Method for counting number of hospital daytime outpatient service patients based on mobile phone signaling

Abstract

The invention relates to a method for counting the number of hospital daytime outpatient service visitors based on mobile phone signaling, comprising the following steps: (1) dividing a target area according to urban functional areas, where office areas and industrial parks are collectively called working areas; (2) finding out a mobile phone user of which the working place or the residence place is not located in a medical area by using a mobile phone signaling, and naming the mobile phone user as a first user; (3) calculating the movement speed and the staying duration of the first user based on the mobile phone signaling, and if the first user meets a preset condition, considering that the first user is in a staying state in a medical area, and naming the first user as a second user for short; and (4) when the staying time period of the second user is the working time of the hospital daytime outpatient service and the staying time length meets a preset staying time length range, identifying that the second user is a patient seeing the daytime outpatient service. According to the invention, the number of daily outpatient service patients in each region can be accurately counted, and the problem that official published or field survey data can only cover part of hospitals is effectively overcome.

Description

Technical Field

[0001] The invention relates to a method for counting the number of people visiting a functional area based on mobile phone signaling, and in particular to a method for counting the number of patients visiting a daytime outpatient clinic in a hospital based on mobile phone signaling. Background Art

[0002] Hospital outpatient volume refers to the total number of outpatients received by the hospital within a specific period of time. The statistics and analysis of outpatient volume are of great significance to hospital management and public health. For example, by analyzing the outpatient volume and its influencing factors, especially the changes in different seasons, different climates and different types of diseases, hospitals can promptly discover the outbreak trend of epidemics or infectious diseases. At present, the acquisition of outpatient volume data usually relies on official data release or tracking surveys of the population. The data lacks timeliness and comprehensiveness, and cannot meet the actual needs of statistical analysis in the current public health field.

[0003] Some algorithms based on mobile phone signaling to identify visitors to functional areas can be used to count outpatient visits. For example, patent CN115809378A discloses a method for obtaining travel information of users arriving at the hospital after excluding users whose place of employment and residence are located in the medical area. However, travel is not the same as medical treatment. Some users who pass by the medical area are also counted. The data obtained cannot directly replace the daytime outpatient volume data. In addition, this method does not take into account the medical habits of outpatients in different regions, and does not further exclude interfering user data who stay in the medical area for a short time but do not receive medical treatment based on the movement characteristics and rules of local outpatients. The data accuracy needs to be further improved.

[0004] Therefore, there is still a lack of a method that can adapt to different regional characteristics, accurately eliminate interference data, and quickly and accurately count the number of patients visiting hospital day clinics based on mobile phone signaling. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method which can adapt to the characteristics of different regions, accurately eliminate interference data, and quickly and accurately count the number of patients visiting the hospital daytime clinic based on mobile phone signaling.

[0006] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0007] This embodiment discloses a method for counting the number of patients visiting a hospital day clinic based on mobile phone signaling, including the following steps:

[0008] (1) The target area is divided into medical area, residential area, recreation area, commercial service area, office area and industrial park according to urban functional areas, among which office area and industrial park are collectively referred to as work area;

[0009] (2) Using mobile phone signaling, find out the mobile phone users whose workplaces or residences are located in the medical area. The mobile phone users in the target area excluding those whose workplaces or residences are located in the medical area are called first users.

[0010] (3) The movement speed and stay time of the first user are calculated based on the mobile phone signaling. If the first user meets one of the following conditions, the first user is considered to be in the stay state of the medical area and is referred to as the second user:

[0011] Speed ​​<V1;

[0012] The speed is in the interval [V1, V2], and the dwell time is > T1;

[0013] The speed is in the interval (V2, V3], and the dwell time is > T2;

[0014] Among them, V1<V2<V3; V1, V2, V3, T1, T2 are all preset values;

[0015] (4) When the second user's stay time period is within the working hours of the hospital's day clinic and the stay time meets the preset stay time range, the second user is identified as a day clinic patient, and the sum of the day clinic patients is the number of day clinic patients.

[0016] The beneficial effects brought by the present invention are: compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0017] 1. The data is more accurate: In the process of identifying patients, the present invention not only excludes users whose place of employment and residence is located in the medical area, but also excludes users whose exercise status does not conform to the medical treatment rules, so that the statistical data is more accurate.

[0018] 2. The method has strong adaptability: The retention time condition for excluding interference data in the present invention is summarized based on the medical treatment habits of users in different regions. Therefore, it can break through geographical restrictions and accurately count the number of daytime outpatient clinic patients in various regions, effectively overcoming the problem that official releases or field survey data can only cover some hospitals.

[0019] 3. Fast data update speed: The present invention is based on statistical data obtained by mobile phone signaling. Since the mobile phone signaling data is updated very quickly, the data on the number of daytime clinic patients can be updated in real time, which can better make up for the lack of timeliness of official releases or field surveys, and can assist the government or relevant scientific researchers to promptly detect the outbreak trend of epidemics or infectious diseases.

[0020] 4. Classified statistical data lays the foundation for subsequent research: The present invention classifies and statistics the data of weekends and weekdays, tertiary hospitals and non-tertiary hospitals. If the number of long-term daytime outpatients in primary hospitals in the target area is low, and the number of tertiary hospitals in the area is too high, it can be judged that there are certain problems with the quality of medical services in primary hospitals in the area, which can help the government and relevant institutions evaluate whether the allocation of medical resources in the target area is reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below in conjunction with the accompanying drawings.

[0022] Figure 1 It is a flowchart of identifying day clinic patients based on mobile phone signaling in this embodiment.

