WO2019196283A1 - 流行病预测方法、计算机装置及非易失性可读存储介质 - Google Patents
流行病预测方法、计算机装置及非易失性可读存储介质 Download PDFInfo
- Publication number
- WO2019196283A1 WO2019196283A1 PCT/CN2018/099663 CN2018099663W WO2019196283A1 WO 2019196283 A1 WO2019196283 A1 WO 2019196283A1 CN 2018099663 W CN2018099663 W CN 2018099663W WO 2019196283 A1 WO2019196283 A1 WO 2019196283A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- probability
- popular season
- turning point
- popular
- season
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/80—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for detecting, monitoring or modelling epidemics or pandemics, e.g. flu
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Definitions
- the present application relates to the field of disease prediction technologies, and in particular, to a method for predicting epidemics, a computer device, and a non-volatile readable storage medium.
- epidemiological prediction An important part of the epidemic warning is epidemiological prediction. Most of the existing epidemiological predictions are based on the prediction model to calculate epidemiological thresholds. The epidemiological surveillance data is compared with epidemiological thresholds to classify the menor season and the epidemic non-popular season. The prediction effects of different algorithms are greatly affected by their parameters. The parameters need to be adjusted to get better prediction results. However, the time required for parameter adjustment is longer.
- a first aspect of the present application provides an epidemiological prediction method, the method comprising:
- each popular season/non-popular season turning point is determined, including a rising turning point and a falling turning point, the rising turning point being the starting point of the epidemic epidemic season, and the falling turning point being the ending point of the epidemic epidemic season.
- a second aspect of the present application provides a computer apparatus comprising a memory and a processor, the memory for storing at least one computer readable instruction, the processor for executing at least one computer readable instruction to implement The method of predicting epidemics.
- a third aspect of the present application provides a non-volatile readable storage medium having stored thereon at least one computer readable instruction, the at least one computer readable instruction being executed by a processor
- the epidemiological prediction method is implemented at the time.
- the present application obtains time series data of epidemiological monitoring; calculates a probability that each time point of each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtains a set of probability sequences; determining the probability sequence The probability peak value is obtained, the probability peak sequence is obtained; the probability peaks in the probability peak sequence are screened, and the popular season/non-popular season turning point of the time series data is obtained according to the filtered probability peak; determining each popular season/non The type of popular season turning point, including the rising turning point and the falling turning point, the rising turning point is the starting point of the epidemic epidemic season, and the falling turning point is the ending point of the epidemic epidemic season. Since the probability calculation is not affected by the parameters of the prediction model, the present application avoids a large amount of time to adjust the parameters, and achieves an efficient and rapid epidemiological prediction.
- FIG. 1 is a flowchart of a method for predicting epidemics according to Embodiment 1 of the present application.
- FIG. 2 is a structural diagram of a disease prediction apparatus according to Embodiment 2 of the present application.
- FIG. 3 is a schematic diagram of a computer device according to Embodiment 3 of the present application.
- the epidemiological prediction method of the present application is applied to one or more computer devices.
- the computer device is a device capable of automatically performing numerical calculation and/or information processing according to an instruction set or stored in advance, and the hardware thereof includes but is not limited to a microprocessor and an application specific integrated circuit (ASIC). , Field-Programmable Gate Array (FPGA), Digital Signal Processor (DSP), embedded devices, etc.
- ASIC application specific integrated circuit
- FPGA Field-Programmable Gate Array
- DSP Digital Signal Processor
- embedded devices etc.
- the computer device may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
- the computer device can perform human-computer interaction with the user through a keyboard, a mouse, a remote controller, a touch panel, or a voice control device.
- FIG. 1 is a flowchart of a method for predicting epidemics according to Embodiment 1 of the present application.
- the epidemiological prediction method is applied to a computer device.
- the epidemiological prediction method can calculate the probability that each epidemiological monitoring data corresponding time point belongs to a popular season/non-popular season turning point in time series data of epidemiological monitoring, and determine a popular season/non-popular season turning point according to the probability, and determine The type of turning point for each popular season/non-popular season, which divides the epidemic season and the epidemic non-popular season.
- the epidemiological prediction method specifically includes the following steps:
- Step 101 Obtain time series data of epidemiological monitoring.
- An epidemiological surveillance network composed of a plurality of monitoring points may be established in a preset area (eg, province, city, region), and epidemiological monitoring data is obtained from the monitoring points, and the epidemiological monitoring data constitutes an epidemiological monitoring time.
- Sequence data Medical institutions, schools and child care institutions, pharmacies, etc. can be selected as monitoring points to conduct epidemiological surveillance and data collection for the corresponding target population.
- a place that meets the preset conditions can be selected as the monitoring point.
- the preset condition may include a number of people, a scale, and the like. For example, select a school with a predetermined number of schools and child care institutions as monitoring points. Another example is to select a pharmacy that has reached the preset size (for example, by daily turnover) as a monitoring point. For another example, select a hospital (for example, the number of people who seek medical treatment in Japan) to reach a preset size as a monitoring point.
- Epidemiological surveillance data at different times constitute time series data for epidemiological surveillance.
- epidemiological surveillance data collected on a daily basis can be used to form time series data for epidemiological surveillance.
- epidemiological surveillance data collected on a weekly basis can be used to form time series data for epidemiological surveillance.
- medical institutions, schools, child care institutions, and pharmacies are mainly selected to collect epidemiological surveillance data.
- data sources does not limit the addition or replacement of other focused populations or sites in other embodiments as a source of data for monitoring.
- hotels can be included in the epidemiological surveillance area to obtain epidemiological surveillance data for hotel residents.
- epidemiological surveillance data collected by any type of monitoring point can constitute time series data of epidemiological surveillance.
- epidemiological surveillance data collected by hospitals can be used to form time series data for epidemiological surveillance.
- epidemiological surveillance data collected from multiple types of monitoring points can be combined to form time series data for epidemiological surveillance.
- the epidemiological surveillance data collected by the hospital can be used as a supplement, and the epidemiological monitoring data participated by the pharmacy is supplemented to constitute time series data of epidemiological surveillance.
- Epidemiological surveillance data can include the number of visits to epidemics, the rate of visits, the number of cases, or the incidence.
- the number of daily visits to an epidemic eg, flu
- the number of daily visits to an epidemic can be obtained from a medical institution (eg, a hospital), and the number of daily visits to an epidemic (eg, flu) can be used as epidemiological surveillance data.
- the number of daily epidemics of a student's epidemic eg, flu
- the number of daily epidemics of an epidemic eg, flu
- Step 102 Calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences.
- the probability that the time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point may be sequentially calculated in time series.
- the probability that the time point of each epidemiological monitoring data in the time series data belongs to the popular season/non-popular season turning point may be calculated according to the following method:
- the probability of r t at time t contains three elements:
- the true value x t of the time point t is based on the probability P(x t
- ⁇ 1: t-1 represents the mean of [x 0 , ..., x t-1 ]
- ⁇ 1: t-1 represents the standard deviation of [x 0 , ..., x t-1 ].
- r t-1 ) probability is an artificially adjustable parameter value.
- r t-1 ) is larger, it indicates that the next time point t is the probability of the popular season/non-popular season turning point.
- time point t belongs to the popular season/non-popular season turning point probability P t is:
- the probability that the time point 0 to the time point t belong to the popular season/non-popular season turning point is calculated, and the probability sequence is obtained:
- P [P 0 , P 1 , P 2 , ..., P t ].
- Step 103 Determine a probability peak in the probability sequence to obtain a sequence of probability peaks.
- the probability of each probability in the probability sequence and its adjacent probability may be compared in turn. If one probability in the probability sequence is greater than its adjacent probability, the probability is the probability peak in the probability sequence. For example, P 1 , P 2 , and P 3 are adjacent, P 1 ⁇ P 2 and P 3 ⁇ P 2 , and P 2 is the probability peak in the probability sequence.
- Step 104 Filter the probability peaks in the probability peak sequence, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peaks.
- Predetermined percentiles (or quantiles) of the sequence of probabilistic peaks can be calculated, with the predetermined percentile (or quantile) as a threshold for screening if the probability in the sequence of probabilities is peaked If the peak value is greater than the preset percentile (or quantile), then the probability peak is a popular season/non-popular season turning point. For example, the 70th percentile (or 0.7 quantile) of the probability peak sequence can be calculated if the probability peak in the probability peak sequence is greater than the 70th percentile of the probability peak sequence (or 0.7) The quantile is the peak of the popular season/non-popular season.
- an average of the probability peak sequences may be calculated, and the average is used as a screening threshold. If the probability peak in the probability peak sequence is greater than the average, the probability peak is a popular season/non-popular The turning point of the season.
- a median of the probability peak sequence may be calculated, using the median as a screening threshold, and if the probability peak in the probability peak sequence is greater than the median, the probability peak is a popular season / Non-popular season turning point.
