and are persuasive. The objection of claim 9 has been withdrawn.
Applicant's arguments, filed 11/26/2025, with respect to the rejection of claims 1-32 under 35 U.S.C. § 101 have been fully considered but they are not persuasive.
Applicant contends that claims 1-32 are not directed to an abstract idea since the claims comprise machine learning-based (ML-based) models to process data and calculate confidence scores corresponding to a disease state being present in a patient sample. Examiner respectfully disagrees.
Independent claims 1 and 17 are directed to an abstract idea since the language is similar to concepts relating to organizing or analyzing information in a way that can be performed mentally or are analogous to human mental work. The language reads on a human performing the claimed behavioral observation and analysis of animals, and calculation of a subsequent confidence score corresponding to a disease state mentally. The claims do not require the use of a computer beyond the recitation of a general-purpose processor to gather information about a subject, therefore they are not self-evidently patent eligible. They appear to be directed to an abstract idea of gathering data for analysis. For instance, a medical professional could observe the behavior of a detection animal and utilize their judgment and prior knowledge to determine whether the behavior of the detection animal indicates that a disease state is present.
Applicant's arguments, filed 11/26/2025, with respect to the rejection of claims 1-32 under 35 U.S.C. § 103 have been fully considered but they are not persuasive. Additionally, since the amendments to independent claims 1 and 17 change the scope of claims 1-32 and do not merely incorporate limitations from previous dependent claims, where Applicant amended to include specific sensor measurements that were not previously required, a new grounds of rejection is made in view of previously applied references as well as new reference Mark Danieli et al. (U.S. Pub. No. 2023/0062575 A1) as explained in further detail below.
Mark teaches a diagnosis system and method for detecting various Volatile Organic Compounds (VOCs) in a biological sample by utilizing a smart platform configured to harness animal bio-sensors trained to detect various VOC's which may indicate various pathologies, and by conducting various data collection practices and analysis methods in order to produce output results (Abstract), and further teaches that the sensor data comprises data received from one or more behavioral sensors (¶[0087], where “According to some embodiments, detection means 110 may be a motion, vibration or an IR sensor or a visual camera, a microphone or any other known sensing device,” ¶[0100], where “the animal bio-sensor may sniff said VOCs during a certain time frame while being constantly monitored by detection means 110”) which measure one of more of:
a) a duration of a sniff (¶[0122], where “In operation 507, the animal bio-sensor may then sniff the VOCs emanating from the biological sample for a certain period of time. As disclosed above, detection means 110 are configured to monitor the animal bio-sensor and measure, inter alia, the time it has been actively sniffing said VOCs”) from the detection canines (Examiner takes the position that the detection canines are taught by Kjellsen.), b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff.
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., “wherein the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff from the detection canines, b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff”) are taught by Mark, newly cited, in the rejection below.
Claim Objections
Claim 1 is objected to because of the following informalities: in line 6, the claim reads “one or more behavioral sensors which measure one of more of” but should read “one or more behavioral sensors which measure one or more of”. Appropriate correction is required.
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 1-32 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., an abstract idea) without significantly more.
Eligibility Step 1 – The Four Categories of Statutory Subject Matter
Claims 1-10, 12-26, and 28-32 fall within one of the four categories of statutory subject matter. Claims 1-10 and 12-16 are drawn to “a system” (i.e., a machine) and claims 17-26 and 28-32 are drawn to “a method” (i.e., a process), and thus fall within one of the four statutory categories.
Eligibility Step 2A, Prong One
Claims 1-10, 12-26, and 28-32 recite an abstract idea:
Regarding independent claims 1 and 17, the limitation of “a machine learning-based (ML-based) disease-detection model trained on a dataset of detection events” in independent claim 1, and the limitation of “processing, using a ML-based disease-detection model trained on a dataset of detection events” in independent claim 17 is directed to an abstract idea.
This claim language is identified as an abstract idea, because in MPEP §2106.04(a)(2) III B. this language is similar to concepts relating to organizing or analyzing information in a way that can be performed mentally or are analogous to human mental work. For example, Synopsys, Inc. v. Mentor Graphics Corp., 839 F.3d 1138, 120 USPQ2d 1473 (Fed. Cir. 2016). In Synopsys, the patentee claimed methods of logic circuit design, comprising converting a functional description of a level sensitive latch into a hardware component description of the latch. 839 F.3d at 1140; 120 USPQ2d at 1475. Although the patentee argued that the claims were intended to be used in conjunction with computer-based design tools, the claims did not include any limitations requiring computer implementation of the methods and thus do not involve the use of a computer in any way. 839 F.3d at 1145; 120 USPQ2d at 1478-79. The court therefore concluded that the claims “read on an individual performing the claimed steps mentally or with pencil and paper,” and were directed to a mental process of “translating a functional description of a logic circuit into a hardware component description of the logic circuit.” 839 F.3d at 1149-50; 120 USPQ2d at 1482-83.
In the instant case, the identified abstract idea is similar to Synopsys because the language reads on a human performing the claimed behavioral observation and analysis of animals, and calculation of a subsequent confidence score corresponding to a disease state mentally. The claims do not require the use of a computer beyond the recitation of a general-purpose processor to gather information about a subject, therefore they are not self-evidently patent eligible. They appear to be directed to an abstract idea of gathering data for analysis. For instance, a medical professional could observe the behavior of a detection animal and utilize their judgment and prior knowledge to determine whether the behavior of the detection animal indicates that a disease state is present.
Eligibility Step 2A, Prong Two
Claims 1-10, 12-26, and 28-32 do not recite additional elements that integrate the judicial exception into a practical application:
Regarding independent claim 1, the limitations of “receive sensor data,” “process the sensor data,” and “calculate … one or more confidence scores” generally link the use of the mental process to a particular field and merely use a computer as a tool to perform the mental process.
Regarding dependent claim 2, the limitation of “one or more behavioral sensors, one or more physiological sensors, or one or more environmental sensors” is merely insignificant, extra-solution activity used for data gathering.
