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 .
Status of Claims
In the amendment filed 06/09/2026 the following occurred: Claims 1, 4, 8-10, 13, 16, and 19-20 were amended; Claims 7, 11-12, 14, and 17-18 were canceled; and Claims 21-26 were added as new. Claims 1-6, 8-10, 13, 15-16, and 19-26 are presented for examination.
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 1-6, 8-10, 13, 15-16, and 19-26 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Step 1
Claims 1-6, 8-10, 13, 15-16, and 19-26 are drawn to a method, a system and a non-transitory computer-readable storage medium, which is/are statutory categories of invention (Step 1: YES).
Step 2A Prong One
Independent claim 1 recites determining, based on the anonymized patient information, a first set of features; determining and based on the retinal images, a second set of features; determining a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features; determining by providing the image and the patient data as inputs, a confidence level indicative of a likelihood that the first disease is associated with the patient; determining based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease.
Independent claim 10 recites determining, based on the anonymized patient information, a first set of features; determining, based on the retinal images, a second set of features; determining a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features; determining, by providing the image and the patient data as inputs, a confidence level indicative of a likelihood that the first disease is associated with the patient; determining, based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease.
Independent claim 16 recites determine, based on the anonymized patient information, a first set of features; determine, based on the retinal images, a second set of features; determine a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features; determine, by providing the image and the EMR data as inputs, a confidence level indicative of a likelihood that the first disease is associated with the patient; and determine, based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease.
The respective dependent claims 2-6, 8-9, 13, 15, and 19-26, but for the inclusion of the additional elements specifically addressed below, provide recitations further limiting the invention of the independent claim(s).
Said recited limitations above, as drafted, under their broadest reasonable interpretation, cover certain methods of organizing human activity, as reflected in the specification, which states that having “detecting potential diseases based on retinal images and patient health records, and providing recommendations for further screening related to the detected diseases” (see: specification paragraph 2). If a claim limitation, under its broadest reasonable interpretation, covers managing personal behavior or relationships or interactions between people, then it falls within the “Certain Methods of Organizing Human Activity” grouping of abstract ideas. The present claims cover certain methods of organizing human activity because they address problems where “a primary care doctor may not be able to review the retinal scans for signs of disease” and “manual analysis of retinal scans of a patient by a retina specialist, in addition to adding to a cost and complexity of a health screening, may also fail to flag early signs of a disease because the retina specialist may not be familiar with a patient's overall medical history” (see: specification paragraph 3). These problems are addressed by an invention “to screen a patient for a host of potential diseases automatically” (see: specification paragraph 4), and particularly, the recited limitations address these problems “to a disease identification system programmed or otherwise configured to generate a recommendation for further screening” (see: specification paragraph 17). Accordingly, the claims recite an abstract idea(s) (Step 2A Prong One: YES).
Step 2A Prong Two
This judicial exception is not integrated into a practical application. The claims are abstract but for the inclusion of the additional elements including an “by a processor…by the processor and…by the processor…by the processor…training, by the processor, a machine learning (ML) model based on the correlation…by the processor…by the processor…to the trained ML model…by the processor and…” (claim 1), “by the processor…by the processor, a training dataset…updating, by the processor, the ML model by re-training with the augmented training dataset…” (claim 4), “the ML model is trained, based on a training dataset…” (claim 6), “the ML model comprises an expert system…” (claim 9), “memory; a processor; and computer-executable instructions stored in the memory and executable by the processor to perform operations comprising:…training a machine learning (ML) model based on the correlation…to the trained ML model…” (claim 10), “a training dataset…and updating the ML model by re-training with the augmented training dataset” (claim 13), “A non-transitory computer-readable storage medium storing processor-executable instructions that, when executed, cause one or more processors to:…train a machine learning (ML) model based on the correlation…to the trained ML model…” (claim 16), “by the processor…by the processor…by the processor…wherein the ML model is further trained based on the additional correlation” (claim 21), “the processor…” (claim 22), “by the processor…by the processor…by the processor…wherein the ML model is further trained based on the additional correlation” (claim 23), and “wherein the ML model is further trained based on the temporal correlation” (claim 25), which are additional elements that are recited at a high level of generality (e.g., the “processor” is configured through no more than a statement than that functions are performed “by” said processor; the “memory” and “processor” are configured through no more than a statement than that “instructions” stored in said memory are “executable” by said processor to perform operations; the “machine learning (ML) model” is “trained” through no more than a statement than that training is “based on” a correlation of features or “based on” a training dataset to produce desired results, and that “re-training” is performed “with” the augmented training dataset or the model is further trained “based on” an additional or temporal correlation; the “non-transitory computer-readable storage medium” is configured through no more than a statement than that stored “instructions” stored by said medium “cause” one or more processors to perform functions “when executed”) such that they amount to no more than mere instruction to apply the exception using generic computer elements. See: MPEP 2106.05(f).
