Automatic reminding system and reminding method for medicine bottle

Abstract

The present application relates to the technical field of medicine taking reminding, in particular to an automatic reminding system and method for medicine bottles. The system comprises an acquisition module for acquiring historical taking time and taking interval; an influence analysis module for determining the taking influence weight of the taking interval and the fitting error of the correction interval data corresponding to each taking interval; then determining the influence degree of the taking interval; an interval prediction module for data prediction to determine the predicted taking interval of the next taking; weighting the predicted taking interval according to the influence degree to determine the target interval of the next taking; a reminding record module for determining the target time of the next taking according to the target interval and the current taking time, and reminding and recording the next taking according to the target time. The present application can adaptively adjust the reminding duration according to all taking time, thereby enhancing the reminding effect of the medicine bottle.

Description

Technical Field

[0001] The present invention relates to the technical field of medicine taking reminders, and in particular to an automatic reminder system and reminder method for a medicine bottle. Background Art

[0002] The medicine bottle with smart label, automatic reminder and recording system can remind users to take medicine and record the amount of medicine used each time. It can remind users to take medicine when they need to take medicine, so as to achieve the purpose of allowing users to take medicine in time.

[0003] In the related art, a timer alarm is directly set to remind the user to take medicine, for example, once every 8 hours, or once at a fixed time every day (such as 12 o'clock). In this way, due to changes in the user's life, the user may not be able to take the medicine in time. For example, the medicine should be taken at 12 o'clock in the afternoon, but due to being away from home or other reasons, it is delayed until 2 o'clock in the afternoon. In addition, the overall medication time is not fixed. At this time, it is too rigid to still remind according to the fixed time, which will affect the effect of the medicine bottle reminder. Summary of the Invention

[0004] In order to solve the technical problem in the related art that reminders based on fixed time are too rigid, which leads to disordered medication and thus affects the reminder effect of the medicine bottle, the present invention provides an automatic reminder system for medicine bottles, and the technical solution adopted is as follows:

[0005] The present invention proposes an automatic reminder system for medicine bottles, comprising:

[0006] An acquisition module is used to obtain the historical access times of the user accessing the contents of the medicine bottle, and determine the time interval between two adjacent access times as the access interval for the latter of the two adjacent access times;

[0007] The impact analysis module is used to determine the corrected interval data for each access interval based on the duration of all access intervals; determine the access impact weight of the access interval based on the corrected interval data of any access interval and the preset number of access intervals closest to it in the time series; perform data fitting on all corrected interval data in the time series to determine the fitting error of the corrected interval data corresponding to each access interval; and determine the impact degree of the access interval by combining the access impact weight and the fitting error;

[0008] An interval prediction module is used to perform data prediction after the current access by combining all the corrected interval data in the time series to determine the predicted access interval for the next access; weight the duration of the predicted access interval according to the impact degree of the predicted access interval, and determine the target interval for the next access in combination with the access interval of the current access;

[0009] The reminder recording module is used to determine the target time of next access according to the target interval and the current access time, and to remind and record the next access according to the target time.

[0010] Furthermore, the determining of the corrected interval data of each access interval according to the duration of all access intervals includes:

[0011] Calculate the average duration of all the intervals, and use the difference between the duration of each interval and the average duration as the corrected interval data.

[0012] Furthermore, the determining of the access impact weight of an access interval based on the corrected interval data of a preset number of access intervals that are closest in time sequence to any access interval includes:

[0013] When the corrected interval data of two adjacent access intervals in time sequence are one positive and one negative, the latter of the two adjacent access intervals is marked as a first mark, otherwise, it is marked as a second mark, wherein the first mark is different from the second mark;

[0014] The number of all two adjacent access intervals in the preset number of access intervals that are closest to any access interval in time sequence is taken as the total number of influences;

[0015] For any access interval, the ratio of the number of items marked as the second mark to the total number of influences is calculated to obtain the access influence weight of any access interval.

[0016] Furthermore, performing data fitting on all the correction interval data in the time series to determine the fitting error of the correction interval data corresponding to each sampling interval includes:

[0017] A two-dimensional coordinate system is constructed with the time series as the horizontal axis and the value of the correction interval data as the vertical axis to determine the two-dimensional coordinates used each time;

[0018] Perform curve fitting on all the two-dimensional coordinates taken for each time to determine the fitting curve;

[0019] The Euclidean distance between the two-dimensional coordinates at each sampling moment and the coordinates at the same moment in the fitting curve is taken as the fitting error of the correction interval data corresponding to each sampling interval.