Regarding independent claim 17, the limitations of “accessing sensor data,” “processing … the sensor data,” and “calculating … one or more confidence scores” generally link the use of the mental process to a particular field and merely use a computer as a tool to perform the mental process, and the limitation of “a test kit” is merely insignificant, extra-solution activity used for data gathering.
Regarding dependent claim 18, the limitation of “one or more behavioral sensors, one or more physiological sensors, or one or more environmental sensors” is merely insignificant, extra-solution activity used for data gathering.
Eligibility Step 2B
Claims 1 and 17 do not amount to significantly more than the abstract ideas recited therein:
Regarding independent claim 1, the limitations of “receive sensor data,” “process the sensor data,” and “calculate … one or more confidence scores” generally link the use of the mental process to a particular field and merely use a computer as a tool to perform the mental process.
Regarding independent claim 17, the limitations of “accessing sensor data,” “processing … the sensor data,” and “calculating … one or more confidence scores” generally link the use of the mental process to a particular field and merely use a computer as a tool to perform the mental process, and the limitation of “a test kit” is merely insignificant, extra-solution activity used for data gathering.
Regarding dependent claims 2-10, 12-16, 18-26, and 28-32, the limitations of these claims further define the limitations already indicated as being directed to the abstract idea as recited in claims 1 and 17.
Dependent claims 2 and 18 further define the abstract idea.
Dependent claims 2 and 18 further define the data gathering, where the limitation of “one or more behavioral sensors, one or more physiological sensors, or one or more environmental sensors” is merely insignificant, extra-solution activity used for data gathering.
Dependent claims 3-11, 12-16, 19-26, and 28-32 further define the abstract idea.
Therefore, these additional elements do not amount to significantly more than the judicial exception and the claimed subject matter appears to be ineligible under §101.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-9 and 12-16 are rejected under 35 U.S.C. 103 as being unpatentable over Kjellsen et al. (hereinafter “Kjellsen”) (U.S. Pub. No. 2022/0087220 A1, IDS reference No. 12 from IDS dated 12/07/2023) in view of Datta et al. (hereinafter “Datta”) (U.S. Pub. No. 2018/0293430 A1, IDS reference No. 8 from IDS dated 12/07/2023), Mark Danieli et al. (hereinafter “Mark”) (U.S. Pub. No. 2023/0062575 A1), and Shi et al. (hereinafter “Shi”) (WO 2022/015700 A1, IDS reference No. 15 from IDS dated 12/07/2023).
Regarding claim 1, Kjellsen teaches a system for disease-detection (Abstract, where “Systems and methods for training canines to detect a virus by scent.” Examiner takes the position that a virus is a type of disease.) comprising:
a disease-detection system operable to: receive sensor data associated with one or more detection canines that have been exposed to a biological sample of a patient (¶[0040], where “Live odor training includes exposing the canine to the target scent, where the target scent is the scent of a live virus. The canine is exposed to the target scent by a box, can, scent wheel, or another container with a sample of the virus inside. The canine is imprinted on live volatile organic compounds (VOCs) specific to the virus.” Examiner takes the position that the sample with the virus inside is equivalent to a biological sample. ¶[0056], where “canine is trained to generate a separate response for each virus and/or disease of the plurality of viruses and/or diseases. Advantageously, the at least one wearable device is configured to detect the different responses for each virus and/or disease,” ¶[0058], where “When the canine finds the source of the target scent or odor, the canine performs an alert behavior … passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors”), wherein the sensor data comprises data received from one or more behavioral sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include heartrate detectors, sound detectors, motion detectors, breathing detectors, temperature detectors, canine body language detectors, pressure sensors, air sensors, GPS devices, accelerometers and/or other sensors,” where an accelerometer is a type of behavioral sensor.).
Although Kjellsen teaches a system for disease detection utilizing a canine, Kjellsen does not teach a machine learning-based (ML-based) disease-detection model trained on a dataset of detection events, that the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff from the detection canines, b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff; wherein the model is operable to: process the sensor data to generate one or more feature representations; and calculate, based on the one or more feature representations, one or more confidence scores corresponding to one or more disease states associated with the biological sample, wherein each confidence score indicates a probability of the respective disease state being present in the patient and a confidence prediction interval for the respective disease state being present in the patient.