- step 105 the type of each popular season/non-popular season turning point is determined, and the type includes a rising turning point and a falling turning point.
- the rising turning point is the starting point of the epidemic epidemic season (ie, the end point of the epidemic non-popular season), and the falling turning point is the ending point of the epidemic epidemic season (ie, the starting point of the epidemic non-popular season), rising from The turning point to the adjacent falling turning point is the epidemic epidemic season, from the falling point to the adjacent rising turning point is the epidemic non-popular season.
- the popular season/non-popular turning point in the time series data may be calculated to the previous popular season/non-popular season turning point (excluding the popular season/non-popular season) The turning point, including the mean of the previous popular season/non-popular season turning point. If the epidemic monitoring data corresponding to the popular season/non-popular turning point is greater than the mean, the popular season/non-popular turning point is the rising turning point. . Otherwise, if the epidemic monitoring data corresponding to the popular season/non-popular season turning point is small, the turning point of the popular season/non-popular season is a falling turning point.
- the time-series data may be calculated, and the popular season/non-popular season turning point is changed to the previous popular season/non-popular season turning point (excluding the popular season/non-).
- the average turning point of the popular season including the average of the previous popular season/non-popular season turning point, and the turning point of the popular season/non-popular season to the next popular season/non-popular season turning point (excluding the popular season/non-popular season turning point, Includes the mean of the next popular season/non-popular season turning point.
- the popular season / Non-popular season turning point is the rising turning point. Otherwise, if the mean value of the popular season/non-popular season turning point to the previous popular season/non-popular season turning point is greater than the mean value of the popular season/non-popular season turning point to the next popular season/non-popular season turning point, then The turning point of the popular season/non-popular season is the turning point of decline.
- the time-series data may be calculated, and the popular season/non-popular season turning point is changed to the previous popular season/non-popular season turning point (excluding the popular season/non- The median turning point of the popular season, including the previous popular season/non-popular turning point) and the transition period of the popular season/non-popular season to the next popular season/non-popular season turning point (excluding the popular season/non-popular season) The median of the turning point, including the next popular season/non-popular turning point.
- the popular season/non-popular season turning point is a rising turning point. Otherwise, if the median value of the popular season/non-popular season turning point to the previous popular season/non-popular season turning point is greater than the median value of the popular season/non-popular season turning point to the next popular season/non-popular season turning point.
- the turning point of the popular season/non-popular season is the turning point of decline.
- Step 101 Obtain time series data of epidemiological surveillance [2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, 5], wherein 2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, and 5 correspond to time points 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, respectively.
- Step 102 Calculate the corresponding time point of each epidemic monitoring data in the time series data [2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, 5] belongs to the popular season/non-popular season turning point The probability of obtaining a set of probability sequences [0.01, 0.02, 0.01, 0.03, 0.03, 0.06, 0.04, 0.4, 0.02, 0.04, 0.05].
- the above examples have two consecutive rising turning points.
- the first rising turning point indicates the beginning of the epidemic season, and the second rising turning point indicates that the lower risk level is entering the higher risk level. .
- the epidemiological prediction method of the first embodiment acquires time series data of the epidemic monitoring; and calculates a probability that the time point of each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtains a set of probability sequences; Determining a probability peak in the probability sequence to obtain a probability peak sequence; screening a probability peak in the probability peak sequence, and obtaining a popular season/non-popular season turning point of the time series data according to the filtered probability peak; determining The type of each popular season/non-popular season turning point, including the rising turning point and the falling turning point, the rising turning point being the starting point of the epidemic epidemic season, and the falling turning point being the ending point of the epidemic epidemic season. Since the probability calculation is not affected by the prediction model parameters, the first embodiment avoids a large amount of time to adjust the parameters, and achieves an efficient and rapid epidemiological prediction.
- FIG. 2 is a structural diagram of a disease prediction apparatus according to Embodiment 2 of the present application.
- the epidemic prediction apparatus 10 may include an acquisition unit 201, a calculation unit 202, a first determination unit 203, a screening unit 204, and a second determination unit 205.
- the obtaining unit 201 is configured to acquire time series data of epidemiological monitoring.
- An epidemiological monitoring network composed of a plurality of monitoring points may be established in a preset area (for example, a province, a city, a region), and epidemiological monitoring data is obtained from the monitoring points, and the epidemiological monitoring data constitutes an epidemiological monitoring data.
- Time series data Medical institutions, schools and child care institutions, pharmacies, etc. can be selected as monitoring points to conduct epidemiological surveillance and data collection for the corresponding target population.
- a place that meets the preset conditions can be selected as the monitoring point.
- the preset condition may include a number of people, a scale, and the like. For example, select a school with a predetermined number of schools and child care institutions as monitoring points. Another example is to select a pharmacy that has reached the preset size (for example, by daily turnover) as a monitoring point. For another example, select a hospital (for example, the number of people who seek medical treatment in Japan) to reach a preset size as a monitoring point.
- Epidemiological surveillance data at different times constitute time series data for epidemiological surveillance.
- epidemiological surveillance data collected on a daily basis can be used to form time series data for epidemiological surveillance.
- epidemiological surveillance data collected on a weekly basis can be used to form time series data for epidemiological surveillance.
- medical institutions, schools, child care institutions, and pharmacies are mainly selected to collect epidemiological surveillance data.
- data sources does not limit the addition or replacement of other focused populations or sites in other embodiments as a source of data for monitoring.
- hotels can be included in the epidemiological surveillance area to obtain epidemiological surveillance data for hotel residents.
- epidemiological surveillance data collected by any type of monitoring point can constitute time series data of epidemiological surveillance.
- epidemiological surveillance data collected by hospitals can be used to form time series data for epidemiological surveillance.
- epidemiological surveillance data collected from multiple types of monitoring points can be combined to form time series data for epidemiological surveillance.
- the epidemiological surveillance data collected by the hospital can be used as a supplement, and the epidemiological monitoring data participated by the pharmacy is supplemented to constitute time series data of epidemiological surveillance.
- Epidemiological surveillance data can include the number of visits to epidemics, the rate of visits, the number of cases, or the incidence.
- the number of daily visits to an epidemic eg, flu
- the number of daily visits to an epidemic can be obtained from a medical institution (eg, a hospital), and the number of daily visits to an epidemic (eg, flu) can be used as epidemiological surveillance data.
- the number of daily epidemics of a student's epidemic eg, flu
- the number of daily epidemics of an epidemic eg, flu
- the calculating unit 202 is configured to calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences.
- the probability that the time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point may be sequentially calculated in time series.
- the probability that the time point of each epidemiological monitoring data in the time series data belongs to the popular season/non-popular season turning point may be calculated according to the following method:
- the probability of r t at time t contains three elements:
- the true value x t of the time point t is based on the probability P(x t
- ⁇ 1: t-1 represents the mean of [x 0 , ..., x t-1 ]
- ⁇ 1: t-1 represents the standard deviation of [x 0 , ..., x t-1 ].
- r t-1 ) probability is an artificially adjustable parameter value.
- r t-1 ) is larger, it indicates that the next time point t is the probability of the popular season/non-popular season turning point.
- time point t belongs to the popular season/non-popular season turning point probability P t is:
- the probability that the time point 0 to the time point t belong to the popular season/non-popular season turning point is calculated, and the probability sequence is obtained:
- P [P 0 , P 1 , P 2 , ..., P t ].
- the first determining unit 203 is configured to determine a probability peak in the probability sequence to obtain a sequence of probability peaks.
- the probability of each probability in the probability sequence and its adjacent probability may be compared in turn. If one probability in the probability sequence is greater than its adjacent probability, the probability is the probability peak in the probability sequence. For example, P 1 , P 2 , and P 3 are adjacent, P 1 ⁇ P 2 and P 3 ⁇ P 2 , and P 2 is the probability peak in the probability sequence.
- the screening unit 204 is configured to filter a probability peak in the probability peak sequence, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peak.
- Predetermined percentiles (or quantiles) of the sequence of probabilistic peaks can be calculated, with the predetermined percentile (or quantile) as a threshold for screening if the probability in the sequence of probabilities is peaked If the peak value is greater than the preset percentile (or quantile), then the probability peak is a popular season/non-popular season turning point. For example, a 70th percentile (or 0.7 quantile) of the probability peak sequence can be calculated if the probability peak in the probability peak sequence is greater than the 70th percentile of the probability peak sequence ( Or 0.7 quartile), then the probability peak is the popular season/non-popular season turning point.
- an average of the probability peak sequences may be calculated, and the average is used as a screening threshold. If the probability peak in the probability peak sequence is greater than the average, the probability peak is a popular season/non-popular The turning point of the season.
- a median of the probability peak sequence may be calculated, using the median as a screening threshold, and if the probability peak in the probability peak sequence is greater than the median, the probability peak is a popular season / Non-popular season turning point.