[0020] Furthermore, the combining of the access influence weight and the fitting error to determine the influence degree of the access interval includes:

[0021] The product of the access impact weight of the access interval and the fitting error of the corresponding correction interval data is calculated, and the opposite of the product value is normalized to the maximum and minimum values ​​to obtain the influence degree of the access interval.

[0022] Furthermore, the step of combining all the corrected interval data in the time series to perform data prediction and determine the predicted access interval for the next access includes:

[0023] Based on the least squares method, the least squares analysis is performed on all the correction interval data in the time series to obtain the predicted correction data for the next use;

[0024] The sum of the predicted correction data and the average of the durations of all the access intervals is used as the predicted access interval for the next access.

[0025] Furthermore, weighting the duration of the predicted access interval according to the impact degree of the predicted access interval and determining the target interval for the next access in combination with the current access interval includes:

[0026] Calculate the difference between the unit value 1 and the influence degree as the current weight;

[0027] Calculate the product of the impact degree and the predicted access interval as a first weighted index; calculate the product of the current weight and the currently access interval as a second weighted index;

[0028] The sum of the first weighted index and the second weighted index is used as the target interval for next use.

[0029] Furthermore, the next access is reminded and recorded according to the target time, including:

[0030] Determine the preset reminder advance time, and use the preset reminder advance time before the target time as the reminder time. At the reminder time, control the medicine bottle to send a reminder signal, and detect whether the contents of the medicine bottle are taken. When taking from the contents of the medicine bottle, record the corresponding taking time.

[0031] Furthermore, a weight sensor is installed at the bottom of the medicine bottle, and an opening monitoring device is installed at the bottle mouth to detect whether the contents of the medicine bottle have been taken out, specifically including:

[0032] For any suspected moment, if the weight detected by the weight sensor decreases before and after the suspected moment, and the opening monitoring device detects that the bottle mouth is opened at the suspected moment, it is determined that the contents are taken; otherwise, it is determined that the contents are not taken.

[0033] In order to better apply the above system, an automatic reminder method for using medicine bottles is also proposed, which includes:

[0034] Step S1: obtaining the time of the user taking out the contents of the medicine bottle in history, and determining the time interval between two adjacent taking out times as the taking interval for the latter of the two adjacent taking out times;

[0035] Step S2: determining the corrected interval data for each access interval based on the duration of all access intervals; determining the access impact weight of the access interval based on the corrected interval data of any access interval and the preset number of access intervals closest to it in the time series; performing data fitting on all corrected interval data in the time series to determine the fitting error of the corrected interval data corresponding to each access interval; and determining the degree of influence of the access interval by combining the access impact weight and the fitting error;

[0036] Step S3: After the current access, all the modified interval data in the time series are combined to perform data prediction to determine the predicted access interval for the next access; weight the duration of the predicted access interval according to the impact degree of the predicted access interval, and determine the target interval for the next access in combination with the current access interval;

[0037] Step S4: determining the target time for the next access according to the target interval and the current access time, and reminding and recording the next access according to the target time.

[0038] The present invention has the following beneficial effects:

[0039] The embodiment of the present invention determines the withdrawal interval by obtaining the historical withdrawal time of the user to withdraw the contents of the medicine bottle, and then performs an impact analysis based on the change in the length of the withdrawal intervals adjacent in time to determine the impact degree of the withdrawal interval. The impact degree represents the impact of the withdrawal interval under normal data fluctuations, that is, the more normal the withdrawal interval is among all withdrawal interval characteristics, the greater the impact degree value, and the more consistent it is with the normal medication pattern. Then, data prediction is performed in combination with all corrected interval data to determine the predicted withdrawal interval for the next withdrawal. The predicted withdrawal interval represents the withdrawal characteristics under normal circumstances. Therefore, the duration of the predicted withdrawal interval is directly weighted according to the impact degree of the predicted withdrawal interval, and the target interval for the next withdrawal is determined in combination with the withdrawal interval of the current withdrawal. The target interval can effectively combine all withdrawal intervals for analysis, thereby accurately determining the next withdrawal situation. At the same time, it can minimize the withdrawal abnormality. The target time for the next withdrawal is determined based on the target interval and the current withdrawal time, and the next withdrawal is reminded and recorded based on the target time.

