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 .
Specification
The disclosure is objected to because of the following informalities:
Examiner suggests amending Para [0001] of the Specification to state: “ This application is a continuation of and claims the benefit of priority to U.S. Patent Application Serial No. 18/150,493, now U.S. Patent No. 11,974,863 entitled “GLUCOSE ESTIMATION WITHOUT CONTINUOUS GLUCOSE MONITORING” and filed January 5, 2023, which is a continuation of and claims the benefit of priority to U.S. Patent Application Serial No. 17/178,087 entitled “MACHINE LEARNING-BASED SYSTEM FOR ESTIMATING GLUCOSE VALUES” and filed February 17, 2021, which is a continuation of and claims the benefit of priority to U.S. Patent Application Serial No. 16/987,330 entitled “MACHINE LEARNING-BASED SYSTEM FOR ESTIMATING GLUCOSE VALUES” and filed August 6, 2020, each of which is incorporated herein by reference”
Appropriate correction is required.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-15 are rejected under 35 U.S.C. 102 (a)(1) as being anticipated by US 2017/0252513 Buck, JR. et al., hereinafter “Buck” (cited previously).
Regarding claim 1, Buck discloses a processor-implemented method (Abstract and Para 22) comprising: receiving input data associated with a user via a window filter (Figure 4, element 40), the window filter configured to divide the input data into a plurality of segments corresponding to a respective plurality of time windows (See Figure 6 that clearly shows the input data 40 being segmented in a plurality of different time segments such as 70, 71, and 72, and the time periods in between), each of the plurality of segments of the input data comprising discrete glucose measurement data associated with the user (Figure 4, element 40 and Para 28 and Figure 6, element 40), and at least a portion of contextual data associated with the user (Para 57); and using an estimation model (Figure 4, elements 52 and 62) and at least a portion of the one or more segments of input data associated with the user received via the window filter (Figure 4, element 40 and Figure 6, element 40, they could use all the data or just the good time segments as described by Para 46), generating, in real-time, one or more estimated glucose values associated with the user corresponding to at least one of the plurality of time windows (Figure 4, predicted glucose level; see also Para 57; Figure 7 clearly shows that the prediction is happening in real-time as the actual values vs predicted values are happening sequentially on the timeline as shown); wherein the estimation model is personalized to the user in real-time based on training glucose measurement data associated with the user received from a blood glucose meter and training activity data associated with the user (After further consideration of reference Buck, examiner believes this limitation is covered by the reference. Consider Para 48, 54, 64-66 and Figure 6; the recursive filter that is receiving the data is disclosed to be a Kalman filter in Para 48. Consider this “The state vector may be updated every time a new input is received (i.e., recursively).” Which inherently means, as new data comes in, the state vector is updated and a new glucose prediction is generated. Para 65 also says exactly that “As new data are measured they may be incorporated into the prediction algorithm and/or training function. There are many possible algorithms for including new data. These include adding the data to the training set when 1) A predetermined period of time has elapsed, 2) The prediction failed on the specific data, 3) The input data is not represented in the training set, or 4) A patient or care provider manually includes the data, including all new data, if suitable” This disclosure is also clearly showing that new data coming in that is not in the training set is incorporated in there immediately, i.e. in real-time. This therefore creates a personalized prediction algorithm that is based on the users measured blood glucose, from a blood glucose meter, consider Para 54 and 67; Para 57 discloses that the prediction algorithm learns a user specific prediction model indicated by the training model, this prediction algorithm is augmented with input data, which includes exercise data, i.e. activity data. As for the “real-time” limitation, examiner believes Buck does disclose the model being personalized in real-time. Refer to Para 55, 57, and 62-63, Figure 7; As can be seen in the figure, the estimated glucose level 60 is being determined in Real-time as the blood glucose measurements 40 are taken, soon after, the estimation/predicted glucose continues to output data, while the measurements 40 stop. The prediction algorithm has to be working in real-time to create a personalized model to be able to predict glucose at the same time measurements come in and have a fully personalized model at t=0 to continue predicting measurements without actual measurements 40).
Regarding claim 2, Buck discloses controlling an insulin delivery device (Figure 2, element 31) based on the generated one or more estimated glucose values (Para 35).
Regarding claim 3, Buck discloses the generating of the one or more estimated glucose values associated with the user comprises generating a plurality of estimated glucose values associated with the user, the plurality of estimated glucose values comprising a combination of at least (i) one or more estimated glucose values associated with a first time window of the plurality of time windows and (ii) one or more estimated glucose values associated with a second time window of the plurality of time windows (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements).