The claims recite the additional elements of “receiving: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients, and respective retinal images corresponding to the disease diagnoses…receiving, by the processor, an image of a retina of an eye of a patient; receiving, by the processor, patient data corresponding to the patient…” (claim 1), “wherein the anonymized patient information and the respective retinal images are extracted from an electronic medical records (EMR) system” (claim 8), “receiving: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients, and respective retinal images corresponding to the disease diagnoses…receiving an image of a retina of an eye of a patient; receiving patient data corresponding to the patient…” (claim 10), and “receive: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients, and respective retinal images corresponding to the disease diagnoses…receive, from an optical imaging device, an image of a retina of an eye of a patient; access, from an electronic medical record (EMR) storage, EMR data of the patient…” (claim 16), which are nominal or tangential additions to the abstract idea(s) and amount to extra-solution activity concerning mere data gathering. The addition of an insignificant extra-solution activity limitation does not impose meaningful limits on the claims such that it is not nominally or tangentially related to the invention. In the claimed context, these claimed additional elements are incidental to the performance of the recited abstract idea(s) as outlined in the recitations above. Similarly, the claims recite the additional elements of “providing, by the processor and to an output device, an output indicating the recommendation” (claim 1) and “providing, to an electronic medical record (EMR) of the patient, an output indicating the recommendation” (claim 10), which are considered insignificant post-solution activity concerning an insignificant application, and similarly, the addition of an insignificant extra-solution activity does not impose meaningful limits on the claims such that it is not nominally or tangentially related to the invention. In the claimed context, these claimed additional elements are incidental to the performance of the recited abstract idea(s) as outlined in the recitations above. See: MPEP 2106.05(g).
The combination of these additional elements is no more than mere instructions to apply the exception using generic computer elements and limitations directed toward extra-solution activity. Accordingly, even in combination, these additional elements do not integrate the abstract idea(s) into a practical application because they do not impose any meaningful limits on practicing the abstract idea(s). Accordingly, the claims are directed to an abstract idea(s) (Step 2A Prong Two: NO).
Step 2B
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea(s) into a practical application, using the additional elements to perform the abstract idea(s) amounts to no more than mere instructions to apply the exception using generic elements. Mere instructions to apply an exception using generic elements cannot provide an inventive concept. See MPEP 2106.05(f).
The claimed additional elements directed toward extra-solution activity, identified above, are not sufficient to amount to significantly more than the judicial exception because they are understood by the courts to perform well-understood, routine, and conventional activities previously known to the industry. The claimed limitations of receiving patient information and retinal images, accessing and extracting patient EMR data from an electronic medical record (EMR) storage system, and providing information to an output device or patient EMR are representative of concepts to electronic recordkeeping, receiving or transmitting data over a network, such as using the Internet to gather data or sending messages over a network, and storing and retrieving information in memory. Said representative concepts have been identified by the courts as well-understood, routine, and conventional activities; therefore, the corresponding claimed limitations are shown to be well-understood, routine, and conventional activities. See: MPEP 2106.05(d)(II).
Additionally, or alternately, the claimed additional elements directed toward extra-solution activity, identified above, are not sufficient to amount to significantly more than the judicial exception because they are generic elements that are configured to perform well-understood, routine, and conventional activities previously known to the industry. See: MPEP 2106.05(d). Said additional elements are recited at a high level of generality and provide conventional functions that do not add meaningful limits to practicing the abstract idea(s). The originally filed specification supports this conclusion:
Paragraph 23, where “In some examples, the EMR system 110 may be connected to a clinical device 114 via a network 116. The clinical device 114 can include a computing device, such as a device including at least one processor configured to perform operations. In some cases, the operations are stored in memory in an executable format. Examples of computing devices include a personal computer, a tablet computer, a smart television (TV), a mobile device, a mobile phone, or an Internet of Things (loT) device. In some examples, the clinical device 114 may be operated by the operator 104, and may receive the image(s) 108 captured by the optical imaging device 106…In examples, the clinical device 114 or the optical imaging device 106 may store the image(s) 108 of the eye(s) of the patient 102 in the EMR system 110 in association with the EMR data 112 of the patient 102.”
Paragraph 24, where “In examples, the network 116 may represent one or more communication networks. Examples of communication networks include at least one wired interface (e.g., an ethernet interface, an optical cable interface, etc.) and/or at least one wireless interface (e.g., a BLUETOOTH interface, a WI-FI interface, a near-field communication (NFC) interface, a LongTerm Evolution (LTE) interface, a New Radio (NR) interface, etc.). In some examples, data or other signals may be transmitted between elements of FIG. 1 over a wide area network (WAN), such as the Internet. In some cases, the data may include one or more data packets (e.g., Internet Protocol (IP) data packets), datagrams, or a combination thereof.”
Paragraph 25, where “In various examples, the clinical device 114 may be connected, via the network 116, to a remote computing device 118, such as a server implemented on a cloud platform. In examples, the clinical device 114 may upload, to the remote computing device 118, the image(s) 108 captured by the optical imaging device 106. The remote computing device 118 may implement an image analysis system 120 that receives, as an input, the image(s) 108 captured by the optical imaging device 106…In other examples, the optical imaging device 106 may be in direct communication with the remote computing device 118 to upload the image(s) 108...”