Datta teaches a system for studying the behavior of an animal (Abstract) with a machine learning-based (ML-based) disease-detection model trained on a dataset of detection events (¶[0016], where “3D depth cameras with analytic methods that extract comprehensive morphometric data and classify mouse behaviors through mathematical clustering algorithms that are independent of human intervention or bias,” ¶[0097], where “a system for automatically discovering, characterizing, classifying and semi-automatically labeling animal behavior for a particular species. In the example case of a mouse, the animal is tracked using a 3D depth camera both before and after some experimental intervention (or two animals are compared that represent two separate experimental conditions),” ¶[0102], where “Dynamically, changes in the overall behavioral state of the animal can be identified by examining the probabilities for which the animal transitions between postural clusters (under the conditions described above),” ¶[0114], where “dimensions may be explicitly combined, subtracted, or eliminated by a suite of dimensionality reduction methods. These methods include principal components analysis (PCA), singular value decomposition (SVD), independent components analysis (ICA), locally linear embedding (LLE) or neural networks”), wherein the model is operable to:
process the sensor data to generate one or more feature representations (¶[0012], where “software the present inventors have developed can effectively segment individual mice from the arena background, determine the orientation of the rodent (defining head and tail), and then quantitatively describe its three-dimensional contour, location, velocity, orientation and more than 20 additional morphological descriptors, all in realtime”); and
calculate, based on the one or more feature representations, one or more confidence scores corresponding to one or more disease states associated with the biological sample (¶[0012], where “Using this morphometric information the present inventors have developed algorithms that identify mathematical patterns in the data that are stable over short timescales, each of which represents a behavioral state of the animals (FIG. 8). The present inventors refer to each of these mathematical clusters as QBPs—Quantitative Behavioral Primitives—and can demonstrate that complex behaviors can be represented as individual QBPs or sequences of QBPs; one can use these QBPs to automatically and in real-time detect stereotyped postures and behaviors of mice,” ¶[0102], where “Dynamically, changes in the overall behavioral state of the animal can be identified by examining the probabilities for which the animal transitions between postural clusters (under the conditions described above),” ¶[0202], where “odors altering QBP dynamics. FIG. 10 includes a transition matrix plotting the probability of transitions between behavioral states (from the dataset shown in FIG. 10); the likelihood that the state in the column occurs after the state in the row is plotted, with the log probabilities within each square heatmapped”), wherein each confidence score indicates a probability of the respective disease state being present in the patient (¶[0202], where “ a transition matrix plotting the probability of transitions between behavioral states (from the dataset shown in FIG. 10); the likelihood that the state in the column occurs after the state in the row is plotted, with the log probabilities within each square heatmapped”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches a machine learning-based (ML-based) disease-detection model trained on a dataset of detection events, wherein the model is operable to: process the sensor data to generate one or more feature representations; and calculate, based on the one or more feature representations, one or more confidence scores corresponding to one or more disease states associated with the biological sample, wherein each confidence score indicates a probability of the respective disease state being present in the patient, with the invention of Kjellsen in order to analyze and classify data independent of human intervention or bias (Datta ¶[0016]), to automatically and in real-time detect stereotyped postures and behaviors (Datta ¶[0012]), and to characterize the overall behavioral state of the animal and to describe how this state is altered by differences in stimulus or genotype without direct reference to natural language descriptors (Datta ¶[0203]).
Neither Kjellsen nor Datta teaches that the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff from the detection canines, b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff; wherein each confidence score indicates a confidence prediction interval for the respective disease state being present in the patient.
Mark teaches a diagnosis system and method for detecting various Volatile Organic Compounds (VOCs) in a biological sample by utilizing a smart platform configured to harness animal bio-sensors trained to detect various VOC's which may indicate various pathologies, and by conducting various data collection practices and analysis methods in order to produce output results (Abstract), and further teaches that the sensor data comprises data received from one or more behavioral sensors (¶[0087], where “According to some embodiments, detection means 110 may be a motion, vibration or an IR sensor or a visual camera, a microphone or any other known sensing device,” ¶[0100], where “the animal bio-sensor may sniff said VOCs during a certain time frame while being constantly monitored by detection means 110”) which measure one of more of:
a) a duration of a sniff (¶[0122], where “In operation 507, the animal bio-sensor may then sniff the VOCs emanating from the biological sample for a certain period of time. As disclosed above, detection means 110 are configured to monitor the animal bio-sensor and measure, inter alia, the time it has been actively sniffing said VOCs”) from the detection canines (Examiner takes the position that the detection canines are taught by Kjellsen.), b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff.
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Mark, which teaches that the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff, with the modified invention of Kjellsen in order to determine whether there is a positive or negative biological sample indicative of a possible pathological finding (Mark ¶[0123]-¶[0124]).
None of Kjellsen, Datta, nor Mark teaches that each confidence score indicates a confidence prediction interval for the respective disease state being present in the patient.
Shi teaches classifier models, computer implemented systems, machine learning systems and methods thereof for classifying asymptomatic patients into a risk category for having or developing cancer and/or classifying a patient with an increased risk of having or developing cancer into an organ system-based malignancy class membership and/or into a specific cancer class membership (Abstract), and further teaches that each confidence score indicates a confidence prediction interval for the respective disease state being present in the patient (¶[00203], where “the cross-external validation, the AUC ROC values of the LSTM models for screening cancers were at the 95% confidence interval”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that each confidence score indicates a confidence prediction interval for the respective disease state being present in the patient, with the modified invention of Kjellsen in order to classify the cases into low, mild, moderate, and high-risk groups based on the levels of the prediction score (Shi ¶[00203]).
Regarding claim 2, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 1 as described in the rejection above.
Kjellsen teaches that the sensor data comprises data received from one or more of: one or more behavioral sensors, one or more physiological sensors, or one or more environmental sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include heartrate detectors, sound detectors, motion detectors, breathing detectors, temperature detectors, canine body language detectors, pressure sensors, air sensors, GPS devices, accelerometers and/or other sensors”).
Regarding claim 3, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors further measure one or more of: a pose of the detection canine (¶[0058], where “the wearable sensor is configured to determine when the canine has a rapid change in direction and/or movement”), or whether the detection canine looks at its handler.
Regarding claim 4, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors comprise one or more of: one or more audio sensors, one or more image sensors, one or more accelerometers, or one or more pressure sensors (¶[0058], where “The sensors further include … sound detectors, motion detectors, … canine body language detectors, pressure sensors, … accelerometers and/or other sensors”).
Regarding claim 5, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors comprise one or more image sensors (¶[0051], where “trained canines are used in conjunction with cameras, devices configured to access the Global Positioning System (GPS), and/or other devices with geopositioning or imaging capabilities”) that measure one or more of: the duration of a sniff from the detection canine, a pose of the detection canine, whether the detection canine looks at its handler, or a number of repeated sniffs (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include … canine body language detectors, … and/or other sensors”).