Regarding claim 4, Buck discloses the generating the one or more estimated glucose values associated with the user further comprises joining or linking the at least (i) one or more estimated glucose values associated with the first time window of the plurality of time windows and (ii) one or more estimated glucose values associated with the second time window of the plurality of time windows into the generated one or more estimated glucose values (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements; this limitation simply means using all the measurements).
Regarding claim 5, Buck discloses the one or more estimated glucose values associated with the second time window is generated based at least on at least a portion of the one or more estimated glucose values associated with the first time window provided as input to the estimation model (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements; the values 60 are predicted in real-time, therefore it is using the time segment just before to predict the ones in real-time so every second time period will rely on data from the first).
Regarding claim 6, Buck discloses the respective plurality of time windows include the first time window and the second time window; and each of the first time window and the second time window is sequential and comprises a time period that is less than a time period over which the input data is collected (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements; the time periods can be sequential as shown by Figure 6 and a time period such as elements 70, 71, 72, or the periods in between are all less than the entire measurement period shown).
Regarding claim 7, Buck discloses the at least the portion of the one or more segments of input data corresponds to a specified time period (Para 19; over a period of time); and the estimation model is configured to generate the one or more estimated glucose values associated with the user based on the at least portion of the one or more segments of input data over the specified time period (Para 19 and Figure 7).
Regarding claim 8, Buck discloses the discrete glucose measurement data associated with the user is obtainable from a blood glucose meter configured to measure glucose levels directly from blood (Para 19).
Regarding claim 9, Buck discloses applying a calibration model to the estimation model, the calibration model configured to reduce an error associated with the generated one or more estimated glucose values as compared to without applying the calibration model (Para 19 and 28).
Regarding claim 10, Buck discloses the calibration model is generated based on one or more calibration sources, the one or more calibration sources (Para 28) comprising: discrete glucose measurement data from a blood glucose meter configured to measure glucose levels directly from blood; historical data relating to estimated glucose values associated with the user; statistical data relating to the estimated glucose values associated with the user; variability data relating to the estimated glucose values associated with the user; or a combination thereof (Para 19 and 28 disclose the calibration of discrete glucose measurement data).
Regarding claim 11, Buck discloses a system (Abstract and Figure 2, element 26) comprising: one or more processors (Figure 2, element 32 and Para 22); and a computer-readable apparatus comprising a storage medium, the storage medium comprising a plurality of instructions (Para 22 and 23) configured to, when executed by the one or more processors, cause the system to: receive input data associated with a user via a window filter (Figure 4, element 40), the window filter configured to divide the input data into a plurality of segments corresponding to a respective plurality of time windows (See Figure 6 that clearly shows the input data 40 being segmented in a plurality of different time segments such as 70, 71, and 72, and the time periods in between), each of the plurality of segments of the input data comprising discrete glucose measurement data associated with the user (Figure 4, element 40 and Para 28 and Figure 6, element 40), and at least a portion of contextual data associated with the user (Para 57); and using an estimation model (Figure 4, elements 52 and 62) and at least a portion of the plurality of segments of input data associated with the user received via the window filter (Figure 4, element 40 and Figure 6, element 40, they could use all the data or just the good time segments as described by Para 46), generate, in real-time, one or more estimated glucose values associated with the user corresponding to at least one of the plurality of time windows (Figure 4, predicted glucose level; see also Para 57; Figure 7 clearly shows that the prediction is happening in real-time as the actual values vs predicted values are happening sequentially on the timeline as shown); wherein the estimation model is personalized to the user in real-time based on training glucose measurement data associated with the user received from a blood glucose meter and training activity data associated with the user (After further consideration of reference Buck, examiner believes this limitation is covered by the reference. Consider Para 48, 54, 64-66 and Figure 6; the recursive filter that is receiving the data is disclosed to be a Kalman filter in Para 48. Consider this “The state vector may be updated every time a new input is received (i.e., recursively).” Which inherently means, as new data comes in, the state vector is updated and a new glucose prediction is generated. Para 65 also says exactly that “As new data are measured they may be incorporated into the prediction algorithm and/or training function. There are many possible algorithms for including new data. These include adding the data to the training set when 1) A predetermined period of time has elapsed, 2) The prediction failed on the specific data, 3) The input data is not represented in the training set, or 4) A patient or care provider manually includes the data, including all new data, if suitable” This disclosure is also clearly showing that new data coming in that is not in the training set is incorporated in there immediately, i.e. in real-time. This therefore creates a personalized prediction algorithm that is based on the users measured blood glucose, from a blood glucose meter, consider Para 54 and 67; Para 57 discloses that the prediction algorithm learns a user specific prediction model indicated by the training model, this prediction algorithm is augmented with input data, which includes exercise data, i.e. activity data. As for the “real-time” limitation, examiner believes Buck does disclose the model being personalized in real-time. Refer to Para 55, 57, and 62-63, Figure 7; As can be seen in the figure, the estimated glucose level 60 is being determined in Real-time as the blood glucose measurements 40 are taken, soon after, the estimation/predicted glucose continues to output data, while the measurements 40 stop. The prediction algorithm has to be working in real-time to create a personalized model to be able to predict glucose at the same time measurements come in and have a fully personalized model at t=0 to continue predicting measurements without actual measurements 40).