[0040] In an embodiment of the present invention, data analysis is performed on the overall medication time to automatically remind users to take medicine. The automatic reminder process is not extracted based on fixed times, but analyzes medication habits to reduce abnormal medication characteristics, thereby effectively reminding users to take medicine and avoiding fixed times that do not conform to actual medication times. At the same time, the reminder duration can be adaptively adjusted according to all medication taking times, thereby enhancing the medicine bottle reminder effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A diagram of an automatic reminder system for medicine bottles provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0043] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of an automatic reminder system for medication bottles according to the present invention. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0045] The following describes in detail a specific solution of an automatic reminder system for medicine bottles provided by the present invention with reference to the accompanying drawings.

[0046] See also Figure 1 , which shows an automatic reminder system for medicine bottles provided by one embodiment of the present invention, including:

[0047] The acquisition module 101 is used to obtain the historical access times of the user accessing the contents of the medicine bottle, and determine the time interval between two adjacent access times as the access interval for the latter of the two adjacent access times.

[0048] The medicine bottle with smart label, automatic reminder and recording system can remind users to take medicine and record the amount of medicine used each time. It can remind users to take medicine when they need to take medicine, so as to achieve the purpose of allowing users to take medicine in time.

[0049] In the related art, a timer alarm is directly set to remind the user to take medicine, for example, once every 5 hours, or once at a fixed time every day (such as 12 o'clock). In this way, due to changes in the user's life, the user may not be able to take the medicine in time. For example, the medicine should be taken at 12 o'clock in the afternoon, but it is delayed until 2 o'clock in the afternoon due to being away from home or other reasons. At this time, in order to ensure the efficacy of the medicine, the next time to take the medicine needs to be delayed, and a specific analysis is needed for the time of the next medication.

[0050] It should be noted that in the embodiment of the present invention, a weight sensor is installed at the bottom of the medicine bottle, and an opening monitoring device is installed at the bottle mouth. When the weight detected by the weight sensor decreases before and after any time, and the opening monitoring device detects that the bottle mouth is opened at the corresponding time, the contents are determined to have been removed. The corresponding time is recorded as the removal time.

[0051] In an embodiment of the present invention, multiple medication collection times can be counted in advance, for example, medication collection times within 10 days before the current time, and then the time interval between two adjacent medication collection times is used as a medication collection interval, that is, the time interval between two consecutive medication collections.

[0052] The impact analysis module 102 is used to determine the corrected interval data of each access interval based on the duration of all access intervals; determine the access impact weight of the access interval based on the corrected interval data of any access interval and the preset number of access intervals closest to it in time series; perform data fitting on all corrected interval data in time series to determine the fitting error of the corrected interval data corresponding to each access interval; and determine the impact degree of the access interval by combining the access impact weight and the fitting error.

[0053] Among them, the corrected interval data represents the characteristic indicators of each medication interval in terms of duration, and is mainly used to indicate whether medication is taken early or late, as well as the specific impact duration of the early or delayed action, and further indicates abnormal conditions of the medication interval.

[0054] Since medication is a long-term process, and patients may experience unusual circumstances in their daily lives, such as being busy at work or not getting up in the morning, using the average of all the intervals between medications to remind patients to take their medications may result in users not being able to take their medications on time in some cases. Therefore, we use the intervals between patients' previous medications to predict the intervals between their next medications.

[0055] Since the average of all the interval data for each medication can be used to remind patients to take medication under normal circumstances, the present invention mainly calculates the degree to which the interval data for each medication is affected by the interval data for other medications under abnormal circumstances. Therefore, the corrected interval data for each interval data is calculated as a numerical value representing the degree of abnormality.

[0056] It can be understood that since the overall medication is close to regular, that is, all medication intervals can show a certain regularity, therefore, based on the duration of all medication intervals, a discrete characteristic analysis can be performed on each medication interval to determine the corrected interval data for each medication interval.

[0057] Furthermore, in some embodiments of the present invention, the corrected interval data of each access interval is determined based on the duration of all access intervals, including: calculating the average duration of all access intervals, and taking the difference between the duration of each access interval and the average duration as the corrected interval data.