Paragraph 34, where “In examples, the remote computing device 118 may also implement an EMR data extractor component 122. The EMR data extractor component 122 may process the EMR data 112 to extract data of interest…”
Paragraph 36, where “…receive, from the clinical device 114 and/or directly from the EMR system 110, the EMR data 112 associated with the patient 102…”
Paragraph 38, where “…where the recommendation may be output, via a user interface, to the operator 104…”
Paragraph 42, where “…Additionally, although FIG. 1 illustrates the optical imaging device 106, the EMR system 110, the clinical device 114, and the remote computing device 118 as separate entities, in some implementations, one or more of these entities may be correspond to the same computing device.”
Paragraph 46, where “…the EMR data extractor 122 may issue a query to the EMR system 110 (e.g., to a database storing EMR information) requesting anonymized data, and may receive, in response, the EMR data 206, and associations with corresponding data of a same patient in the retinal images 204…”
Paragraph 71, where “…In examples, the recommendation may be provided to the patient and/or a healthcare provider caring for the patient, and may be added to the electronic health record(s) of the patient.”
Paragraph 80, where “…The memory 602 may include volatile and nonvolatile memory and/or removable and non-removable media implemented in any type of technology for storage of information, such as computer-readable instructions, data structures, program modules, or other data.”
Paragraph 81, where “As described herein, the processor(s) 604, can be a single processing unit or a number of processing units, and can include single or multiple processing cores, comprising a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), or both CPU and GPU, or other processing unit known in the art.)…”
Paragraph 82, where “The device(s) 600 can also include additional data storage devices (removable and/or non-removable) such as, for example, magnetic disks, optical disks, or tape. Such additional storage is illustrated in FIG. 6 by removable storage 606 and non-removable storage 608. Tangible computer-readable media can include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, or other data. The memory 602, removable storage 606, and non-removable storage 608 are all examples of computer-readable storage media. Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Discs (DVDs), Content-Addressable Memory (CAM), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by the device(s) 600. Any such tangible computer-readable media can be part of the device(s) 600.”
Paragraph 84, where “As illustrated in FIG. 6, the device(s) 600 can also include one or more wired or wireless transceiver(s) 614. For example, the transceiver(s) 614 can include a Network Interface Card (NIC), a network adapter, a LAN adapter, or a physical, virtual, or logical address to connect to the various base stations or networks (e.g., the network 116) contemplated herein, for example, or the various user devices and servers. To increase throughput when exchanging wireless data, the transceiver(s) 614 can utilize Multiple-Input/Multiple-Output (MIMO) technology. The transceiver(s) 614 can include any sort of wireless transceivers capable of engaging in wireless, Radio Frequency (RF) communication. The transceiver(s) 614 can also include other wireless modems, such as a modem for engaging in Wi-Fi, WiMAX, Bluetooth, or infrared communication.”
The claims recite the additional elements directed to pre-solution and post-solution activity, as recited and indicated above, each of which amount to extra-solution activity. The specification (e.g., as excerpted above) does not indicate that the additional element(s) provide anything other than well‐understood, routine, and conventional functions when claimed in a merely generic manner (as they are presently). See: MPEP 2106.05(g).
Viewing the limitations as an ordered combination, the claims simply instruct the additional elements to implement the concept described above in the identification of abstract idea(s) with routine, conventional activity specified at a high level of generality in a particular technological environment. Hence, the claims as a whole, considering the additional elements individually and as an ordered combination, do not amount to significantly more than the abstract idea(s) (Step 2B: NO).
Dependent claim(s) 2-6, 8-9, 13, 15, and 19-26, when analyzed as a whole, considering the additional elements individually and/or as an ordered combination, are held to be patent ineligible under 35 U.S.C. 101 because the additional recited limitation(s) fail(s) to establish that the claim(s) is/are not directed to an abstract idea(s) without significantly more. These claims fail to remedy the deficiencies of their parent claims above and are therefore rejected for at least the same rationale as applied to their parent claims above and incorporated herein.
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) 1-6, 10, 13, and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over WO 2023/214890 to Rezaei in view of JP-7596092-B2 to Fukushima.