- the second determining unit 205 is configured to determine a type of each popular season/non-popular season turning point, and the type includes a rising turning point and a falling turning point.
- the rising turning point is the starting point of the epidemic epidemic season (ie, the end point of the epidemic non-popular season), and the falling turning point is the ending point of the epidemic epidemic season (ie, the starting point of the epidemic non-popular season), rising from The turning point to the adjacent falling turning point is the epidemic epidemic season, from the falling point to the adjacent rising turning point is the epidemic non-popular season.
- the popular season/non-popular turning point in the time series data may be calculated to the previous popular season/non-popular season turning point (excluding the popular season/non-popular season) The turning point, including the mean of the previous popular season/non-popular season turning point. If the epidemic monitoring data corresponding to the popular season/non-popular turning point is greater than the mean, the popular season/non-popular turning point is the rising turning point. . Otherwise, if the epidemic monitoring data corresponding to the popular season/non-popular season turning point is small, the turning point of the popular season/non-popular season is a falling turning point.
- the time-series data may be calculated, and the popular season/non-popular season turning point is changed to the previous popular season/non-popular season turning point (excluding the popular season/non-).
- the average turning point of the popular season including the average of the previous popular season/non-popular season turning point, and the turning point of the popular season/non-popular season to the next popular season/non-popular season turning point (excluding the popular season/non-popular season turning point, Includes the mean of the next popular season/non-popular season turning point.
- the popular season / Non-popular season turning point is the rising turning point. Otherwise, if the mean value of the popular season/non-popular season turning point to the previous popular season/non-popular season turning point is greater than the mean value of the popular season/non-popular season turning point to the next popular season/non-popular season turning point, then The turning point of the popular season/non-popular season is the turning point of decline.
- the time-series data may be calculated, and the popular season/non-popular season turning point is changed to the previous popular season/non-popular season turning point (excluding the popular season/non- The median turning point of the popular season, including the previous popular season/non-popular turning point) and the transition period of the popular season/non-popular season to the next popular season/non-popular season turning point (excluding the popular season/non-popular season) The median of the turning point, including the next popular season/non-popular turning point.
- the popular season/non-popular season turning point is a rising turning point. Otherwise, if the median value of the popular season/non-popular season turning point to the previous popular season/non-popular season turning point is greater than the median value of the popular season/non-popular season turning point to the next popular season/non-popular season turning point.
- the turning point of the popular season/non-popular season is the turning point of decline.
- the epidemiological prediction device 10 is exemplified below:
- the obtaining unit 201 obtains time series data of epidemiological monitoring [2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, 5], wherein 2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, and 5 correspond to time points 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, respectively.
- the calculating unit 202 calculates, according to the time series data [2, 2.5, 2, 3, 2, 4, 3, 6, 7, 6, 5], the time point corresponding to each epidemic monitoring data belongs to the popular season/non-popular season turning point.
- the probability of obtaining a set of probability sequences [0.01, 0.02, 0.01, 0.03, 0.03, 0.06, 0.04, 0.4, 0.02, 0.04, 0.05].
- the above examples have two consecutive rising turning points.
- the first rising turning point indicates the beginning of the epidemic season, and the second rising turning point indicates that the lower risk level is entering the higher risk level. .
- the epidemic prediction device 10 of the second embodiment acquires time series data of epidemiological monitoring; calculates a probability that each time point of each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtains a set of probability sequences. Determining a probability peak in the probability sequence to obtain a probability peak sequence; screening a probability peak in the probability peak sequence, and obtaining a popular season/non-popular season turning point of the time series data according to the filtered probability peak; The type of each popular season/non-popular season turning point is determined, including a rising turning point and a falling turning point, the rising turning point being the starting point of the epidemic epidemic season, and the falling turning point being the ending point of the epidemic epidemic season. Since the probability calculation is not affected by the prediction model parameters, the second embodiment achieves an efficient and rapid epidemiological prediction.
- FIG. 3 is a schematic diagram of a computer apparatus according to Embodiment 3 of the present application.
- the computer device 1 includes a memory 20, a processor 30, and computer readable instructions 40, such as an epidemiological prediction program, stored in the memory 20 and executable on the processor 30.
- computer readable instructions 40 such as an epidemiological prediction program, stored in the memory 20 and executable on the processor 30.
- the processor 30 executes the computer readable instructions 40, the steps in the embodiment of the epidemiological prediction method are implemented, for example, steps 101-105 shown in FIG.
- Step 101 Obtain time series data of epidemiological monitoring
- Step 102 Calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences;
- Step 103 Determine a probability peak in the probability sequence to obtain a sequence of probability peaks
- Step 104 Filter the probability peaks in the sequence of probability peaks, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peaks;
- step 105 the type of each popular season/non-popular season turning point is determined, and the type includes a rising turning point and a falling turning point.
- the processor 30 executes the computer readable instructions 40, the functions of the modules/units in the epidemiological prediction embodiment are implemented, such as the units 201-205 in FIG. :
- the obtaining unit 201 is configured to obtain time series data of epidemiological monitoring
- the calculating unit 202 is configured to calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences;
- a first determining unit 203 configured to determine a probability peak in the probability sequence, to obtain a sequence of probability peaks
- the screening unit 204 is configured to filter a probability peak in the probability peak sequence, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peak;
- the second determining unit 205 is configured to determine a type of each popular season/non-popular season turning point, and the type includes a rising turning point and a falling turning point.
- the computer readable instructions 40 may be partitioned into one or more modules/units that are stored in the memory 20 and executed by the processor 30, To complete this application.
- the one or more modules/units may be a series of computer readable instruction segments capable of performing a particular function for describing the execution of the computer readable instructions 40 in the computer device 1.
- the computer readable instructions 40 may be divided into the obtaining unit 201, the calculating unit 202, the first determining unit 203, the screening unit 204, and the second determining unit 205 in FIG.
- the computer device 1 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. It will be understood by those skilled in the art that the schematic diagram 3 is merely an example of the computer device 1, and does not constitute a limitation of the computer device 1, and may include more or less components than those illustrated, or may combine some components, or different.
- the components, such as the computer device 1, may also include input and output devices, network access devices, buses, and the like.
- the processor 30 may be a central processing unit (CPU), or may be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
- the general purpose processor may be a microprocessor or the processor 30 may be any conventional processor or the like, and the processor 30 is a control center of the computer device 1, and connects the entire computer device 1 by using various interfaces and lines. Various parts.
- the memory 20 can be used to store the computer readable instructions 40 and/or modules/units by running or executing computer readable instructions 40 and/or modules/units stored in the memory 20.
- the various functions of the computer device 1 are implemented by invoking data stored in the memory 20.
- the memory 20 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be Data (such as audio data, phone book, etc.) created according to the use of the computer device 1 is stored.
- the memory 20 may include a high-speed random access memory, and may also include a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a secure digital (Secure Digital, SD).
- a non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a secure digital (Secure Digital, SD).
- SMC smart memory card
- SD Secure Digital
- Card flash card, at least one disk storage device, flash device, or other volatile solid state storage device.
- the modules/units integrated by the computer device 1 can be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, the present application implements all or part of the processes in the foregoing embodiments, and may also be implemented by computer-readable instructions, which may be stored in a non-volatile manner. In reading a storage medium, the computer readable instructions, when executed by a processor, implement the steps of the various method embodiments described above. Wherein, the computer readable instructions comprise computer readable instruction code, which may be in the form of source code, an object code form, an executable file or some intermediate form or the like.
- the computer readable medium can include any entity or device capable of carrying the computer readable instruction code, a recording medium, a USB flash drive, a removable hard drive, a magnetic disk, an optical disk, a computer memory, a read only memory (ROM, Read-Only) Memory), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media.
- a recording medium a USB flash drive, a removable hard drive, a magnetic disk, an optical disk, a computer memory, a read only memory (ROM, Read-Only) Memory), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media.
- ROM Read Only memory
- RAM random access memory
- the embodiment provides a non-volatile readable storage medium, where the non-readable storage medium stores computer readable instructions, and the computer readable instructions are implemented by the processor to implement the disease abnormal data detecting method.
- Step 101 Obtain time series data of epidemiological monitoring
- Step 102 Calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences;
- Step 103 Determine a probability peak in the probability sequence to obtain a sequence of probability peaks
- Step 104 Filter the probability peaks in the sequence of probability peaks, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peaks;
- step 105 the type of each popular season/non-popular season turning point is determined, and the type includes a rising turning point and a falling turning point.