Regarding claim 6, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Datta teaches that a length of time between a sniff and a signal from the detection canine indicating a positive disease-detection event is input into the ML-based disease-detection model, wherein the signal comprises one or more of: a pose of the detection canine, the detection canine looking at its handler, or a repeated sniff (¶[0121], where “search algorithms are employed that are optimized only for a fixed length behavior, and that ignore any behavior that occurs over timescales that are significantly longer or shorter than the fixed length … the time-series of posture data is scanned in a sliding window, saving vectors in regular intervals, and perform clustering on those saved periods of posture data,” ¶[0144], where “these cameras are placed overhead and the disposition of the animal is recorded over time,” ¶[0146], where “an aversive odorant causes the animal to change his posture … This revealed sniffing behavior would be difficult or impossible to identify from data limited to two dimensions (as it cannot be disambiguated from any other posture that compresses the aspect ratio). Thus the presentation of a stimulus does not simply cause a change in the position of the animal over time (as would be typically assessed and shown as FIG. 4A), but rather induces a wholesale change in the behavioral state of the animal, one which is best assessed in three dimensions instead of two”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches a length of time between a sniff and a signal from the detection canine indicating a positive disease-detection event is input into the ML-based disease-detection model, wherein the signal comprises one or more of: a pose of the detection canine, the detection canine looking at its handler, or a repeated sniff, with the invention of Kjellsen since sniffing behavior would be difficult or impossible to identify from data limited to two dimensions (Datta ¶[0146]) and to enable efficient segmentation of an animal from any given background (¶[0151]).
Regarding claim 7, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Kjellsen teaches that the one or more physiological sensors comprises one or more of: one or more heart rate sensors, one or more heart rate variability sensors, one or more temperature sensors, one or more breath rate sensors (¶[0058], where “The sensors further include heartrate detectors, … breathing detectors, temperature detectors, … and/or other sensors”), one or more sweat rate sensors, one or more galvanic skin response (GSR) sensors, one or more electroencephalogram (EEG) sensors, one or more functional near-infrared spectroscopy (fNIR) sensors, one or more functional magnetic resonance imaging (fMRI) scanners, or one or more magnetic resonance imaging (MRI) scanners.
Regarding claim 8, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 2 as described in the rejection above.
Kjellsen teaches that the one or more environmental sensors comprise one or more of: one or more temperature sensors, one or more humidity sensors, one or more audio sensors, one or more gas sensors, or one or more air particulate sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. … The sensors further include … sound detectors, … temperature detectors, … air sensors, … and/or other sensors”).
Regarding claim 9, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 1 as described in the rejection above.
Shi teaches that the ML-based disease-detection model is further operable to: receive patient data corresponding to the patient, wherein the patient data comprises one or more of: family medical history, patient medical history, patient age, patient gender, or demographical data (¶[0038], where “the term “cohort” or “cohort population” refers to a group or segment of human subjects with shared factors or influences, such as age, family history, cancer risk factors, environmental influences, medical histories, etc. In one instance, as used herein, a “cohort” refers to a group of human subjects with shared cancer risk factors; this is also referred to herein as a “disease cohort”. In another instance, as used herein, a “cohort” refers to a normal population group matched, for example by age, to the cancer risk cohort; also referred to herein as a “normal cohort”. A “same cohort” refers to a group of human subjects having the same shared cancer risk factors as the individual undergoing assessment for a risk of having a disease such as cancer,” ¶[0049], where “classifier models, generation of those models, computer implemented systems, machine learning systems and methods thereof for classifying asymptomatic patients into a risk category for having or developing cancer. The machine learning system disclosed herein generated the present classifier models using a long short term memory (LSTM) algorithm and input values from longitudinal data of a cohort of over 157,000 asymptomatic patients”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the ML-based disease-detection model is further operable to: receive patient data corresponding to the patient, wherein the patient data comprises one or more of: family medical history, patient medical history, patient age, patient gender, or demographical data, with the modified invention of Kjellsen in order to train the classifier models and create highly accurate classifier models (Shi ¶[0048]).
Regarding claim 12, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 1 as described in the rejection above.
Shi teaches that the one or more disease states comprises one or more types of cancer (¶[0014], where “the classifier model provides binary outcomes selected from increased risk of having cancer or developing cancer above a pre-determined threshold or no increased risk of having or developing cancer below a pre-determined threshold”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states comprises one or more types of cancer, with the modified invention of Kjellsen in order to predict an increased risk of having or developing cancer (Shi ¶[0014]).
Regarding claim 13, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 12 as described in the rejection above.
Shi teaches that the one or more disease states further comprise one or more stages corresponding to the one or more types of cancer (¶[00199], where “One patient (mp #1) was classified as having an increased risk of having cancer as 5 out of 100 (positive predictive value) and the other (mp #2) was classified as having an increased risk of having cancer as 12 out of 100. Mp #1 was subsequently diagnosed with stage 1 liver cancer and mp #2 was subsequently diagnosed with stage 1 bladder cancer,” ¶[00200], where “One patient (fp #1) was classified as having an increased risk of having cancer as 2 out of 100 (positive predictive value) and the other (fp #2) was classified as having an increased risk of having cancer as 3 out of 100. Fp # was subsequently diagnosed with sragelB lung cancer and fp #2 was subsequently diagnosed with stage 2B breast cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states further comprise one or more stages corresponding to the one or more types of cancer, with the modified invention of Kjellsen in order to predict an increased risk of having or developing cancer (Shi ¶[0014]) and to identify tumor markers that would normally not raise concern (Shi ¶[00199] and ¶[00200]).
Regarding claim 14, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 12 as described in the rejection above.
Shi teaches that the one or more disease states further comprises one or more sources corresponding to the one or more types of cancer (¶[0037], where “The terms “cancer” and “cancerous” refer to or describe the physiological condition in mammals that is typically characterized by unregulated cell growth. Examples of cancer include but are not limited to, lung cancer, breast cancer, colon cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, and brain cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states further comprises one or more sources corresponding to the one or more types of cancer, with the modified invention of Kjellsen in order to detect multiple types of cancer.
Regarding claim 15, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 12 as described in the rejection above.
Shi teaches that the one or more types of cancer are selected from a group comprising: breast cancer, lung cancer, prostate cancer, and colon cancer (¶[0037], where “The terms “cancer” and “cancerous” refer to or describe the physiological condition in mammals that is typically characterized by unregulated cell growth. Examples of cancer include but are not limited to, lung cancer, breast cancer, colon cancer, prostate cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more types of cancer are selected from a group comprising: breast cancer, lung cancer, prostate cancer, and colon cancer, with the modified invention of Kjellsen in order to detect multiple types of cancer.