As per claim 1, Rezaei teaches a method, comprising:
receiving, by a processor: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients , and respective retinal images corresponding to the disease diagnoses (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc; paragraph 54, is met by Because not all measurements are related to the CVD risk, irrelevant columns were discarded according to the expert advice. As a result, 35 columns corresponding to 21 fields remained, including: age, sex, ethnicity, deprivation score, family history, smoking, systolic blood pressure, BMI, TC/HDL, HbAlc, state of diabetes (Y /N), diabetic type, atrial fibrillation, anti hypertensives, antithrombotic medication, lipid lowering medication, eGFR, metolazone prior 6 months, lipids in prior 6 months, LLD prior 6 months, anticoagulation medication prior 6 months, antiplay prior 6 months, CVD event and date, etc.; and paragraph 96-99, is met by disclose that the UK Biobank data was used);
receiving, by a processor (see: Rezaei, Fig. 1; and paragraph 48, is met by the processing system 1002 may have processing facilities represented by one or more processors 1004. memory 1006, and other components typically present in such computing environments. In the exemplary embodiment illustrated the memory 1006 stores information accessible by processor 1004, the information comprising instructions 1008 that may be executed by the processor 1004 and data 1010 that may be retrieved, manipulated or stored by the processor 1004), an image of a retina of an eye of a patient (see: Rezaei, Fig. 2A; and paragraph 56, is met by at input stage 2002 one or more fundus images are received - for example a collection of fundus photographs of an individual);
receiving, by the processor and, patient data corresponding to the patient (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc);
determining, by the processor and by providing the image and the patient data as inputs to the trained ML model, a confidence level indicative of a likelihood that the first disease is associated with the patient (see: Rezaei, Fig. 2A; and paragraph 74-75, is met by the processing pipelines of fundus images 2016 and meta-information 2020 are completed, the individual-level fundus image feature vector and meta-information vector are concatenated together to form an individual feature vector at step 2022 and that the individual feature vector is processed by a CVD risk prediction neural network model 2024 utilizing a fully connected neural network (FCNN). See also Figs. 4A-D);
determining, by the processor and based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease (see: Rezaei, Fig. 4D; and paragraph 32 and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), a referral recommendation 4076 is included in the report - for example recommending a consultation with a cardiologist); and
providing, by the processor and to an output device, an output indicating the recommendation (see: Rezaei, Fig. 4D; and paragraph 89, is met by exemplary individual report; may be sent to a healthcare provider for further follow ups).
Rezaei fails to specifically teach the following limitations met by Fukushima as cited:
determining, by the processor and based on the anonymized patient information, a first set of features (see: Fukushima, page 9, is met by the input data are bioinformation, such as blood pressure, BMI index, age, sex, medical history, or smoking habit);
determining, by the processor, and based on the retinal images, a second set of features (see: Fukushima, page 9, is met by the input data are feature quantities obtained from fundus images, such as arterial diameter, venous diameter, ratio of arterial diameter to venous diameter, branching angle of blood vessels, asymmetry of the branching, arterial venous stenosis or vascular tortuosity);
determining, by the processor, a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features (see: Fukushima, page 10, is met by the correlation is learned between the feature quantities and the probability of developing diseases assessed from the bioinformation);
training, by the processor, a machine learning (ML) model based on the correlation (see: Fukushima, page 9-10, is met by inputting the feature amounts and biometric information acquired to a trained model generated in advance, thereby estimating the risk of the subject developing a disease);
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the model training and data as taught by Rezaei to include inputting the feature amounts and biometric information acquired to learn a correlation between the feature quantities and the probability of developing diseases assessed from the bioinformation in order to train a model to estimate the risk of the subject developing a disease as taught by Fukushima with the motivation of improving the accuracy of disease risk estimation at least by inputting multiple features (see: Fukushima, page 9-10).
As per claim 2, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the second set of features comprises at least one of: a brightness level of an optic disc of the retina, a diameter of blood vessels of the retina, a topology of the blood vessels of the retina, an edema of the optic disc, or an arteriovenous ratio (AVR) (see: Rezaei, paragraph 20, is met by a brightness adjustment process may be performed on the one or more fundus images; paragraph 24 and 67, is met by the CNNs may look for "localized" signs of biological changes and physiological changes (e.g. microaneurysms, oedema, etc.) changes, and or "global" changes in an image that could indicate presence of glycaemic control, blood pressure, cholesterol, and exposure to smoking (e.g. pigmentary changes in the peripapillary region, arterial/venous crossing deformations, vascular tortuosity changes, vascular calibre changes, etc.). In examples the signs may include, but not be limited to: drusen appearance, clustering, and/or location; pigmentation change in density and/or location; arteriovenous crossing; change in arteriovenous crossing calibre and/or thickness change; arteriovenous tortuosity; retinal oedema size and/or pattern; and/or microaneurysms concentration).
As per claim 3, Rezae and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the patient data comprises at least one of: an age of the patient, a sex of the patient, a race of the patient, a smoking status of the patient, a blood pressure measurement of the patient, or one or more medical test results associated with the patient (see: Rezaei, paragraph 73 is met by gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, etc, and other is numerical data such as age, HbAlc, etc.).
As per claim 4, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
receiving, by the processor, follow-up information indicating whether the patient was diagnosed with the first disease; augmenting, by the processor, a training dataset to include a data point comprising the follow-up information, the image, and the patient data; and updating, by the processor, the ML model by re-training with the augmented training dataset (see: Rezaei, Fig. 4D; and paragraph 30, 32, 76, and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), and the training data includes labels for each individual as to whether they encounter a CVD event (e.g., heart failure) after the fundus images been taken and meta-information has been recorded. Therefore, we can measure the Al model predicted risk with the truth using the cross-entropy loss…the back-propagation method is used to calculate the gradients…the parameters are updated at the negative gradients direction).