- the computer readable instructions when executed by the processor, implement the functions of the various modules/units in the apparatus embodiments described above, such as units 201-205 in FIG. 2:
- the obtaining unit 201 is configured to obtain time series data of epidemiological monitoring
- the calculating unit 202 is configured to calculate a probability that a time point corresponding to each epidemic monitoring data in the time series data belongs to a popular season/non-popular season turning point, and obtain a set of probability sequences;
- a first determining unit 203 configured to determine a probability peak in the probability sequence, to obtain a sequence of probability peaks
- the screening unit 204 is configured to filter a probability peak in the probability peak sequence, and obtain a popular season/non-popular season turning point of the time series data according to the filtered probability peak;
- the second determining unit 205 is configured to determine a type of each popular season/non-popular season turning point, and the type includes a rising turning point and a falling turning point.
- each functional unit in each embodiment of the present application may be integrated in the same processing unit, or each unit may exist physically separately, or two or more units may be integrated in the same unit.
- the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software function modules.
Landscapes
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
一种流行病预测方法,所述方法获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。本申请还提供一种计算机装置及可读存储介质。本申请可以实现高效快速的流行病预测。
Description
本申请要求于2018年04月11日提交中国专利局,申请号为201810322421.4发明名称为“流行病预测方法、计算机装置及计算机可读存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及疾病预测技术领域,具体涉及一种流行病预测方法、计算机装置和非易失性可读存储介质。
随着社会和自然环境的变化,流行病的病原体、传播途径、发病特点以及影响因素也发生很大变化,如何能早期识别到流行病突发公共卫生事件,及时发出预警,尽早采取相应的控制措施,将突发公共卫生事件造成的损失降到最低,是公共卫生领域长期以来关注的焦点,也是卫生应急工作的重要内容。突发公共卫生事件预警,是通过对有关数据的收集,整理、分析和整合,运用计算机、网络、通讯等现代先进的技术,对事件的征兆进行监测、识别、诊断与评价,及时报警,告知有关部门和公众做好相关的应对和准备工作,及时采取有效的防控措施,尽可能阻止或减缓突发事件的发生或减少事件的危害。
流行病预警中的一个重要环节是流行病预测。现有的流行病预测大多是根据预测模型计算流行病阈值,将流行病监测数据与流行病阈值进行比较来划分流形季与流行病非流行季,不同算法的预测效果受其参数影响较大,需要对参数进行调整,以获得较佳的预测效果。然而,参数调整所需要的时间较长。
发明内容
鉴于以上内容,有必要提出一种流行病预测方法及装置、计算机装置和非易失性可读存储介质,其可以实现高效快速的流行病预测。
本申请的第一方面提供一种流行病预测方法,所述方法包括:
获取流行病监测的时间序列数据;
计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;
确定所述概率序列中的概率峰值,得到概率峰值序列;
对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得 到所述时间序列数据的流行季/非流行季转折点;
确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。
本申请的第二方面提供一种计算机装置,所述计算机装置包括存储器和处理器,所述存储器用于存储至少一个计算机可读指令,所述处理器用于执行至少一个计算机可读指令以实现所述流行病预测方法。
本申请的第三方面提供一种非易失性可读存储介质,所述非易失性可读存储介质上存储有至少一个计算机可读指令,所述至少一个计算机可读指令被处理器执行时实现所述流行病预测方法。
本申请获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。由于概率计算不受预测模型参数的影响,因此,本申请避免了调整参数要花费大量时间,实现了高效快速的流行病预测。
图1是本申请实施例一提供的流行病预测方法的流程图。
图2是本申请实施例二提供的流行病预测装置的结构图。
图3是本申请实施例三提供的计算机装置的示意图。
为了能够更清楚地理解本申请的上述目的、特征和优点,下面结合附图和具体实施例对本申请进行详细描述。需要说明的是,在不冲突的情况下,本申请的实施例及实施例中的特征可以相互组合。
在下面的描述中阐述了很多具体细节以便于充分理解本申请,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
除非另有定义,本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同。本文中在本申请的说明书中所使用的术语只是为了描述具体的实施例的目的,不是旨在于限制本申请。
优选地,本申请的流行病预测方法应用在一个或者多个计算机装置中。 所述计算机装置是一种能够按照事先设定或存储的指令,自动进行数值计算和/或信息处理的设备,其硬件包括但不限于微处理器、专用集成电路(Application Specific Integrated Circuit,ASIC)、可编程门阵列(Field-Programmable Gate Array,FPGA)、数字处理器(Digital Signal Processor,DSP)、嵌入式设备等。
所述计算机装置可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述计算机装置可以与用户通过键盘、鼠标、遥控器、触摸板或声控设备等方式进行人机交互。
实施例一
图1是本申请实施例一提供的流行病预测方法的流程图。所述流行病预测方法应用于计算机装置。所述流行病预测方法可以计算流行病监测的时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,根据该概率确定流行季/非流行季转折点,并确定每个流行季/非流行季转折点的类型,从而划分出流行病流行季与流行病非流行季。
如图1所示,所述流行病预测方法具体包括以下步骤:
步骤101,获取流行病监测的时间序列数据。
可以在预设区域(例如省、市、地区)建立由多个监测点组成的流行病监测网络,从所述监测点获取流行病监测数据,由所述流行病监测数据构成流行病监测的时间序列数据。可以选择医疗机构、学校和幼托机构、药店等作为监测点,分别对相应的目标人群进行流行病监测及数据采集。可以选择满足预设条件的场所作为监测点。所述预设条件可以包括人数、规模等。例如,选择学生人数达到预设数量的学校和幼托机构作为监控点。又如,选择规模(例如以日营业额统计)达到预设规模的药店作为监控点。再如,选择规模(例如以日就医人数统计)达到预设规模的医院作为监控点。
不同时间的流行病监测数据构成流行病监测的时间序列数据。例如,可以将以日为单位采集到的流行病监测数据构成流行病监测的时间序列数据。或者,可以将以周为单位采集到的流行病监测数据构成流行病监测的时间序列数据。
医疗机构(主要包括医院)是最能捕捉流行病早期暴发预兆的场所,是开展流行病监测的首选。可以根据病人就诊情况,获取流行病监测数据。
一部分流行病人会自行去药店购药来缓解早期症状,因此,可以根据药店的药品销售情况,获取流行病监测数据。
儿童和青少年是流行病的高危人群以及流行病传播过程中的重要环节,也应该加强对该人群的监测。学校和幼托机构是监测儿童和青少年流行病发病情况的较佳场所。可以根据学校和幼托机构的儿童和青少年的请假情况,获得流行病监测数据。
因此,本申请中主要选择医疗机构、学校和幼托机构、药店这三类场所进行流行病监测数据的采集。当然,上述对数据源的选择,并不能限制在另外的实施方案中增加或替换其他重点关注人群或场所作为监测的数据源。例如,可以将宾馆纳入流行病监测范围,获取宾馆入住人员的流行病监测数据。
根据需要,可以取任意一类监控点(例如医疗机构)采集的流行病监测数据构成流行病监测的时间序列数据。例如,可以取医院采集的流行病监测数据构成流行病监测的时间序列数据。或者,可以结合多类监控点采集的流行病监测数据构成流行病监测的时间序列数据。例如,可以以医院采集的流行病监测数据为主,以药店参加的流行病监测数据作为补充,构成流行病监测的时间序列数据。
流行病监测数据可以包括流行病的就诊数、就诊率、发病数或发病率等。例如,可以从医疗机构(例如医院)获取流行病(例如流感)的每日就诊数,将流行病(例如流感)的每日就诊数作为流行病监测数据。又如,可以从学校获取学生的流行病(例如流感)的每日发病数,将流行病(例如流感)的每日发病数作为流行病监测数据。
步骤102,计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列。
可以按照时间顺序依次计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率。
可以按照以下方法计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率:
流行病监测的时间序列数据为X=[x
0,x
1,x
2,…,x
t],时间点0为已知的上一个流行季/非流行季转折点,时间点t为给定时间点。
以r
t表示时间点t到上一个流行季/非流行季转折点的间距。若时间点t是流行季/非流行季转折点,则r
t=0。例如,时间点t=3是流行季/非流行季转折点,则r
t=3=0,且r
t=5=2,表示时间点t=5和上一个流行季/非流行季转折点t=3的间距为2。
因此,对于时间点t来说,若其下一时间点r
t+1不是流行季/非流行季转折点,则r
t+1=r
t+1;若其下一时间点r
t+1是流行季/非流行季转折点,则r
t+1=0。
初始化时将时间序列数据中第一个时间点属于流行季/非流行季转折点的概率设置为1,即
P
t(r
t=0)=1。
对于时间点t下r
t取值的概率含有3个要素:
(1)时间点t的真实值x
t基于[x
0…x
t-1]序列分布的概率P(x
t|x
1:t-1),
其中μ
1:t-1表示[x
0,…,x
t-1]的均值,σ
1:t-1表示[x
0,…,x
t-1]的标准差。
(2)上一时间点r
t-1的概率P(r
t-1)。
r
t都是由上一时间点r
t-1变化而来,r
t=r
t-1+1或r
t=0。
(3)r
t-1转变成r
t的概率P(r
t|r
t-1)。
P(r
t|r
t-1)概率是一个人为可调的参数值。当P(r
t|r