[0058] In an embodiment of the present invention, the corrected interval data is determined in the form of mean difference, that is, when the absolute value of the corrected interval data is larger, it can be said that the difference between the interval data and the mean is larger, and the drug release of the corresponding medication at the interval is more unstable. When the absolute value of the corrected interval data is smaller, it means that it is more consistent with the mean, that is, the overall drug release is more stable.

[0059] The corrected interval data itself is the difference between the duration of the interval and the mean duration, that is, the corrected interval data has positive and negative signs, which represent the advance and delay of medication, that is, the positive sign indicates delayed medication, and the negative sign indicates early medication. Based on this, impact analysis can be performed.

[0060] In an embodiment of the present invention, a usage impact weight for any medication access interval is determined based on the corrected interval data for a preset number of medication access intervals that are closest in time series. The usage impact weight represents the effect of a corresponding medication access on subsequent medication accesses. For example, if a medication access delay occurs during a particular medication access, the next medication access will also be affected and delayed. The greater the usage impact weight of that access, the greater the fluctuation in the access interval within its temporal neighborhood.

[0061] The preset number is the number of adjacent access intervals when performing temporal neighborhood analysis on the access influence weights. Optionally, the preset number can be specifically 10, for example, and there is no limitation on this.

[0062] Furthermore, in some embodiments of the present invention, the usage impact weight of an access interval is determined based on the corrected interval data of a preset number of access intervals that are closest to any access interval in time sequence, including: when the corrected interval data of two adjacent access intervals in time sequence are one positive and one negative, the latter of the two adjacent access intervals is marked as a first mark, otherwise, it is marked as a second mark, wherein the first mark is different from the second mark; the number of all two adjacent access intervals in the preset number of access intervals that are closest to any access interval in time sequence is taken as the total impact number; for any access interval, the ratio of the number marked as the second mark to the total impact number is calculated to obtain the usage impact weight of any access interval.

[0063] It is understood that when the corrected interval data of two adjacent sampling intervals is one positive and one negative in time sequence, the later of the two adjacent sampling intervals is marked as the first mark; when the corrected interval data are all negative or all positive, the later of the two adjacent sampling intervals is marked as the second mark. It should be noted that the first mark and the second mark are merely for distinction. Specifically, the first mark can be 0 and the second mark can be 1. Of course, different letters, symbols, special characters, etc. can also be used to distinguish the first mark and the second mark, and this is not limited to this.

[0064] In an embodiment of the present invention, the number of all two adjacent access intervals in the preset number of access intervals that are closest to any access interval in time sequence is taken as the total number of influences. That is, when the preset number is 10, plus the access interval itself, there are a total of 11 access intervals, and the number of the corresponding two adjacent access intervals is 10. That is, when the preset number is 10, the total number of influences is 10.

[0065] When the mark is the second mark, it indicates a positive or negative sign, which means that the usage interval is expanded or reduced. Therefore, in the embodiment of the present invention, the ratio of the number marked as the second mark to the total number of influences is calculated to obtain the usage influence weight of any usage interval. The larger the value of the usage influence weight, the greater the influence, which is reflected in the abnormal increase or decrease of all usage intervals within the corresponding time range. In normal actual conditions, the medication usage interval should remain stable without changing its positive or negative sign, that is, the usage influence weight is at a smaller value, that is, the usage influence weight also represents the abnormal effect.

[0066] Afterwards, a fitting feature analysis is performed based on the fitting of the corrected interval data. Data fitting is performed on all corrected interval data in the time series to determine the fitting error of the corrected interval data corresponding to each access interval. This includes: constructing a two-dimensional coordinate system with the time series as the horizontal axis and the value of the corrected interval data as the vertical axis to determine the two-dimensional coordinates of each access; performing curve fitting on all the two-dimensional coordinates of each access to determine the fitting curve; and using the Euclidean distance between the two-dimensional coordinate at each access moment and the coordinate at the same moment in the fitting curve as the fitting error of the corrected interval data corresponding to each access interval.

[0067] Because patients' lives tend to be highly regular, this means that if they are not affected by major external factors, the time between medications taken on different days will be relatively similar. However, this does not mean that the time between medications taken on different days is exactly the same. Under normal circumstances, when the interval between medications increases on one day, the interval between the next medication decreases. On other days, the interval between medications increases on one day and also increases on the next. Simultaneous increases indicate an abnormal effect, which can be effectively analyzed through fitting.