As per claim 5, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the first disease comprises one of: obstructive sleep apnea (OSA), anemia, heart disease, kidney disease, multiple sclerosis (MS), or Alzheimer's disease (see: Rezaei, Fig. 4D; and paragraph 94, is met by kidney disease).
As per claim 6, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the ML model is trained, based on a training dataset, to identify, based on the image and the patient data as inputs, the confidence level associated with the first disease (see: Rezaei, Fig. 4D; and paragraph 30, 32, 76, and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), and the training data includes labels for each individual as to whether they encounter a CVD event (e.g., heart failure) after the fundus images been taken and meta-information has been recorded. Therefore, we can measure the Al model predicted risk with the truth using the cross-entropy loss).
As per claim 10, Rezaei teaches a system, comprising:
memory; a processor; and computer-executable instructions stored in the memory and executable by the processor to perform operations comprising (see: Rezaei, Fig. 1; and paragraph 48, is met by the processing system 1002 may have processing facilities represented by one or more processors 1004. memory 1006, and other components typically present in such computing environments. In the exemplary embodiment illustrated the memory 1006 stores information accessible by processor 1004, the information comprising instructions 1008 that may be executed by the processor 1004 and data 1010 that may be retrieved, manipulated or stored by the processor 1004):
receiving: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients , and respective retinal images corresponding to the disease diagnoses (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc; paragraph 54, is met by Because not all measurements are related to the CVD risk, irrelevant columns were discarded according to the expert advice. As a result, 35 columns corresponding to 21 fields remained, including: age, sex, ethnicity, deprivation score, family history, smoking, systolic blood pressure, BMI, TC/HDL, HbAlc, state of diabetes (Y /N), diabetic type, atrial fibrillation, anti hypertensives, antithrombotic medication, lipid lowering medication, eGFR, metolazone prior 6 months, lipids in prior 6 months, LLD prior 6 months, anticoagulation medication prior 6 months, antiplay prior 6 months, CVD event and date, etc.; and paragraph 96-99, is met by disclose that the UK Biobank data was used);
receiving an image of a retina of an eye of a patient (see: Rezaei, Fig. 2A; and paragraph 56, is met by at input stage 2002 one or more fundus images are received - for example a collection of fundus photographs of an individual);
receiving patient data corresponding to the patient (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc);
determining, by providing the image and the patient data as inputs to the trained ML model, a confidence level indicative of a likelihood that the first disease is associated with the patient (see: Rezaei, Fig. 2A; and paragraph 74-75, is met by the processing pipelines of fundus images 2016 and meta-information 2020 are completed, the individual-level fundus image feature vector and meta-information vector are concatenated together to form an individual feature vector at step 2022 and that the individual feature vector is processed by a CVD risk prediction neural network model 2024 utilizing a fully connected neural network (FCNN). See also Figs. 4A-D);
determining, based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease (see: Rezaei, Fig. 4D; and paragraph 32 and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), a referral recommendation 4076 is included in the report - for example recommending a consultation with a cardiologist); and
providing, to an electronic medical record (EMR) of the patient, an output indicating the recommendation (see: Rezaei, Fig. 4D; and paragraph 89, is met by exemplary individual report; may be sent to a healthcare provider for further follow ups).
Rezaei fails to specifically teach the following limitations met by Fukushima as cited:
determining, based on the anonymized patient information, a first set of features (see: Fukushima, page 9, is met by the input data are bioinformation, such as blood pressure, BMI index, age, sex, medical history, or smoking habit);
determining, based on the retinal images, a second set of features (see: Fukushima, page 9, is met by the input data are feature quantities obtained from fundus images, such as arterial diameter, venous diameter, ratio of arterial diameter to venous diameter, branching angle of blood vessels, asymmetry of the branching, arterial venous stenosis or vascular tortuosity);
determining a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features (see: Fukushima, page 10, is met by the correlation is learned between the feature quantities and the probability of developing diseases assessed from the bioinformation);
training a machine learning (ML) model based on the correlation (see: Fukushima, page 9-10, is met by inputting the feature amounts and biometric information acquired to a trained model generated in advance, thereby estimating the risk of the subject developing a disease);
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the model training and data as taught by Rezaei and Sharma to include inputting the feature amounts and biometric information acquired to learn a correlation between the feature quantities and the probability of developing diseases assessed from the bioinformation in order to train a model to estimate the risk of the subject developing a disease as taught by Fukushima with the motivation of improving the accuracy of disease risk estimation at least by inputting multiple features (see: Fukushima, page 9-10).
As per claim 13, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 10, and further teach:
receiving follow-up information indicating whether the patient was diagnosed with the first disease; augmenting a training dataset to include a data point comprising the follow-up information, the image, and the patient data; and updating the ML model by re-training with the augmented training dataset (see: Rezaei, Fig. 4D; and paragraph 30, 32, 76, and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), and the training data includes labels for each individual as to whether they encounter a CVD event (e.g., heart failure) after the fundus images been taken and meta-information has been recorded. Therefore, we can measure the Al model predicted risk with the truth using the cross-entropy loss…the back-propagation method is used to calculate the gradients…the parameters are updated at the negative gradients direction).