Regarding claim 16, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 1 as described in the rejection above.
Datta teaches that the ML-based disease-detection model is trained using a dataset of target odors and detection events, wherein the detection events comprise one or more of canine behaviors, physiological signals, or neurological signals (¶[0123], where “Hidden Markov Models, Bayes Nets and Restricted Boltzmann Machines (RBMs) have been formulated that have explicit notions of time and causality, and these are incorporated into the inventive method. For example, RBMs can be trained,” ¶[0124], where “Providing plain-language labels for the resulting clusters is a simple matter of presenting recorded video of the animal while it is performing a behavior or exhibits a posture defined by a cluster, and asking a trained observer to provide a label. So, minimal intervention is required to label a “training set” of 3D video with the inventive method, and none is required by the user, because the results of the automatic training are included into the client-side software. As mentioned above, these natural language labels will not be applied to all postural clusters, although all postural clusters are considered behaviorally meaningful,” ¶[0125], where “FIGS. 2A and 2B illustrate how clustering in accordance with the inventive principles indicates overall animal behavioral state changes when an animal is offered odor stimuli. FIG. 2A shows plots of six principle components (PC1-PC6) versus time generated when a mouse was presented with blank, fearful (TMT) and mildly positive odor … the mouse behavior was analyzed, the six principle components were found to capture most of the variance in the mouse posture,” ¶[0126], where “FIG. 2B shows the data of FIG. 2A clustered in accordance with the principles of the invention irrespective of the odor treatment. It was found that different treatments preferentially elicited different behaviors. The clusters were inspected by trained observers and some were given English language labels”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches that the ML-based disease-detection model is trained using a dataset of target odors and detection events, wherein the detection events comprise one or more of canine behaviors, physiological signals, or neurological signals, with the invention of Kjellsen in order to train a model to model high-dimensional time-series containing nonlinear interactions between variables (Datta ¶[0123]).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Kjellsen in view of Datta, Mark, and Shi as applied to claim 1 above, and further in view of Smith et al. (hereinafter “Smith”) (U.S. Pub. No. 2021/0298661 A1, IDS reference No. 10 from IDS dated 12/07/2023).
Regarding claim 10, Kjellsen in combination with Datta, Mark, and Shi teaches all limitations of claim 1 as described in the rejection above.
None of Kjellsen, Datta, Mark, nor Shi teaches that the ML-based disease- detection model is further operable to receive data comprising a number of exposures of the detection canine to the biological sample.
Smith teaches an apparatus, module, methods and systems for automated, standardized assessment and analysis of a human olfactory system's odor detection ability (Abstract) where the ML-based disease-detection model is further operable to receive data comprising a number of exposures of the detection canine to the biological sample (¶[0052], where “the cartridge houses a chip or other identifier preprogrammed to identify the odorant(s), total actuatable mass or volume of the odorant(s), and/or concentration of the odorant(s) in the headspace of the cartridge, all of which are communicated to the device … The testing device controls and varies the delivered test air volume and test frequency. The delivered test air volume will vary based on the desired concentration to deliver to the subject and from presentation to presentation for the subject. The test frequency or trial rate (number of tests per time period) will vary based on the previous test air volume delivered”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Smith, which teaches that the ML-based disease-detection model is further operable to receive data comprising a number of exposures of the detection canine to the biological sample, with the modified invention of Kjellsen in order to ensure that the cartridge headspace has the proper odor vapor concentration, which is dictated by odorant volatility and airflow (Smith ¶[0052]).
Claims 17-25 and 28-32 are rejected under 35 U.S.C. 103 as being unpatentable over Kjellsen in view of Datta, Mark, Shi, and Alburty et al. (hereinafter “Alburty”) (U.S. Pub. No. 2022/0125333 A1, IDS reference No. 14 from IDS dated 12/07/2023).
Regarding claim 17, Kjellsen teaches a method of disease-detection (Abstract, where “Systems and methods for training canines to detect a virus by scent.” Examiner takes the position that a virus is a type of disease.) comprising:
exposing the biological sample to one or more detection canines (¶[0040], where “Live odor training includes exposing the canine to the target scent, where the target scent is the scent of a live virus. The canine is exposed to the target scent by a box, can, scent wheel, or another container with a sample of the virus inside. The canine is imprinted on live volatile organic compounds (VOCs) specific to the virus.” Examiner takes the position that the sample with the virus inside is equivalent to a biological sample.);
accessing sensor data associated with the detection animals (¶[0056], where “canine is trained to generate a separate response for each virus and/or disease of the plurality of viruses and/or diseases. Advantageously, the at least one wearable device is configured to detect the different responses for each virus and/or disease,” ¶[0058], where “When the canine finds the source of the target scent or odor, the canine performs an alert behavior … passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors”), wherein the sensor data comprises data received from one or more behavioral sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include heartrate detectors, sound detectors, motion detectors, breathing detectors, temperature detectors, canine body language detectors, pressure sensors, air sensors, GPS devices, accelerometers and/or other sensors,” where an accelerometer is a type of behavioral sensor.).
Kjellsen does not teach receiving a test kit, wherein the test kit comprises a biological sample from a patient; wherein the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff from the detection canines, b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff; processing, using a ML-based disease-detection model trained on a dataset of detection events, the sensor data to generate one or more feature representations; and calculating, based on the one or more feature representations, one or more confidence scores corresponding to a probability of the respective disease state being present in the patient and a confidence prediction interval for the respective disease state being present in the patient.