[0068] It is understandable that the fitting curve is obtained by performing two-dimensional time series fitting on the modified interval data. The specific fitting process is a technique well known to those skilled in the art, such as the least squares method, and is not limited thereto.

[0069] Under normal circumstances, the overall trend should present a stable fluctuating curve, while when large fluctuations occur, it indicates an abnormal situation.

[0070] After curve fitting, the difference between the value of the fitted curve and the value of the corrected interval data at the same time can be calculated. In the embodiment of the present invention, the Euclidean distance is directly used to implement the numerical difference analysis, that is, the Euclidean distance of the coordinates is calculated as the fitting error of the corrected interval data corresponding to each sampling interval.

[0071] Therefore, the larger the fitting error, the less it corresponds to the normal fitting characteristics. Then, the fitting error and the influence weight can be combined to achieve specific impact analysis.

[0072] The influence degree of the access interval is determined by combining the access influence weight and the fitting error, including: calculating the product of the access influence weight of the access interval and the fitting error of the corresponding correction interval data, normalizing the opposite of the product value to the maximum and minimum values, and obtaining the influence degree of the access interval.

[0073] Since the larger the value of the access influence weight is, the greater the impact is, which is reflected in the abnormal increase or decrease of all access intervals within the corresponding time range. In essence, it means that the access interval is more abnormal within the neighborhood range, and the larger the fitting error is, the less it conforms to the normal fitting characteristics.

[0074] Therefore, we directly calculate the product of the use impact weight of the use interval and the fitting error of the corresponding corrected interval data. The inverse of this product is normalized to its minimum and maximum values ​​to obtain the degree of influence of the use interval. The degree of influence represents the impact of the use interval under normal data fluctuations. In other words, the more normal the use interval characteristics are among all the use intervals, the greater the degree of influence, and the more consistent it is with normal medication use patterns. This facilitates subsequent weighted analysis of normal predictions based on the degree of influence.

[0075] The interval prediction module 103 is used to perform data prediction after the current access by combining all the modified interval data in the time series to determine the predicted access interval for the next access; weight the duration of the predicted access interval according to the influence of the predicted access interval, and determine the target interval for the next access in combination with the access interval of the current access.

[0076] In an embodiment of the present invention, all correction interval data can be combined for prediction analysis. Furthermore, in some embodiments of the present invention, based on the least squares method, least squares analysis is performed on all correction interval data in the time series to obtain predicted correction data for the next use; the sum of the predicted correction data and the mean duration of all use intervals is used as the predicted use interval for the next use.

[0077] The least squares method is a data fitting prediction algorithm well known to those skilled in the art and is not further limited or elaborated on. The predicted usage interval represents the usage interval predicted under normal circumstances, which has a certain degree of credibility, i.e., the degree of influence. Therefore, it can be weighted according to the degree of influence.

[0078] Furthermore, in some embodiments of the present invention, the duration of the predicted access interval is weighted according to the degree of influence of the predicted access interval, and the target interval for the next access is determined in combination with the access interval currently accessed, including: calculating the difference between the unit value 1 and the degree of influence as the current weight; using the degree of influence as the duration weight of the predicted access interval, and the current weight as the duration weight of the access interval currently accessed, and weighting to obtain the target interval for the next access.

[0079] In this weighting process, since the predicted access interval is the time interval for normal data analysis, its credibility is represented by the degree of influence. In the face of uncontrollable changes, this scheme uses the current access interval as the change impact, that is, the current weight is used as the duration weight of the current access interval.

[0080] Specific weighted calculation: take the impact degree as the duration weight of the predicted access interval, and the current weight as the duration weight of the current access interval, and weight them to obtain the target interval for the next access, including: calculating the product of the impact degree and the predicted access interval as the first weighted index; calculating the product of the current weight and the current access interval as the second weighted index; and taking the sum of the first weighted index and the second weighted index as the target interval for the next access.

[0081] That is, by weighting, the predicted usage interval and the current usage interval are combined to predict the next usage interval and obtain the target interval. The target interval is combined with the fluctuation analysis of all usage to obtain the predicted interval value, which has a certain objectivity and reliability.

[0082] The reminder recording module 104 is used to determine the target time for the next access according to the target interval and the current access time, and to remind and record the next access according to the target time.