As per claim 15, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 10, and further teach:
wherein the first disease is one of: obstructive sleep apnea (OSA), anemia, heart disease, kidney disease, multiple sclerosis (MS), or Alzheimer's disease (see: Rezaei, Fig. 4D; and paragraph 94, is met by kidney disease).
Claim(s) 8, 16, and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over WO 2023/214890 to Rezaei in view of U.S. Patent Application Publication 2020/0388386 to Sharma further in view of JP-7596092-B2 to Fukushima.
As per claim 8, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the anonymized patient information (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc; paragraph 54, is met by Because not all measurements are related to the CVD risk, irrelevant columns were discarded according to the expert advice. As a result, 35 columns corresponding to 21 fields remained, including: age, sex, ethnicity, deprivation score, family history, smoking, systolic blood pressure, BMI, TC/HDL, HbAlc, state of diabetes (Y /N), diabetic type, atrial fibrillation, anti-hypertensives, antithrombotic medication, lipid lowering medication, eGFR, metolazone prior 6 months, lipids in prior 6 months, LLD prior 6 months, anticoagulation medication prior 6 months, antiplay prior 6 months, CVD event and date, etc.; and paragraph 96-99, is met by disclose that the UK Biobank data was used) and the respective retinal images are extracted (see: Rezaei, Fig. 2A; and paragraph 67-69 and 73, is met by good quality images with related labels pass through a plurality of Al models. These Al models include sets of risk contributing factor (RCF) CNNs 201 0a to 201 On that are trained to detect indicators of: glycaemic control, blood pressure, cholesterol, and exposure to smoking. These indicators include, but are not limited to: drusen appearance, clustering, and/or location; pigmentation change in density and/or location; arteriovenous crossing; change in arteriovenous crossing calibre and/or thickness change; arteriovenous tortuosity; retinal oedema size and/or pattern; and/or microaneurysms concentration. Moreover, these outputted features are then concatenated to form a one-dimensional array (i.e. the individual-level fundus image feature vector) at 2016).
Rezaei fails to specifically teach receiving the patient data and images from an electronic medical records (EMR) system; however, Sharma teaches patient data and/or medical image may be obtained from memory, such as a computerized patient record or a picture archiving and communications system (see: Sharma, paragraph 32 and 42).
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the receiving of the metadata and images as taught by Rezaei and Fukushima to include obtaining patient data and/or medical image from a computerized patient record or a picture archiving and communications system as taught by Sharma with the motivation of providing patient-specific optimization (see: Sharma, paragraph 45).
As per claim 16, Rezaei teaches a non-transitory computer-readable storage medium storing processor-executable instructions that, when executed, cause one or more processors to (see: Rezaei, Fig. 1; and paragraph 48, is met by the processing system 1002 may have processing facilities represented by one or more processors 1004. memory 1006, and other components typically present in such computing environments. In the exemplary embodiment illustrated the memory 1006 stores information accessible by processor 1004, the information comprising instructions 1008 that may be executed by the processor 1004 and data 1010 that may be retrieved, manipulated or stored by the processor 1004):
receiving: anonymized patient information of a plurality of patients, respective disease diagnoses of the plurality of patients , and respective retinal images corresponding to the disease diagnoses (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc; paragraph 54, is met by Because not all measurements are related to the CVD risk, irrelevant columns were discarded according to the expert advice. As a result, 35 columns corresponding to 21 fields remained, including: age, sex, ethnicity, deprivation score, family history, smoking, systolic blood pressure, BMI, TC/HDL, HbAlc, state of diabetes (Y /N), diabetic type, atrial fibrillation, anti hypertensives, antithrombotic medication, lipid lowering medication, eGFR, metolazone prior 6 months, lipids in prior 6 months, LLD prior 6 months, anticoagulation medication prior 6 months, antiplay prior 6 months, CVD event and date, etc.; and paragraph 96-99, is met by disclose that the UK Biobank data was used);
receive, from an optical imaging device, an image of a retina of an eye of a patient (see: Rezaei, Fig. 2A; and paragraph 56, is met by at input stage 2002 one or more fundus images are received - for example a collection of fundus photographs of an individual);
access data of the patient (see: Rezaei, Fig. 2A; and paragraph 73, is met by meta-information of an individual associated with the one or more fundus images is received and that meta-information includes gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, and other is numerical data such as age, HbAlc, etc);
determine, providing the image and the EMR data as inputs to the trained ML model, a confidence level indicative of a likelihood that the first disease is associated with the patient (see: Rezaei, Fig. 2A; and paragraph 74-75, is met by the processing pipelines of fundus images 2016 and meta-information 2020 are completed, the individual-level fundus image feature vector and meta-information vector are concatenated together to form an individual feature vector at step 2022 and that the individual feature vector is processed by a CVD risk prediction neural network model 2024 utilizing a fully connected neural network (FCNN). See also Figs. 4A-D); and
determine, based on the confidence level being higher than a threshold, a recommendation for screening of the patient based on the first disease (see: Rezaei, Fig. 4D; and paragraph 32 and 94, is met by instances in which the CVD risk is above a threshold (in this example 15%), a referral recommendation 4076 is included in the report - for example recommending a consultation with a cardiologist).