Datta teaches processing, using a ML-based disease-detection model trained on a dataset of detection events (¶[0016], where “3D depth cameras with analytic methods that extract comprehensive morphometric data and classify mouse behaviors through mathematical clustering algorithms that are independent of human intervention or bias,” ¶[0097], where “a system for automatically discovering, characterizing, classifying and semi-automatically labeling animal behavior for a particular species. In the example case of a mouse, the animal is tracked using a 3D depth camera both before and after some experimental intervention (or two animals are compared that represent two separate experimental conditions),” ¶[0102], where “Dynamically, changes in the overall behavioral state of the animal can be identified by examining the probabilities for which the animal transitions between postural clusters (under the conditions described above),” ¶[0114], where “dimensions may be explicitly combined, subtracted, or eliminated by a suite of dimensionality reduction methods. These methods include principal components analysis (PCA), singular value decomposition (SVD), independent components analysis (ICA), locally linear embedding (LLE) or neural networks”), the sensor data to generate one or more feature representations (¶[0012], where “software the present inventors have developed can effectively segment individual mice from the arena background, determine the orientation of the rodent (defining head and tail), and then quantitatively describe its three-dimensional contour, location, velocity, orientation and more than 20 additional morphological descriptors, all in realtime”); and
calculating, based on the one or more feature representations, one or more confidence scores (¶[0012], where “Using this morphometric information the present inventors have developed algorithms that identify mathematical patterns in the data that are stable over short timescales, each of which represents a behavioral state of the animals (FIG. 8). The present inventors refer to each of these mathematical clusters as QBPs—Quantitative Behavioral Primitives—and can demonstrate that complex behaviors can be represented as individual QBPs or sequences of QBPs; one can use these QBPs to automatically and in real-time detect stereotyped postures and behaviors of mice,” ¶[0102], where “Dynamically, changes in the overall behavioral state of the animal can be identified by examining the probabilities for which the animal transitions between postural clusters (under the conditions described above),” ¶[0202], where “odors altering QBP dynamics. FIG. 10 includes a transition matrix plotting the probability of transitions between behavioral states (from the dataset shown in FIG. 10); the likelihood that the state in the column occurs after the state in the row is plotted, with the log probabilities within each square heatmapped”) corresponding to a probability of the respective disease state being present in the patient (¶[0202], where “ a transition matrix plotting the probability of transitions between behavioral states (from the dataset shown in FIG. 10); the likelihood that the state in the column occurs after the state in the row is plotted, with the log probabilities within each square heatmapped”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches processing, using a ML-based disease-detection model trained on a dataset of detection events, the sensor data to generate one or more feature representations; and calculating, based on the one or more feature representations, one or more confidence scores corresponding to a probability of the respective disease state being present in the patient, with the invention of Kjellsen in order to analyze and classify data independent of human intervention or bias (Datta ¶[0016]), to automatically and in real-time detect stereotyped postures and behaviors (Datta ¶[0012]), and to characterize the overall behavioral state of the animal and to describe how this state is altered by differences in stimulus or genotype without direct reference to natural language descriptors (Datta ¶[0203]).
Neither Kjellsen nor Datta teaches receiving a test kit, wherein the test kit comprises a biological sample from a patient; wherein the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff from the detection canines, b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff; nor one or more confidence scores corresponding to a confidence prediction interval for the respective disease state being present in the patient.
Mark teaches that the sensor data comprises data received from one or more behavioral sensors (¶[0087], where “According to some embodiments, detection means 110 may be a motion, vibration or an IR sensor or a visual camera, a microphone or any other known sensing device,” ¶[0100], where “the animal bio-sensor may sniff said VOCs during a certain time frame while being constantly monitored by detection means 110”) which measure one of more of:
a) a duration of a sniff (¶[0122], where “In operation 507, the animal bio-sensor may then sniff the VOCs emanating from the biological sample for a certain period of time. As disclosed above, detection means 110 are configured to monitor the animal bio-sensor and measure, inter alia, the time it has been actively sniffing said VOCs”) from the detection canines (Examiner takes the position that the detection canines are taught by Kjellsen.), b) a sniff intensity, c) a number of repeated sniffs, d) a pressure of the detection canine's nose against a sniffing port, or e) or auditory features of the sniff.
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Mark, which teaches that the sensor data comprises data received from one or more behavioral sensors which measure one of more of: a) a duration of a sniff, with the modified invention of Kjellsen in order to determine whether there is a positive or negative biological sample indicative of a possible pathological finding (Mark ¶[0123]-¶[0124]).
None of Kjellsen, Datta, nor Mark teaches receiving a test kit, wherein the test kit comprises a biological sample from a patient; nor one or more confidence scores corresponding to a confidence prediction interval for the respective disease state being present in the patient.
Shi teaches calculating, based on the one or more feature representations, one or more confidence scores corresponding to a confidence prediction interval for the respective disease state being present in the patient (¶[00203], where “the cross-external validation, the AUC ROC values of the LSTM models for screening cancers were at the 95% confidence interval”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches one or more confidence scores corresponding to a confidence prediction interval for the respective disease state being present in the patient, with the modified invention of Kjellsen in order to classify the cases into low, mild, moderate, and high-risk groups based on the levels of the prediction score (Shi ¶[00203]).
None of Kjellsen, Datta, Mark, nor Shi teaches receiving a test kit, wherein the test kit comprises a biological sample from a patient.
Alburty teaches devices and methods for using normal human breath to separately capture particles from inhaled and exhaled breath for analysis (Abstract), and further teaches receiving a test kit (¶[0075], where “When the user has worn the device for the appropriate amount of time, the mask is removed and packaged for transport to a laboratory or analyzed onsite”), wherein the test kit comprises a biological sample from a patient (¶[0084], where “when wearing the mask 500, exhaled and inhaled air can pass through the protective filter material 501 and through the opening in the protective filter material 501 and through the collection filter assembly 504. This allows for increased collection of target particles”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Alburty, which teaches receiving a test kit, wherein the test kit comprises a biological sample from a patient, with the modified invention of Kjellsen in order to allow for increased collection of target particles into the collection filter assembly while providing easy access to the collection filter assembly by health care workers (Alburty ¶[0084]).