t-1)越大,则表明下一时间点t是流行季/非流行季转折点的概率越大。
则时间点t属于流行季/非流行季转折点概率P
t为:
P
t(r
t)=P
t-1(r
t-1)*P
t(r
t|r
t-1)*P(x
t|x
1:t-1)。
按照上述方法计算出时间点0到时间点t属于流行季/非流行季转折点的概率,得到概率序列:
P=[P
0,P
1,P
2,…,P
t]。
步骤103,确定所述概率序列中的概率峰值,得到概率峰值序列。
可以依次比较概率序列中每个概率与其前后相邻概率的大小,若概率序列中的一个概率同时大于其前后相邻概率,则该概率为所述概率序列中的概率峰值。例如,P
1、P
2、P
3相邻,P
1<P
2且P
3<P
2,则P
2就是概率序列中的概率峰值。
所述概率序列中的所有概率峰值构成所述概率峰值序列。
步骤104,对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点。
可以计算所述概率峰值序列的预设百分位数(或分位数),以所述预设百分位数(或分位数)作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述预设百分位数(或分位数),则所述概率峰值为流行季/非流行季转折点。例如,可以计算所述概率峰值序列的第70百分位数(或0.7分位数),若所述概率峰值序列中的概率峰值大于所述概率峰值序列的第70百分位数(或0.7分位数),则所述概率峰值为流行季/非流行季转折点。
或者,可以计算所述概率峰值序列的平均数,以所述平均数作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述平均数,则所述概率峰值为流行季/非流行季转折点。
或者,可以计算所述概率峰值序列的中位数,以所述中位数作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述中位数,则所述概率峰值为流行季/非流行季转折点。
步骤105,确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
所述上升转折点为流行病流行季的起始点(即流行病非流行季的结束点),所述下降转折点为流行病流行季的终止点(即流行病非流行季的起始点),从上升转折点至相邻下降转折点为流行病流行季,从下降折点至相邻上升转折点为流行病非流行季。
对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的均值,若所述流行季/非流行季转折点对应的流行病监测数据大于所述均值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点对应的流行病监测数据小所述均值,则所述流行季/非流行季转折点为下降转折点。
或者,对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括后一个流行季/非流行季转折点)的均值。若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值小于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值大于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,则所述流行季/非流行季转折点为下降转折点。
或者,对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的中位值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括后一个流行季/非流行季转折点)的中位值。若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的中位值小于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的中位值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的中位值大于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的中位值,则所述流行季/非流行季转折点为下降转折点。
下面对所述流行病预测方法进行举例说明:
步骤101,获取流行病监测的时间序列数据[2,2.5,2,3,2,4,3,6,7,6,5],其中 2,2.5,2,3,2,4,3,6,7,6,5分别对应时间点0,1,2,3,4,5,6,7,8,9,10。
步骤102,计算所述时间序列数据[2,2.5,2,3,2,4,3,6,7,6,5]中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列[0.01,0.02,0.01,0.03,0.03,0.06,0.04,0.4,0.02,0.04,0.05]。
步骤103,确定概率序列[0.01,0.02,0.01,0.03,0.03,0.06,0.04,0.4,0.02,0.04,0.05]中的概率峰值为时间点t=1的0.02、时间点t=5的0.06,和时间点t=7的0.4,得到概率峰值序列[0.02,0.06,0.4]。
步骤104,对所述概率峰值序列[0.02,0.06,0.4]中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点。例如,取第0.2分位数0.026作为阈值,删除所有小于0.026的概率峰值,即将时间点t=1的0.02删除,剩下时间点t=5的0.06和时间点t=7的0.4,则时间点t=5和t=7为流行季/非流行季转折点。
步骤105,确定时间点t=5和t=7为流行季/非流行季转折点的类型。时间点t=5前的流行病监测数据为[2,2.5,2,3,2],均值为2.3,由于时间点t=5为4大于2.3,所以时间点t=5为上升转折点。时间点t=7到时间点t=5之间的流行病监测数据为[4,3],均值为3.5,时间点t=7为6大于3.5,所以时间点t=7为上升转折点。
需要说明的是,上述举例出现了两个连续的上升转折点,第一个上升转折点表示流行病流行季的起始,第二个上升转折点表示从较低风险等级流行季进入较高风险等级流行季。
实施例一的流行病预测方法获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。由于概率计算不受预测模型参数的影响,因此,实施例一避免了调整参数要花费大量时间,实现了高效快速的流行病预测。
实施例二
图2为本申请实施例二提供的流行病预测装置的结构图。
如图2所示,所述流行病预测装置10可以包括:获取单元201、计算单元202、第一确定单元203、筛选单元204、第二确定单元205。
获取单元201,用于获取流行病监测的时间序列数据。
可以在预设区域(例如省、市、地区)建立由多个监测点组成的流行病 监测网络,从所述监测点获取流行病监测数据,由所述流行病监测数据构成构成流行病监测的时间序列数据。可以选择医疗机构、学校和幼托机构、药店等作为监测点,分别对相应的目标人群进行流行病监测及数据采集。可以选择满足预设条件的场所作为监测点。所述预设条件可以包括人数、规模等。例如,选择学生人数达到预设数量的学校和幼托机构作为监控点。又如,选择规模(例如以日营业额统计)达到预设规模的药店作为监控点。再如,选择规模(例如以日就医人数统计)达到预设规模的医院作为监控点。
不同时间的流行病监测数据构成流行病监测的时间序列数据。例如,可以将以日为单位采集到的流行病监测数据构成流行病监测的时间序列数据。或者,可以将以周为单位采集到的流行病监测数据构成流行病监测的时间序列数据。
医疗机构(主要包括医院)是最能捕捉流行病早期暴发预兆的场所,是开展流行病监测的首选。可以根据病人就诊情况,获取流行病监测数据。
一部分流行病人会自行去药店购药来缓解早期症状,因此,可以根据药店的药品销售情况,获取流行病监测数据。
儿童和青少年是流行病的高危人群以及流行病传播过程中的重要环节,也应该加强对该人群的监测。学校和幼托机构是监测儿童和青少年流行病发病情况的较佳场所。可以根据学校和幼托机构的儿童和青少年的请假情况,获得流行病监测数据。
因此,本申请中主要选择医疗机构、学校和幼托机构、药店这三类场所进行流行病监测数据的采集。当然,上述对数据源的选择,并不能限制在另外的实施方案中增加或替换其他重点关注人群或场所作为监测的数据源。例如,可以将宾馆纳入流行病监测范围,获取宾馆入住人员的流行病监测数据。
根据需要,可以取任意一类监控点(例如医疗机构)采集的流行病监测数据构成流行病监测的时间序列数据。例如,可以取医院采集的流行病监测数据构成流行病监测的时间序列数据。或者,可以结合多类监控点采集的流行病监测数据构成流行病监测的时间序列数据。例如,可以以医院采集的流行病监测数据为主,以药店参加的流行病监测数据作为补充,构成流行病监测的时间序列数据。
流行病监测数据可以包括流行病的就诊数、就诊率、发病数或发病率等。例如,可以从医疗机构(例如医院)获取流行病(例如流感)的每日就诊数,将流行病(例如流感)的每日就诊数作为流行病监测数据。又如,可以从学校获取学生的流行病(例如流感)的每日发病数,将流行病(例如流感)的每日发病数作为流行病监测数据。
计算单元202,用于计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列。
可以按照时间顺序依次计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率。
可以按照以下方法计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率:
流行病监测的时间序列数据为X=[x
0,x
1,x
2,…,x
t],时间点0为已知的上一个流行季/非流行季转折点,时间点t为给定时间点。
以r
t表示时间点t到上一个流行季/非流行季转折点的间距。若时间点t是流行季/非流行季转折点,则r
t=0。例如,时间点t=3是流行季/非流行季转折点,则r
t=3=0,且r
t=5=2,表示时间点t=5和上一个流行季/非流行季转折点t=3的间距为2。
因此,对于时间点t来说,若其下一时间点r
t+1不是流行季/非流行季转折点,则r
t+1=r
t+1;若其下一时间点r
t+1是流行季/非流行季转折点,则r
t+1=0。
初始化时将时间序列数据中第一个时间点属于流行季/非流行季转折点的概率设置为1,即
P
t(r
t=0)=1。
对于时间点t下r
t取值的概率含有3个要素:
其中μ
1:t-1表示[x
0,…,x
t-1]的均值,σ
1:t-1表示[x
0,…,x
t-1]的标准差。
(2)上一时间点r
t-1的概率P(r
t-1)。
r
t都是由上一时间点r
t-1变化而来,r
t=r
t-1+1或r
t=0。
(3)r
t-1转变成r
t的概率P(r
t|r
t-1)。
P(r
t|r
t-1)概率是一个人为可调的参数值。当P(r
t|r
t-1)越大,则表明下一时间点t是流行季/非流行季转折点的概率越大。
则时间点t属于流行季/非流行季转折点概率P
t为:
P
t(r
t)=P
t-1(r
t-1)*P
t(r
t|r
t-1)*P(x
t|x
1:t-1)。
按照上述方法计算出时间点0到时间点t属于流行季/非流行季转折点的概率,得到概率序列:
P=[P
0,P
1,P
2,…,P
t]。
第一确定单元203,用于确定所述概率序列中的概率峰值,得到概率峰 值序列。
可以依次比较概率序列中每个概率与其前后相邻概率的大小,若概率序列中的一个概率同时大于其前后相邻概率,则该概率为所述概率序列中的概率峰值。例如,P
1、P
2、P
3相邻,P
1<P
2且P
3<P
2,则P
2就是概率序列中的概率峰值。
所述概率序列中的所有概率峰值构成所述概率峰值序列。
筛选单元204,用于对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点。
可以计算所述概率峰值序列的预设百分位数(或分位数),以所述预设百分位数(或分位数)作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述预设百分位数(或分位数),则所述概率峰值为流行季/非流行季转折点。例如,可以计算所述概率峰值序列的第70百分位数(或0.7分位数),若所述所述概率峰值序列中的概率峰值大于所述概率峰值序列的第70百分位数(或0.7分位数),则所述概率峰值为流行季/非流行季转折点。
或者,可以计算所述概率峰值序列的平均数,以所述平均数作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述平均数,则所述概率峰值为流行季/非流行季转折点。
或者,可以计算所述概率峰值序列的中位数,以所述中位数作为筛选的阈值,若所述概率峰值序列中的概率峰值大于所述中位数,则所述概率峰值为流行季/非流行季转折点。