Rezaei fails to specifically teach receiving the patient data from an electronic medical record (EMR) storage, EMR data; however, Sharma teaches patient data and/or medical image may be obtained from memory, such as a computerized patient record or a picture archiving and communications system (see: Sharma, paragraph 32 and 42).
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the receiving of the metadata as taught by Rezaei to include obtaining patient data and/or medical image from a computerized patient record or a picture archiving and communications system as taught by Sharma with the motivation of providing patient-specific optimization (see: Sharma, paragraph 45).
Rezaei and Sharma fail to specifically teach the following limitations met by Fukushima as cited:
determining, based on the anonymized patient information, a first set of features (see: Fukushima, page 9, is met by the input data are bioinformation, such as blood pressure, BMI index, age, sex, medical history, or smoking habit);
determining, based on the retinal images, a second set of features (see: Fukushima, page 9, is met by the input data are feature quantities obtained from fundus images, such as arterial diameter, venous diameter, ratio of arterial diameter to venous diameter, branching angle of blood vessels, asymmetry of the branching, arterial venous stenosis or vascular tortuosity);
determining a correlation between a first disease of the disease diagnoses and: (i) the first set of features, and (ii) the second set of features (see: Fukushima, page 10, is met by the correlation is learned between the feature quantities and the probability of developing diseases assessed from the bioinformation);
training a machine learning (ML) model based on the correlation (see: Fukushima, page 9-10, is met by inputting the feature amounts and biometric information acquired to a trained model generated in advance, thereby estimating the risk of the subject developing a disease);
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the model training and data as taught by Rezaei and Sharma to include inputting the feature amounts and biometric information acquired to learn a correlation between the feature quantities and the probability of developing diseases assessed from the bioinformation in order to train a model to estimate the risk of the subject developing a disease as taught by Fukushima with the motivation of improving the accuracy of disease risk estimation at least by inputting multiple features (see: Fukushima, page 9-10).
As per claim 19, Rezaei, Sharma, and Fukushima teach the invention as claimed, see discussion of claim 16, and further teach:
wherein: the first set of features comprises at least one of: an age of the patient, a blood pressure measurement of the patient, or one or more medical test results associated with the patient, and the second set of features comprises at least one of: a brightness level of an optic disc of the retina, a diameter of blood vessels of the retina, an edema of the optic disc, or an arteriovenous ratio (AVR) (see: Rezaei, paragraph 73 is met by gender, ethnicity, HbAlc, TCHDL, etc. Some of the meta-information is categorical data such as gender, ethnicity, deprivation value, medicine, etc, and other is numerical data such as age, HbAlc, etc.).
As per claim 20, Rezaei, Sharma, and Fukushima teach the invention as claimed, see discussion of claim 16, and further teach:
wherein the first disease comprises one of: obstructive sleep apnea (OSA), anemia, heart disease, kidney disease, multiple sclerosis (MS), or Alzheimer's disease (see: Rezaei, Fig. 4D; and paragraph 94, is met by kidney disease).
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over WO 2023/214890 to Rezaei in view of JP-7596092-B2 to Fukushima further in view of U.S. Patent Application Publication 2024/0029885 to Holden.
As per claim 9, Rezaei and Fukushima teach the invention as claimed, see discussion of claim 1, and further teach:
wherein the ML model indicating rules correlating one or more first features of the first set of features and one or more second features of the second set of features with a probability of occurrence of the first disease (see: Rezaei, Fig. 2A; and paragraph 74-75, is met by the processing pipelines of fundus images 2016 and meta-information 2020 are completed, the individual-level fundus image feature vector and meta-information vector are concatenated together to form an individual feature vector at step 2022 and that the individual feature vector is processed by a CVD risk prediction neural network model 2024 utilizing a fully connected neural network (FCNN). See also Figs. 4A-D).
Rezaei fails to specifically teach that the neural network model comprises an expert system; however, Holden teaches expert/AI system (hereafter referred to simply as the “expert system”) can include a traditional rules-based expert system and/or a machine learning system, the machine learning system including neural networks and/or knowledge graphs (see: Holden, paragraph 5, 44, and 61).
It would have been obvious to one of ordinary skill in the art at the time the invention was field to modify the neural network model as taught by Rezaei and Fukushima to include an expert/AI system as taught by Holden with the motivation of selecting an appropriate course of action (see: Holden, paragraph 45).