Regarding claim 12, Buck discloses the plurality of instructions are further configured to control an insulin delivery device (Figure 2, element 31) based on the generated one or more estimated glucose values (Para 35).
Regarding claim 13, Buck discloses the generation of the one or more estimated glucose values associated with the user comprises generation of a plurality of estimated glucose values associated with the user, the plurality of estimated glucose values comprising a combination of at least (i) one or more estimated glucose values associated with a first time window of the plurality of time windows and (ii) one or more estimated glucose values associated with a second time window of the plurality of time windows (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements).
Regarding claim 14, Buck discloses the generation the one or more estimated glucose values associated with the user further comprises joining or linking the at least (i) one or more estimated glucose values associated with the first time window of the plurality of time windows and (ii) one or more estimated glucose values associated with the second time window of the plurality of time windows into the generated one or more estimated glucose values (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements; this limitation simply means using all the measurements).
Regarding claim 15, Buck discloses the one or more estimated glucose values associated with the second time window is generated based at least on at least a portion of the one or more estimated glucose values associated with the first time window provided as input to the estimation model (Para 19 and 62, see also Figures 6 and 7; the model is using the measured blood glucose levels at a plurality of time periods, including the hypothetical first and second time windows, and outputting estimated values 60 based on those measurements; the values 60 are predicted in real-time, therefore it is using the time segment just before to predict the ones in real-time so every second time period will rely on data from the first).
Allowable Subject Matter
Claims 16-20 are allowed.
Reasons for Allowance
The following is an examiner’s statement of reasons for allowance:
Examiner relies on reference (US 2017/0252513 Buck) to disclose a processor/method that receive input data of blood glucose measurements along with contextual data and feeds them into a learning model in real-time to estimate blood glucose values, thereby enabling the measuring of glucose without it being invasive. However, no reference was found to measure the estimated glucose value only based on the contextual data, i.e. without using discrete glucose measurements. In other words, no reference was found to estimate blood glucose values solely based on contextual data and not utilizing the trained blood glucose values. For this reason, claims 16-20 are allowed.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Response to Arguments
Applicant’s arguments have been fully considered but are moot because the new ground of rejection.
Examiner respectfully disagrees with applicants view point of reference Buck in not teaching the newly amended limitations. Para 57 shows that a patient specific prediction model is generated using the prediction algorithm trained data. This model utilizes the “estimated glucose levels” and “other input data, including meal times, carbohydrates, medications, exercise”. Examiner believes Buck creates a personalized model based on trained data including the ones listed above as the limitation states in the claim.
Regarding applicants arguments on “real-time” examiner also respectfully disagrees. Kindly consider Para 55, 57, and 62-63 and Figure 7. As can be seen in Figure 7, the estimated glucose level 60 is being determined in Real-time as the blood glucose measurements 40 are taken, soon after, the estimation/predicted glucose continues to output data, while the measurements 40 stop. The prediction algorithm has to be working in real-time to create a personalized model to be able to predict glucose at the same time measurements come in and have a fully personalized model at t=0 to continue predicting measurements without actual measurements 40.
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AYA ZIAD BAKKAR whose telephone number is (313)446-6659. The examiner can normally be reached on 7:30 am - 5:00 pm M-Th.
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, Carl Layno can be reached on (571) 272-4949. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/AYA ZIAD BAKKAR/
Examiner, Art Unit 3796
/CARL H LAYNO/Supervisory Patent Examiner, Art Unit 3796
Prosecution Insights