[0083] In an embodiment of the present invention, a reminder can be given for the next use according to a target interval, and a reminder and record can be given for the next use according to a target time, including: determining a preset reminder advance time, using the preset reminder advance time before the target time as the reminder time, controlling the medicine bottle to send a reminder signal at the reminder time, and detecting whether the contents of the medicine bottle have been used, and recording the corresponding use time when the contents of the medicine bottle are used.

[0084] Among them, the preset reminder advance time is the preset advance time. For example, when the reminder should be given at 12 o'clock, in order to leave redundancy for the specific time to pick up the medicine, the reminder needs to be given a period of time in advance, such as half an hour in advance, or 5 minutes in advance. The preset reminder advance time can be adaptively adjusted, and there is no restriction on this.

[0085] The preset reminder time before the target time is used as the reminder time. At the reminder time, the medicine bottle is controlled to send a reminder signal and detect whether the contents of the medicine bottle are taken. When the contents of the medicine bottle are taken, the corresponding taking time is recorded.

[0086] In an embodiment of the present invention, a weight sensor is installed at the bottom of the medicine bottle, and an opening monitoring device is installed at the bottle mouth to detect whether the contents of the medicine bottle have been taken. Specifically, for any suspected moment, when the weight detected by the weight sensor decreases before and after the suspected moment, and the opening monitoring device detects that the bottle mouth is opened at the suspected moment, it is determined that the contents have been taken; otherwise, it is determined that the contents have not been taken.

[0087] That is, the reminder process is an advance reminder, while the actual access record is a real-time record, which is convenient for subsequent access analysis.

[0088] The embodiment of the present invention determines the withdrawal interval by obtaining the historical withdrawal time of the user to withdraw the contents of the medicine bottle, and then performs an impact analysis based on the change in the length of the withdrawal intervals adjacent in time to determine the impact degree of the withdrawal interval. The impact degree represents the impact of the withdrawal interval under normal data fluctuations, that is, the more normal the withdrawal interval is among all withdrawal interval characteristics, the greater the impact degree value, and the more consistent it is with the normal medication pattern. Then, data prediction is performed in combination with all corrected interval data to determine the predicted withdrawal interval for the next withdrawal. The predicted withdrawal interval represents the withdrawal characteristics under normal circumstances. Therefore, the duration of the predicted withdrawal interval is directly weighted according to the impact degree of the predicted withdrawal interval, and the target interval for the next withdrawal is determined in combination with the withdrawal interval of the current withdrawal. The target interval can effectively combine all withdrawal intervals for analysis, thereby accurately determining the next withdrawal situation. At the same time, it can minimize the withdrawal abnormality. The target time for the next withdrawal is determined based on the target interval and the current withdrawal time, and the next withdrawal is reminded and recorded based on the target time. The present invention can automatically remind users to take medicines. The automatic reminder process is not extracted based on fixed time, but analyzes the medicine taking habits and reduces abnormal medicine taking characteristics, so as to effectively remind users to take medicines and avoid the inconsistency between the fixed time and the actual medicine taking time. At the same time, the reminder duration can be adaptively adjusted according to all medicine taking times, thereby enhancing the medicine bottle reminder effect.

[0089] The present invention also provides an automatic reminder method for a medicine bottle, which is used to implement the steps of an automatic reminder system for medicine bottles. The automatic reminder method for a medicine bottle and the automatic reminder system for a medicine bottle are based on the same concept. The specific implementation process is detailed in the system embodiment and will not be repeated here.