Regarding claim 18, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 17 as described in the rejection above.
Kjellsen teaches that the sensor data comprises data received from one or more of: one or more behavioral sensors, one or more physiological sensors, or one or more environmental sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include heartrate detectors, sound detectors, motion detectors, breathing detectors, temperature detectors, canine body language detectors, pressure sensors, air sensors, GPS devices, accelerometers and/or other sensors”).
Regarding claim 19, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors further measure one or more of: a pose of the detection canine (¶[0058], where “the wearable sensor is configured to determine when the canine has a rapid change in direction and/or movement”), or whether the detection canine looks at its handler.
Regarding claim 20, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors comprise one or more of: one or more audio sensors, one or more image sensors, one or more accelerometers, or one or more pressure sensors (¶[0058], where “The sensors further include … sound detectors, motion detectors, … canine body language detectors, pressure sensors, … accelerometers and/or other sensors”).
Regarding claim 21, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Kjellsen teaches that the one or more behavioral sensors comprise one or more image sensors (¶[0051], where “trained canines are used in conjunction with cameras, devices configured to access the Global Positioning System (GPS), and/or other devices with geopositioning or imaging capabilities”) that measure one or more of: the duration of a sniff from the detection canine, a pose of the detection canine, whether the detection canine looks at its handler, or a number of repeated sniffs (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. The sensors include wearable sensors. The sensors further include … canine body language detectors, … and/or other sensors”).
Regarding claim 22, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Datta teaches that a length of time between a sniff and a signal from the detection canine indicating a positive disease-detection event is input into the ML-based disease-detection model, wherein the signal comprises one or more of: a pose of the detection canine, the detection canine looking at its handler, or a repeated sniff (¶[0121], where “search algorithms are employed that are optimized only for a fixed length behavior, and that ignore any behavior that occurs over timescales that are significantly longer or shorter than the fixed length … the time-series of posture data is scanned in a sliding window, saving vectors in regular intervals, and perform clustering on those saved periods of posture data,” ¶[0144], where “these cameras are placed overhead and the disposition of the animal is recorded over time,” ¶[0146], where “an aversive odorant causes the animal to change his posture … This revealed sniffing behavior would be difficult or impossible to identify from data limited to two dimensions (as it cannot be disambiguated from any other posture that compresses the aspect ratio). Thus the presentation of a stimulus does not simply cause a change in the position of the animal over time (as would be typically assessed and shown as FIG. 4A), but rather induces a wholesale change in the behavioral state of the animal, one which is best assessed in three dimensions instead of two”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches a length of time between a sniff and a signal from the detection canine indicating a positive disease-detection event is input into the ML-based disease-detection model, wherein the signal comprises one or more of: a pose of the detection canine, the detection canine looking at its handler, or a repeated sniff, with the invention of Kjellsen since sniffing behavior would be difficult or impossible to identify from data limited to two dimensions (Datta ¶[0146]) and to enable efficient segmentation of an animal from any given background (¶[0151]).
Regarding claim 23, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Kjellsen teaches that the physiological sensor comprises one or more of: one or more heart rate sensors, one or more heart rate variability sensors, one or more temperature sensors, one or more breath rate sensors, one or more sweat rate sensors, one or more galvanic skin response (GSR) sensors, one or more electroencephalogram (EEG) sensors, one or more functional near-infrared spectroscopy (fNIR) sensors, one or more functional magnetic resonance imaging (fMRI) scanners, or one or more magnetic resonance imaging (MRI) scanners (¶[0058], where “The sensors further include heartrate detectors, … breathing detectors, temperature detectors, … and/or other sensors”).
Regarding claim 24, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 18 as described in the rejection above.
Kjellsen teaches that the one or more environmental sensors comprise one or more of: one or more temperature sensors, one or more humidity sensors, one or more audio sensors, one or more gas sensors, or one or more air particulate sensors (¶[0058], where “passive and/or active alert behavior is detected by means of sensors. … The sensors further include … sound detectors, … temperature detectors, … air sensors, … and/or other sensors”).
Regarding claim 25, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 17 as described in the rejection above.
Shi teaches that the ML-based disease-detection model receives patient data, wherein the patient data includes or more of: family medical history, patient medical history, patient age, patient gender, or demographical data (¶[0038], where “the term “cohort” or “cohort population” refers to a group or segment of human subjects with shared factors or influences, such as age, family history, cancer risk factors, environmental influences, medical histories, etc. In one instance, as used herein, a “cohort” refers to a group of human subjects with shared cancer risk factors; this is also referred to herein as a “disease cohort”. In another instance, as used herein, a “cohort” refers to a normal population group matched, for example by age, to the cancer risk cohort; also referred to herein as a “normal cohort”. A “same cohort” refers to a group of human subjects having the same shared cancer risk factors as the individual undergoing assessment for a risk of having a disease such as cancer,” ¶[0049], where “classifier models, generation of those models, computer implemented systems, machine learning systems and methods thereof for classifying asymptomatic patients into a risk category for having or developing cancer. The machine learning system disclosed herein generated the present classifier models using a long short term memory (LSTM) algorithm and input values from longitudinal data of a cohort of over 157,000 asymptomatic patients”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the ML-based disease-detection model receives patient data, wherein the patient data includes or more of: family medical history, patient medical history, patient age, patient gender, or demographical data, with the modified invention of Kjellsen in order to train the classifier models and create highly accurate classifier models (Shi ¶[0048]).
Regarding claim 28, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 17 as described in the rejection above.
Shi teaches that the one or more disease states comprises one or more types of cancer (¶[0014], where “the classifier model provides binary outcomes selected from increased risk of having cancer or developing cancer above a pre-determined threshold or no increased risk of having or developing cancer below a pre-determined threshold”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states comprises one or more types of cancer, with the modified invention of Kjellsen in order to predict an increased risk of having or developing cancer (Shi ¶[0014]).