第二确定单元205,用于确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
所述上升转折点为流行病流行季的起始点(即流行病非流行季的结束点),所述下降转折点为流行病流行季的终止点(即流行病非流行季的起始点),从上升转折点至相邻下降转折点为流行病流行季,从下降折点至相邻上升转折点为流行病非流行季。
对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的均值,若所述流行季/非流行季转折点对应的流行病监测数据大于所述均值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点对应的流行病监测数据小所述均值,则所述流行季/非流行季转折点为下降转折点。
或者,对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点(不包括所述流 行季/非流行季转折点,包括后一个流行季/非流行季转折点)的均值。若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值小于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值大于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,则所述流行季/非流行季转折点为下降转折点。
或者,对于每个流行季/非流行季转折点,可以计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括前一个流行季/非流行季转折点)的中位值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点(不包括所述流行季/非流行季转折点,包括后一个流行季/非流行季转折点)的中位值。若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的中位值小于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的中位值,则所述流行季/非流行季转折点为上升转折点。否则,若所述流行季/非流行季转折点到前一个流行季/非流行季转折点的中位值大于所述流行季/非流行季转折点到后一个流行季/非流行季转折点的中位值,则所述流行季/非流行季转折点为下降转折点。
下面对所述流行病预测装置10进行举例说明:
获取单元201获取流行病监测的时间序列数据[2,2.5,2,3,2,4,3,6,7,6,5],其中2,2.5,2,3,2,4,3,6,7,6,5分别对应时间点0,1,2,3,4,5,6,7,8,9,10。
计算单元202计算所述时间序列数据[2,2.5,2,3,2,4,3,6,7,6,5]中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列[0.01,0.02,0.01,0.03,0.03,0.06,0.04,0.4,0.02,0.04,0.05]。
第一确定单元203确定概率序列[0.01,0.02,0.01,0.03,0.03,0.06,0.04,0.4,0.02,0.04,0.05]中的概率峰值为时间点t=1的0.02、时间点t=5的0.06,和时间点t=7的0.4,得到概率峰值序列[0.02,0.06,0.4]。
筛选单元204对所述概率峰值序列[0.02,0.06,0.4]中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点。例如,取第0.2分位数0.026作为阈值,删除所有小于0.026的概率峰值,即将时间点t=1的0.02删除,剩下时间点t=5的0.06和时间点t=7的0.4,则时间点t=5和t=7为流行季/非流行季转折点。
第二确定单元205确定时间点t=5和t=7为流行季/非流行季转折点的类型。时间点t=5前的流行病监测数据为[2,2.5,2,3,2],均值为2.3,由于时间点t=5为4大于2.3,所以时间点t=5为上升转折点。时间点t=7到时间点t=5 之间的流行病监测数据为[4,3],均值为3.5,时间点t=7为6大于3.5,所以时间点t=7为上升转折点。
需要说明的是,上述举例出现了两个连续的上升转折点,第一个上升转折点表示流行病流行季的起始,第二个上升转折点表示从较低风险等级流行季进入较高风险等级流行季。
实施例二的流行病预测装置10获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。由于概率计算不受预测模型参数的影响,因此,实施例二实现了高效快速的流行病预测。
实施例三
图3为本申请实施例三提供的计算机装置的示意图。所述计算机装置1包括存储器20、处理器30以及存储在所述存储器20中并可在所述处理器30上运行的计算机可读指令40,例如流行病预测程序。所述处理器30执行所述计算机可读指令40时实现上述流行病预测方法实施例中的步骤,例如图1所示的步骤101~105:
步骤101,获取流行病监测的时间序列数据;
步骤102,计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;
步骤103,确定所述概率序列中的概率峰值,得到概率峰值序列;
步骤104,对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;
步骤105,确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
或者,所述处理器30执行所述计算机可读指令40时实现上述流行病预测实施例中各模块/单元的功能,例如图2中的单元201~205。:
获取单元201,用于获取流行病监测的时间序列数据;
计算单元202,用于计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;
第一确定单元203,用于确定所述概率序列中的概率峰值,得到概率峰值序列;
筛选单元204,用于对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;
第二确定单元205,用于确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
示例性的,所述计算机可读指令40可以被分割成一个或多个模块/单元,所述一个或者多个模块/单元被存储在所述存储器20中,并由所述处理器30执行,以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机可读指令段,该指令段用于描述所述计算机可读指令40在所述计算机装置1中的执行过程。例如,所述计算机可读指令40可以被分割成图2中的获取单元201、计算单元202、第一确定单元203、筛选单元204、第二确定单元205,各单元具体功能参见实施例二。
所述计算机装置1可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。本领域技术人员可以理解,所述示意图3仅仅是计算机装置1的示例,并不构成对计算机装置1的限定,可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件,例如所述计算机装置1还可以包括输入输出设备、网络接入设备、总线等。
所称处理器30可以是中央处理单元(Central Processing Unit,CPU),还可以是其他通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field-Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器30也可以是任何常规的处理器等,所述处理器30是所述计算机装置1的控制中心,利用各种接口和线路连接整个计算机装置1的各个部分。
所述存储器20可用于存储所述计算机可读指令40和/或模块/单元,所述处理器30通过运行或执行存储在所述存储器20内的计算机可读指令40和/或模块/单元,以及调用存储在存储器20内的数据,实现所述计算机装置1的各种功能。所述存储器20可主要包括存储程序区和存储数据区,其中,存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等;存储数据区可存储根据计算机装置1的使用所创建的数据(比如音频数据、电话本等)等。此外,存储器20可以包括高速随机存取存储器,还可以包括非易失性存储器,例如硬盘、内存、插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
所述计算机装置1集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请实现上述实施例方法中的全部或部分流程,也可以通过计算机可读指令来指令相关的硬件来完成,所述的计算机可读指令可存 储于一非易失性可读存储介质中,该计算机可读指令在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机可读指令包括计算机可读指令代码,所述计算机可读指令代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机可读指令代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,非易失性可读介质不包括电载波信号和电信信号。
实施例四
本实施例提供一种非易失性可读存储介质,该非易失性可读存储介质上存储有计算机可读指令,该计算机可读指令被处理器执行时实现上述疾病异常数据检测方法实施例中的步骤,例如图1所示的步骤101~105:
步骤101,获取流行病监测的时间序列数据;
步骤102,计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;
步骤103,确定所述概率序列中的概率峰值,得到概率峰值序列;
步骤104,对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;
步骤105,确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
或者,该计算机可读指令被处理器执行时实现上述装置实施例中各模块/单元的功能,例如图2中的单元201~205:
获取单元201,用于获取流行病监测的时间序列数据;
计算单元202,用于计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;
第一确定单元203,用于确定所述概率序列中的概率峰值,得到概率峰值序列;
筛选单元204,用于对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;
第二确定单元205,用于确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点。
在本申请所提供的几个实施例中,应该理解到,所揭露的计算机装置和方法,可以通过其它的方式实现。例如,以上所描述的计算机装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实 现时可以有另外的划分方式。
另外,在本申请各个实施例中的各功能单元可以集成在相同处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在相同单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能模块的形式实现。
对于本领域技术人员而言,显然本申请不限于上述示范性实施例的细节,而且在不背离本申请的精神或基本特征的情况下,能够以其他的具体形式实现本申请。因此,无论从哪一点来看,均应将实施例看作是示范性的,而且是非限制性的,本申请的范围由所附权利要求而不是上述说明限定,因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本申请内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。此外,显然“包括”一词不排除其他单元或步骤,单数不排除复数。计算机装置权利要求中陈述的多个单元或计算机装置也可以由同一个单元或计算机装置通过软件或者硬件来实现。第一,第二等词语用来表示名称,而并不表示任何特定的顺序。
最后应说明的是,以上实施例仅用以说明本申请的技术方案而非限制,尽管参照较佳实施例对本申请进行了详细说明,本领域的普通技术人员应当理解,可以对本申请的技术方案进行修改或等同替换,而不脱离本申请技术方案的精神和范围。
Claims (20)