[0090] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. An automatic reminder system for medicine bottles, characterized in that: The system includes: An acquisition module is used to obtain the historical access times of the user accessing the contents of the medicine bottle, and determine the time interval between two adjacent access times as the access interval for the latter of the two adjacent access times; The impact analysis module is used to determine the corrected interval data for each access interval based on the duration of all access intervals; determine the access impact weight of the access interval based on the corrected interval data of any access interval and the preset number of access intervals closest to it in the time series; perform data fitting on all corrected interval data in the time series to determine the fitting error of the corrected interval data corresponding to each access interval; and determine the impact degree of the access interval by combining the access impact weight and the fitting error; An interval prediction module is used to perform data prediction after the current access by combining all the corrected interval data in the time series to determine the predicted access interval for the next access; weight the duration of the predicted access interval according to the impact degree of the predicted access interval, and determine the target interval for the next access in combination with the access interval of the current access; a reminder recording module, configured to determine a target time for next access based on the target interval and the current access time, and to remind and record the next access based on the target time; Determining the corrected interval data for each access interval based on the duration of all access intervals includes: Calculate the average duration of all access intervals, and use the difference between the duration of each access interval and the average duration as the corrected interval data; The determining of the access impact weight of an access interval based on the corrected interval data of a preset number of access intervals that are closest in time sequence to any access interval includes: When the corrected interval data of two adjacent access intervals in time sequence are one positive and one negative, the latter of the two adjacent access intervals is marked as a first mark, otherwise, it is marked as a second mark, wherein the first mark is different from the second mark; The number of all two adjacent access intervals in the preset number of access intervals that are closest to any access interval in time sequence is taken as the total number of influences; For any access interval, calculate the ratio of the number of items marked as the second mark to the total number of impacts to obtain the access impact weight of any access interval; The step of performing data fitting on all the correction interval data in the time series to determine the fitting error of the correction interval data corresponding to each sampling interval includes: A two-dimensional coordinate system is constructed with the time series as the horizontal axis and the value of the correction interval data as the vertical axis to determine the two-dimensional coordinates used each time; Perform curve fitting on all the two-dimensional coordinates taken for each time to determine the fitting curve; The Euclidean distance between the two-dimensional coordinates at each sampling moment and the coordinates at the same moment in the fitting curve is taken as the fitting error of the correction interval data corresponding to each sampling interval; The combining of the access influence weight and the fitting error to determine the influence degree of the access interval includes: Calculate the product of the access impact weight of the access interval and the fitting error of the corresponding correction interval data, and perform maximum and minimum normalization on the inverse of the product value to obtain the influence degree of the access interval; The step of weighting the predicted access interval according to the influence of the predicted access interval and determining the target access interval for the next access in combination with the current access interval includes: Calculate the difference between the unit value 1 and the influence degree as the current weight; Calculating the product of the impact degree and the predicted access interval as a first weighted index; calculating the product of the current weight and the currently used access interval as a second weighted index; The sum of the first weighted index and the second weighted index is used as the target interval for next use.

2. The automatic reminder system for medicine bottles according to claim 1, characterized in that: The step of combining all the corrected interval data in the time series to perform data prediction and determine the predicted access interval for the next access includes: Based on the least squares method, the least squares analysis is performed on all the correction interval data in the time series to obtain the predicted correction data for the next use; The sum of the predicted correction data and the average of the durations of all the access intervals is used as the predicted access interval for the next access.

3. The automatic reminder system for medicine bottles according to claim 1, characterized in that: Remind and record the next time to use the device according to the target time, including: Determine the preset reminder advance time, and use the preset reminder advance time before the target time as the reminder time. At the reminder time, control the medicine bottle to send a reminder signal, and detect whether the contents of the medicine bottle are taken. When the contents of the medicine bottle are taken, record the corresponding taking time.

4. The automatic reminder system for medicine bottles according to claim 3, characterized in that: A weight sensor is installed at the bottom of the medicine bottle and an opening monitoring device is installed at the bottle mouth to detect whether the contents of the medicine bottle have been taken out. Specifically, the following are included: For any suspected moment, if the weight detected by the weight sensor decreases before and after the suspected moment, and the opening monitoring device detects that the bottle mouth is opened at the suspected moment, it is determined that the contents are taken; otherwise, it is determined that the contents are not taken.

5. An automatic reminder method for a medicine bottle, used in an automatic reminder system for a medicine bottle according to any one of claims 1 to 4, characterized in that: Step S1: Obtain the time at which the user has taken the contents of the medicine bottle in history, and determine the time interval between two adjacent times of taking the contents as the taking interval for the latter of the two adjacent times of taking the contents; Step S2: Determine the corrected interval data for each access interval based on the duration of all access intervals; determine the access impact weight of the access interval based on the corrected interval data of any access interval and the preset number of access intervals closest to it in the time series; perform data fitting on all corrected interval data in the time series to determine the fitting error of the corrected interval data corresponding to each access interval; and determine the degree of influence of the access interval by combining the access impact weight and the fitting error; Step S3: After the current access, data prediction is performed in combination with all the modified interval data in the time series to determine the predicted access interval for the next access; the duration of the predicted access interval is weighted according to the impact degree of the predicted access interval, and the target interval for the next access is determined in combination with the current access interval; Step S4: Determine the target time for next access based on the target interval and the current access time, and remind and record the next access based on the target time.

Images