Novelty of Claims
As per claims 21-26, the closest prior art of record, WO 2023/214890 to Rezaei and JP-7596092-B2 to Fukushima, neither alone nor in combination teach the new limitations as they do not teach, for example, in combination with the other claimed limitations from which they depend, “identifying, by the processor, first retinal images corresponding to the first disease diagnosis, each retinal image of the first retinal images being associated having a respective time of capture; determining, by the processor and based on the first retinal images and the respective times of capture, a temporal trend of a retinal feature represented in the first retinal images; and determining, by the processor and based on the date of disease diagnosis, an additional correlation between the first disease and the temporal trend, wherein the ML model is further trained based on the additional correlation”; therefore, the closest prior art of record does not anticipate or otherwise render the claimed invention obvious. Prior art Fukushima teaches, for exmaple, that the “type of image in the input data at the time of estimation (prediction) may be displayed together with information indicating the likelihood of each type of imageincluded in the correct answer data at the time of learning (for example, a numerical value indicating a ratio)” (see: Fukushima, page 27-28), and that generally its deep learning may include “a neural network thathandles time-series information” (see: Fukushima, page 29-30), but does not detial the temporal tred of a retinal feature, for example, such that correlation between the first disease and said trend is further used to train the ML model. Hence, rejection of the claimed combination of limitations would require an unreasonable combination of the available prior art. Associated dependent claims are novel for incorporating such features.
Response to Arguments
Applicant’s arguments from the response filed on 06/09/2026 have been fully considered and will be addressed below in the order in which they appeared.
In the remarks, Applicant argues in substance that (1) the 35 U.S.C. 101 rejections should be withdrawn because “the independent claims, as amended, recite features that integrate the alleged abstract idea into a practical application of the alleged abstract idea at least because the claims recite features directed to training a specialized machine learning (ML) model based on determined correlation between two different types of features: a first set based on patient information, and a second set based on retinal images. The claimed features are also directed to determining and providing a recommendation to assist a medical provider in monitoring a health status of a patient. For instance, amended claim 1 recites "determining, by the processor, a correlation between a first disease of the disease diagnoses and: (i) the first set of features [based on the anonymized patient information], and (ii) the second set of features [based on the retinal images]," "training, by the processor, a machine learning (ML) model based on the correlation," and "determining ... by providing the image and the patient data as inputs to the trained ML model, a confidence level indicative of a likelihood that the first disease is associated with the patient." At least these features are directed to automatically determining a correlation between features in an anonymized data-set in a data-driven manner, to determine whether a patient is to be recommended for further screening for a disease based on patient data and a retinal image of the patient. The claimed features also include training a ML model based on the correlation to generate a confidence level of the disease in the patient, which cannot be fairly characterized as "extra-solution activity."”
The Examiner respectfully disagrees. Applicant’s arguments are not persuasive.
Similarly to Office Example 47, claim 2, the claimed machine learning in the present application is ineligible because the training is merely “based on” the determined correlation where such training is being performed by a computer such that it amounts to no more than mere instructions to apply the exception using a generic computer. The “machine learning (ML) model” is “trained” through no more than a statement than that training is “based on” a correlation of features or “based on” a training dataset to produce desired results, and that “re-training” is performed “with” the augmented training dataset or the model is further trained “based on” an additional or temporal correlation. With such high-level training, there is no improvement to the training techniques themselves or way in which the training is employed to achieve the desired results claimed. The claims here are not directed to a specific improvement to computer functionality that amount to a practical application. Rather, they are directed to the use of conventional or generic technology in a well-known environment, without any claim that the invention reflects an inventive solution to a technical problem presented by combining the two. In the present case, the claims fail to recite any elements that individually or as an ordered combination transform the identified abstract idea(s) in the rejection into a patent-eligible application of that idea.
In the remarks, Applicant argues in substance that (2) the 35 U.S.C. 103 rejections should be withdrawn in view of the amendments because “Rezaei describes determining an "overall risk of CVD [cardio-vascular disease] to an individual associated with processed fundus images" which also takes into account "contributing factors to the individual CVD risk ... (e.g., based on patient meta-information such as age, gender, and/or ethnicity) and modifiable contributing factors ( e.g., based on glycemic control, blood pressure, cholesterol, and exposure to smoking)," (Rezaei, para. [0030]) Rezaei is understood to describe a predetermined set of contributing factors as inputs to a CVD prediction model. ( emphasis added). Rezaei fails to describe at least the features of amended claim 1 reciting "determining, by the processor, a correlation between a first disease ... and (i) the first set of features ... and (ii) the second set of features," and "training, by the processor, a machine learning (ML) model based on the correlation." Applicant notes that providing a predetermined set of contributing factors for a disease, as Rezaei describes, does not satisfy the above features of amended claim 1, where the correlation between the sets of features and the first disease is determined by the processor, and the ML model is trained based on the determined correlation. Sharma is not cited for, and does not teach or suggest these features, individually, or in combination with Rezaei.”
Applicant’s arguments have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument - see application of prior art Fukushima in the rejection above. However, with regard to new claims 21-26, the claims are novel when considered in combination with the parent claims from which they depend for the reasons given above in the Novelty of Claims section.
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 ROBERT A SOREY whose telephone number is (571)270-3606. The examiner can normally be reached Monday through Friday, 8am to 5pm.
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, Fonya Long can be reached at (571) 270-5096. 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.
/ROBERT A SOREY/Primary Examiner, Art Unit 3682
Prosecution Insights