- 一种流行病预测方法,其特征在于,所述方法包括:获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。
- 如权利要求1所述的方法,其特征在于,所述时间序列数据为X=[x 0,x 1,x 2,…,x t],时间点0为已知的上一个流行季/非流行季转折点,时间点t为给定时间点,r t表示时间点t到上一个流行季/非流行季转折点的间距,所述计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率包括:步骤a:计算时间点t下r t取值的概率:(1)计算时间点t的真实值x t基于[x 0…x t-1]序列分布的概率P(x t|x 1:t-1),其中μ 1:t-1表示[x 0,…,x t-1]的均值,σ 1:t-1表示[x 0,…,x t-1]的标准差;(2)计算上一时间点r t-1的概率P(r t-1),r t都是由上一时间点r t-1变化而来,r t=r t-1+1或r t=0;(3)计算r t-1转变成r t的概率P(r t|r t-1),P(r t|r t-1)为可调的参数值,P(r t|r t-1)越大,表明下一时间点t是流行季/非流行季转折点的概率越大;步骤b:计算时间点t属于流行季/非流行季转折点概率P t:P t(r t)=P t-1(r t-1)*P t(r t|r t-1)*P(x t|x 1:t-1)。
- 如权利要求1所述的方法,其特征在于,所述对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点包括:计算所述概率峰值序列的预设百分位数,以所述预设百分位数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述预设百分位数,则所述概率峰值为流行季/非流行季转折点;或者计算所述概率峰值序列的平均数,以所述平均数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述平均数,则所述概率峰值为流行季/非流行季转折点;或者计算所述概率峰值序列的中位数,以所述中位数作为筛选的阈值,若所 述所述概率峰值序列中的概率峰值大于所述中位数,则所述概率峰值为流行季/非流行季转折点。
- 如权利要求1所述的方法,其特征在于,所述流行病监测数据包括流行病的就诊数、就诊率、发病数或发病率。
- 如权利要求1所述的方法,其特征在于,所述确定每个流行季/非流行季转折点的类型包括:计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型;或者计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型。
- 如权利要求1所述的方法,其特征在于,所述获取流行病监测的时间序列数据包括:在预设区域建立由多个监测点组成的流行病监测网络,从所述监测点获取流行病监测数据,由所述流行病监测数据构成所述时间序列数据。
- 如权利要求6所述的方法,其特征在于,所述监测点包括满足预设人数或规模的医疗机构、学校和幼托机构、药店。
- 如权利要求1所述的方法,其特征在于,所述确定所述概率序列中的概率峰值包括:依次比较概率序列中每个概率与其前后相邻概率的大小,若该概率大于其前后相邻概率,则该概率为所述概率序列中的概率峰值。
- 一种计算机装置,其特征在于,所述计算机装置包括存储器和处理器,所述存储器用于存储至少一个计算机可读指令,所述处理器用于执行至少一个计算机可读指令以实现以下步骤:获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。
- 如权利要求9所述的计算机装置,其特征在于,所述时间序列数据为X=[x 0,x 1,x 2,…,x t],时间点0为已知的上一个流行季/非流行季转折点,时间点t为给定时间点,r t表示时间点t到上一个流行季/非流行季转折点的间距,所述计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季 /非流行季转折点的概率包括:步骤a:计算时间点t下r t取值的概率:(1)计算时间点t的真实值x t基于[x 0…x t-1]序列分布的概率P(x t|x 1:t-1),其中μ 1:t-1表示[x 0,…,x t-1]的均值,σ 1:t-1表示[x 0,…,x t-1]的标准差;(2)计算上一时间点r t-1的概率P(r t-1),r t都是由上一时间点r t-1变化而来,r t=r t-1+1或r t=0;(3)计算r t-1转变成r t的概率P(r t|r t-1),P(r t|r t-1)为可调的参数值,P(r t|r t-1)越大,表明下一时间点t是流行季/非流行季转折点的概率越大;步骤b:计算时间点t属于流行季/非流行季转折点概率P t:P t(r t)=P t-1(r t-1)*P t(r t|r t-1)*P(x t|x 1:t-1)。
- 如权利要求9所述的计算机装置,其特征在于,所述对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点包括:计算所述概率峰值序列的预设百分位数,以所述预设百分位数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述预设百分位数,则所述概率峰值为流行季/非流行季转折点;或者计算所述概率峰值序列的平均数,以所述平均数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述平均数,则所述概率峰值为流行季/非流行季转折点;或者计算所述概率峰值序列的中位数,以所述中位数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述中位数,则所述概率峰值为流行季/非流行季转折点。
- 如权利要求9所述的计算机装置,其特征在于,所述确定每个流行季/非流行季转折点的类型包括:计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型;或者计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型。
- 如权利要求9所述的计算机装置,其特征在于,所述获取流行病监测的时间序列数据包括:在预设区域建立由多个监测点组成的流行病监测网络,从所述监测点获取流行病监测数据,由所述流行病监测数据构成所述时间序列数据。
- 如权利要求9所述的计算机装置,其特征在于,所述确定所述概率 序列中的概率峰值包括:依次比较概率序列中每个概率与其前后相邻概率的大小,若该概率大于其前后相邻概率,则该概率为所述概率序列中的概率峰值。
- 一种非易失性可读存储介质,其特征在于,所述非易失性可读存储介质上存储有至少一个计算机可读指令,所述至少一个计算机可读指令被处理器执行时实现以下步骤:获取流行病监测的时间序列数据;计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率,得到一组概率序列;确定所述概率序列中的概率峰值,得到概率峰值序列;对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点;确定每个流行季/非流行季转折点的类型,所述类型包括上升转折点和下降转折点,所述上升转折点为流行病流行季的起始点,所述下降转折点为流行病流行季的终止点。
- 如权利要求15所述的存储介质,其特征在于,所述时间序列数据为X=[x 0,x 1,x 2,…,x t],时间点0为已知的上一个流行季/非流行季转折点,时间点t为给定时间点,r t表示时间点t到上一个流行季/非流行季转折点的间距,所述计算所述时间序列数据中每个流行病监测数据对应时间点属于流行季/非流行季转折点的概率包括:步骤a:计算时间点t下r t取值的概率:(1)计算时间点t的真实值x t基于[x 0…x t-1]序列分布的概率P(x t|x 1:t-1),其中μ 1:t-1表示[x 0,…,x t-1]的均值,σ 1:t-1表示[x 0,…,x t-1]的标准差;(2)计算上一时间点r t-1的概率P(r t-1),r t都是由上一时间点r t-1变化而来,r t=r t-1+1或r t=0;(3)计算r t-1转变成r t的概率P(r t|r t-1),P(r t|r t-1)为可调的参数值,P(r t|r t-1)越大,表明下一时间点t是流行季/非流行季转折点的概率越大;步骤b:计算时间点t属于流行季/非流行季转折点概率P t:P t(r t)=P t-1(r t-1)*P t(r t|r t-1)*P(x t|x 1:t-1)。
- 如权利要求15所述的存储介质,其特征在于,所述对所述概率峰值序列中的概率峰值进行筛选,根据筛选后的概率峰值得到所述时间序列数据的流行季/非流行季转折点包括:计算所述概率峰值序列的预设百分位数,以所述预设百分位数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述预设百分位数,则所述概率峰值为流行季/非流行季转折点;或者计算所述概率峰值序列的平均数,以所述平均数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述平均数,则所述概率峰值为流行 季/非流行季转折点;或者计算所述概率峰值序列的中位数,以所述中位数作为筛选的阈值,若所述所述概率峰值序列中的概率峰值大于所述中位数,则所述概率峰值为流行季/非流行季转折点。
- 如权利要求15所述的存储介质,其特征在于,所述确定每个流行季/非流行季转折点的类型包括:计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型;或者计算所述时间序列数据中,所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值,根据所述流行季/非流行季转折点到前一个流行季/非流行季转折点的均值以及所述流行季/非流行季转折点到后一个流行季/非流行季转折点的均值确定所述流行季/非流行季转折点的类型。
- 如权利要求15所述的存储介质,其特征在于,所述获取流行病监测的时间序列数据包括:在预设区域建立由多个监测点组成的流行病监测网络,从所述监测点获取流行病监测数据,由所述流行病监测数据构成所述时间序列数据。
- 如权利要求15所述的存储介质,其特征在于,所述确定所述概率序列中的概率峰值包括:依次比较概率序列中每个概率与其前后相邻概率的大小,若该概率大于其前后相邻概率,则该概率为所述概率序列中的概率峰值。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2019572831A JP6950008B2 (ja) | 2018-04-11 | 2018-08-09 | コンピュータ装置による感染症流行の予測方法、コンピュータ装置及び不揮発性読み取り可能な記憶媒体 |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810322421.4 | 2018-04-11 | ||
| CN201810322421.4A CN108630321B (zh) | 2018-04-11 | 2018-04-11 | 流行病预测方法、计算机装置及计算机可读存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019196283A1 true WO2019196283A1 (zh) | 2019-10-17 |
Family
ID=63705116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2018/099663 Ceased WO2019196283A1 (zh) | 2018-04-11 | 2018-08-09 | 流行病预测方法、计算机装置及非易失性可读存储介质 |
Country Status (3)
| Country | Link |
|---|---|
| JP (1) | JP6950008B2 (zh) |
| CN (1) | CN108630321B (zh) |
| WO (1) | WO2019196283A1 (zh) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767842B (zh) * | 2018-12-13 | 2023-08-22 | 平安科技(深圳)有限公司 | 一种疾病预警方法、疾病预警装置及计算机可读存储介质 |
| CN110047593A (zh) * | 2019-04-12 | 2019-07-23 | 平安科技(深圳)有限公司 | 疾病流行季等级的确定方法、装置、设备及可读存储介质 |
| CN111145917B (zh) * | 2020-01-20 | 2021-05-04 | 罗晖 | 一种面向流行病防控的大规模人口接触网络建模方法 |