Regarding claim 29, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 28 as described in the rejection above.
Shi teaches that the one or more disease states further comprise one or more stages corresponding to the one or more types of cancer (¶[00199], where “One patient (mp #1) was classified as having an increased risk of having cancer as 5 out of 100 (positive predictive value) and the other (mp #2) was classified as having an increased risk of having cancer as 12 out of 100. Mp #1 was subsequently diagnosed with stage 1 liver cancer and mp #2 was subsequently diagnosed with stage 1 bladder cancer,” ¶[00200], where “One patient (fp #1) was classified as having an increased risk of having cancer as 2 out of 100 (positive predictive value) and the other (fp #2) was classified as having an increased risk of having cancer as 3 out of 100. Fp # was subsequently diagnosed with sragelB lung cancer and fp #2 was subsequently diagnosed with stage 2B breast cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states further comprise one or more stages corresponding to the one or more types of cancer, with the modified invention of Kjellsen in order to predict an increased risk of having or developing cancer (Shi ¶[0014]) and to identify tumor markers that would normally not raise concern (Shi ¶[00199] and ¶[00200]).
Regarding claim 30, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 28 as described in the rejection above.
Shi teaches that the one or more disease states further comprises one or more sources corresponding to the one or more types of cancer (¶[0037], where “The terms “cancer” and “cancerous” refer to or describe the physiological condition in mammals that is typically characterized by unregulated cell growth. Examples of cancer include but are not limited to, lung cancer, breast cancer, colon cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, and brain cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more disease states further comprises one or more sources corresponding to the one or more types of cancer, with the modified invention of Kjellsen in order to detect multiple types of cancer.
Regarding claim 31, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 28 as described in the rejection above.
Shi teaches that the one or more types of cancer are selected from a group comprising: breast cancer, lung cancer, prostate cancer, and colon cancer (¶[0037], where “The terms “cancer” and “cancerous” refer to or describe the physiological condition in mammals that is typically characterized by unregulated cell growth. Examples of cancer include but are not limited to, lung cancer, breast cancer, colon cancer, prostate cancer”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Shi, which teaches that the one or more types of cancer are selected from a group comprising: breast cancer, lung cancer, prostate cancer, and colon cancer, with the modified invention of Kjellsen in order to detect multiple types of cancer.
Regarding claim 32, Kjellsen in combination with Datta, Mark, Shi, and Alburty teaches all limitations of claim 17 as described in the rejection above.
Datta teaches that the ML-based disease-detection model is trained using a dataset of target odors and detection events, wherein the detection events include animal behavior, physiological signals, or neurological signals (¶[0123], where “Hidden Markov Models, Bayes Nets and Restricted Boltzmann Machines (RBMs) have been formulated that have explicit notions of time and causality, and these are incorporated into the inventive method. For example, RBMs can be trained,” ¶[0124], where “Providing plain-language labels for the resulting clusters is a simple matter of presenting recorded video of the animal while it is performing a behavior or exhibits a posture defined by a cluster, and asking a trained observer to provide a label. So, minimal intervention is required to label a “training set” of 3D video with the inventive method, and none is required by the user, because the results of the automatic training are included into the client-side software. As mentioned above, these natural language labels will not be applied to all postural clusters, although all postural clusters are considered behaviorally meaningful,” ¶[0125], where “FIGS. 2A and 2B illustrate how clustering in accordance with the inventive principles indicates overall animal behavioral state changes when an animal is offered odor stimuli. FIG. 2A shows plots of six principle components (PC1-PC6) versus time generated when a mouse was presented with blank, fearful (TMT) and mildly positive odor … the mouse behavior was analyzed, the six principle components were found to capture most of the variance in the mouse posture,” ¶[0126], where “FIG. 2B shows the data of FIG. 2A clustered in accordance with the principles of the invention irrespective of the odor treatment. It was found that different treatments preferentially elicited different behaviors. The clusters were inspected by trained observers and some were given English language labels”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Datta, which teaches that the ML-based disease-detection model is trained using a dataset of target odors and detection events, wherein the detection events include animal behavior, physiological signals, or neurological signals, with the invention of Kjellsen in order to train a model to model high-dimensional time-series containing nonlinear interactions between variables (Datta ¶[0123]).
Claim 26 is rejected under 35 U.S.C. 103 as being unpatentable over Kjellsen in view of Datta, Mark, Shi, and Alburty as applied to claim 17 above, and further in view of Smith.
None of Kjellsen, Datta, Mark, Shi, nor Alburty teach that the ML-based disease-detection model receives data comprising a number of exposures of the detection canine to the biological sample.
Smith teaches that the ML-based disease-detection model receives data comprising a number of exposures of the detection canine to the biological sample (¶[0052], where “the cartridge houses a chip or other identifier preprogrammed to identify the odorant(s), total actuatable mass or volume of the odorant(s), and/or concentration of the odorant(s) in the headspace of the cartridge, all of which are communicated to the device … The testing device controls and varies the delivered test air volume and test frequency. The delivered test air volume will vary based on the desired concentration to deliver to the subject and from presentation to presentation for the subject. The test frequency or trial rate (number of tests per time period) will vary based on the previous test air volume delivered”).
It would have been obvious to one of ordinary skill in the art at the time of the invention to combine the above-described teachings of Smith, which teaches that the ML-based disease-detection model receives data comprising a number of exposures of the detection canine to the biological sample, with the modified invention of Kjellsen in order to ensure that the cartridge headspace has the proper odor vapor concentration, which is dictated by odorant volatility and airflow (Smith ¶[0052]).
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 SEFRA D. MANOS whose telephone number is (703)756-5937. The examiner can normally be reached M-F: 7:00 AM - 3:30 PM ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Unsu Jung can be reached at (571) 272-8506. 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.
/SEFRA D. MANOS/Examiner, Art Unit 3792
/AMANDA L STEINBERG/Examiner, Art Unit 3792