| JP7570057B2 (ja) * | 2021-02-26 | 2024-10-21 | パナソニックIpマネジメント株式会社 | 情報通知システム、情報通知方法、及びプログラム |
| CN113948219B (zh) * | 2021-10-26 | 2025-07-22 | 平安科技(深圳)有限公司 | 传染病数据处理方法、装置、计算机及存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100274573A1 (en) * | 2006-03-09 | 2010-10-28 | Microsoft Corporation | Data relevation and pattern or event recognition |
| US20170316324A1 (en) * | 2016-04-27 | 2017-11-02 | Virginia Polytechnic Institute And State University | Computerized Event-Forecasting System and User Interface |
| CN107688872A (zh) * | 2017-08-20 | 2018-02-13 | 平安科技(深圳)有限公司 | 预测模型建立装置、方法及计算机可读存储介质 |
| CN107871538A (zh) * | 2016-12-19 | 2018-04-03 | 平安科技(深圳)有限公司 | 基于宏观因子的大数据预测方法及系统 |
Family Cites Families (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7457731B2 (en) * | 2001-12-14 | 2008-11-25 | Siemens Medical Solutions Usa, Inc. | Early detection of disease outbreak using electronic patient data to reduce public health threat from bio-terrorism |
| CN101339078A (zh) * | 2007-07-05 | 2009-01-07 | 王俪谕 | 禽畜流感疾病的预警装置 |
| JP2010146223A (ja) * | 2008-12-18 | 2010-07-01 | Hitachi Ltd | 行動抽出システム、行動抽出方法、及びサーバ |
| CN101894309A (zh) * | 2009-11-05 | 2010-11-24 | 南京医科大学 | 传染病疫情预测预警方法 |
| JP2011128935A (ja) * | 2009-12-18 | 2011-06-30 | Noriaki Aoki | 感染症予測システム |
| WO2014045351A1 (ja) * | 2012-09-19 | 2014-03-27 | 株式会社 東芝 | 時系列データ蓄積装置 |
| US20140095417A1 (en) * | 2012-10-01 | 2014-04-03 | Frederick S.M. Herz | Sdi (sdi for epi-demics) |
| JP6364800B2 (ja) * | 2014-02-10 | 2018-08-01 | オムロン株式会社 | 監視装置及び監視方法 |
| JP6471452B2 (ja) * | 2014-10-17 | 2019-02-20 | 日本電気株式会社 | 感染症流行予測装置、感染症流行予測方法及びプログラム |
| JP6569585B2 (ja) * | 2016-04-19 | 2019-09-04 | 株式会社デンソー | 業務評価システム |
| JP2017204214A (ja) * | 2016-05-13 | 2017-11-16 | 株式会社リコー | 情報処理装置及びプログラム |
| CN106599553B (zh) * | 2016-11-29 | 2019-08-16 | 中国科学院深圳先进技术研究院 | 疾病预警装置 |
| CN107633254A (zh) * | 2017-07-25 | 2018-01-26 | 平安科技(深圳)有限公司 | 建立预测模型的装置、方法及计算机可读存储介质 |
-
2018
- 2018-04-11 CN CN201810322421.4A patent/CN108630321B/zh active Active
- 2018-08-09 JP JP2019572831A patent/JP6950008B2/ja not_active Expired - Fee Related
- 2018-08-09 WO PCT/CN2018/099663 patent/WO2019196283A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100274573A1 (en) * | 2006-03-09 | 2010-10-28 | Microsoft Corporation | Data relevation and pattern or event recognition |
| US20170316324A1 (en) * | 2016-04-27 | 2017-11-02 | Virginia Polytechnic Institute And State University | Computerized Event-Forecasting System and User Interface |
| CN107871538A (zh) * | 2016-12-19 | 2018-04-03 | 平安科技(深圳)有限公司 | 基于宏观因子的大数据预测方法及系统 |
| CN107688872A (zh) * | 2017-08-20 | 2018-02-13 | 平安科技(深圳)有限公司 | 预测模型建立装置、方法及计算机可读存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108630321B (zh) | 2020-02-21 |
| JP6950008B2 (ja) | 2021-10-13 |
| JP2020527787A (ja) | 2020-09-10 |
| CN108630321A (zh) | 2018-10-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11961621B2 (en) | Predicting intensive care transfers and other unforeseen events using machine learning | |
| CN108597617B (zh) | 流行病分级预测方法及装置、计算机装置和可读存储介质 | |
| Kulldorff et al. | A maximized sequential probability ratio test for drug and vaccine safety surveillance | |
| WO2019196283A1 (zh) | 流行病预测方法、计算机装置及非易失性可读存储介质 | |
| US10026508B2 (en) | Geographic utilization of artificial intelligence in real-time for disease identification and alert notification | |
| WO2019196282A1 (zh) | 疾病异常数据检测方法及装置、计算机装置及存储介质 | |
| WO2019196280A1 (zh) | 疾病预测方法及装置、计算机装置及可读存储介质 | |
| WO2019196286A1 (zh) | 疾病预测方法及装置、计算机装置及可读存储介质 | |
| US20210052231A1 (en) | Method and system for analyzing risk associated with respiratory sounds | |
| CN107658023A (zh) | 疾病预测方法、装置、介质和电子设备 | |
| WO2019196279A1 (zh) | 疾病异常数据检测方法及装置、计算机装置及存储介质 | |
| Jarvis et al. | Are observation selection methods important when comparing early warning score performance? | |
| Beil et al. | On predictions in critical care: the individual prognostication fallacy in elderly patients | |
| Huybrechts et al. | Active surveillance of the safety of medications used during pregnancy | |
| Kramer et al. | Severity of illness and predictive models in society of critical care medicine’s first 50 years: A tale of concord and conflict | |
| Arbelle et al. | Multiple chronic disorders-health care system’s modern challenge in the Maccabi Health Care System | |
| Indriany et al. | Mobile Health Assisted Self Monitoring in Heart Failure Patients To Ensure Quality of Life: A Scoping Review | |
| US20180349558A1 (en) | Systems and methods for autonomous discharge queue management | |
| Carson et al. | Predicting adolescent suicidal behavior following inpatient discharge using structured and unstructured data | |
| CN120032787A (zh) | 一种基于二维码腕带伤员信息快速登记方法及系统 | |
| Wang et al. | A conformal regressor for predicting negative conversion time of Omicron patients | |
| Hooft et al. | Validation of a risk-prediction model for pediatric post-discharge mortality after hospital admission in Rwanda | |
| Raghupathi et al. | An empirical analysis of the status of country-level public health | |
| Guevara Rodríguez et al. | Incidence of cerebrovascular disease in Peru from 2015 to 2023 | |
| NYONI et al. | Analyzing New HIV Infections in Pregnant Women at Gweru District Hospital Using Artificial Neural Networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18914333 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2019572831 Country of ref document: JP Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 18914333 Country of ref document: EP Kind code of ref document: A1 |