[0023] Figure 2 This is a base station distribution diagram of the signaling data used in this embodiment.

[0024] Figure 3 This is the spatial distribution diagram of functional areas used in this embodiment to identify patients visiting the day clinic. DETAILED DESCRIPTION

[0025] In order to more clearly introduce the technical process, purpose and significance of the present invention, further explanation is given in conjunction with examples and drawings.

[0026] like Figure 1 As shown, this embodiment provides a method for counting the number of patients visiting the hospital day clinic based on mobile phone signaling, which is as follows:

[0027] (1) The target area is divided into medical area, residential area, recreation area, commercial service area, office area and industrial park according to urban functional areas, among which office area and industrial park are collectively referred to as work area.

[0028] This embodiment takes Nanjing as the target area. First, the target area is divided into a grid, and the row and column numbers corresponding to each base station in the grid are calculated based on the longitude and latitude of the base station, and the grid coordinates are used to replace the longitude and latitude coordinates of the base station: This embodiment is based on the mobile phone signaling (including 10604 people and 1424613 records) provided by the telecom operator and its base station distribution data. According to the service range of the Nanjing base station, Nanjing is taken as the origin (118.350°E, 31.220°N), the direction along the increasing longitude is set as the x-axis, and the direction along the increasing latitude is set as the y-axis, and a coordinate system is constructed to divide Nanjing into a grid (3454 rows × 2208 columns) with an accuracy of 0.0004° (approximately equal to 44m). According to the longitude and latitude of the base station, the row and column numbers corresponding to each base station in the grid are calculated in turn, and the longitude and latitude coordinates of the base station are replaced by this. The following rules apply to base stations that fall on grid boundaries: if a base station is located on two grid boundaries that are adjacent along the x-axis, the grid coordinates on the left of the two grid boundaries are assigned to the base station; if a base station is located on two grid boundaries that are adjacent along the y-axis, the grid coordinates on the upper side of the two grid boundaries are assigned to the base station. The number of grid base stations based on the divided Nanjing grid is as follows: Figure 2 shown.

[0029] Secondly, based on the 2020 Nanjing point of interest data crawled from the open interface of Gaode Map and the 2020 geographic national conditions survey data provided by the Jiangsu Basic Geographic Information Center, the urban functional areas are marked with land use surface data. They are specifically divided into medical areas, residential areas, recreational areas, commercial service areas, office areas, and industrial parks, of which office areas and industrial parks are collectively referred to as work areas. In order to more accurately identify the number of patients, during implementation, the medical area is usually expanded outward to a certain preset range to form a hospital visit identification area, which is used as the medical area for identifying the number of patients. For example: This embodiment takes into account that the service range of a single base station is 250m, and the area with a radius of 250m with the center of each hospital as the center is set as the medical area. The generated functional area data is as follows: Figure 3 shown.

[0030] (2) Using mobile phone signaling, find out the mobile phone users whose workplaces or residences are located in the medical area. The mobile phone users in the target area excluding those whose workplaces or residences are located in the medical area are called the first users:

[0031] First, the noise of the mobile phone signaling data is reduced. In this embodiment, the following method for reducing data noise is preferred. It should be noted that the following method is only exemplary, and other examples may use different methods:

[0032] Eliminate incomplete data: For users whose total signaling duration is less than 20 hours or whose signaling messages are less than 5 in a day, it is impossible to form a complete signaling record time chain, so the records of these users are eliminated.

[0033] Eliminate data drift: Data drift can cause the speed calculated based on the signal to increase sharply and the dwell time to be short, so remove the signal with a dwell time of less than 5 seconds and the signal with a speed greater than 120 km / h. Recalculate the speed of each remaining signal and iterate three times to eliminate data drift.

[0034] Eliminate the ping-pong effect: For the ping-pong effect, for each user, traverse its daily signaling. Starting from the fourth signaling, for each signaling, if at least one of the three previous signals has the same coordinates as the signaling and the stay time is less than 60 seconds, it is considered a ping-pong effect and deleted. Recalculate the stay time of each remaining signaling and iterate twice to eliminate the ping-pong effect.

[0035] Next, find out the mobile phone users whose workplace or residence is located in the medical area: the method for finding out the workplace and residence disclosed in the patent document with publication number CN115809378A can be adopted. The present embodiment preferably uses the following method: within a time range of 2 months, extract the places where each mobile phone user stays the longest between 9:00-11:30 and 14:00-18:00 every day as the places where the longest stay is during the day; extract the places where the longest stay is between 0:00-6:00 and 20:00-24:00 as the places where the longest stay is at night. In this way, a point set of the places where the longest stay is during the day and a point set of the places where the longest stay is at night within the specified time range are formed.

[0036] In the point sets with the longest stay time during the day and at night, find out the places where each user appears most frequently and the frequency of appearance is greater than 25%, and mark them as anchor points; use the anchor point as the center and a buffer zone with a radius of 750m (approximately equal to 15 neighborhoods), delete the points distributed outside the buffer zone, so as to form the work point group and residence point group of each user.

[0037] Finally, the mobile phone users in the target area who work or live in the medical area are removed, and the remaining customers are called the first users.

[0038] (3) The movement speed and stay time of the first user are calculated based on the mobile phone signaling. If the first user meets one of the following conditions, the first user is considered to be in the stay state in the medical area and is referred to as the second user. Preferably, this embodiment proposes the following specific values ​​based on experimental experience. It should be noted that the specific values ​​should be interpreted as merely exemplary rather than limiting, and other examples may have different values:

[0039] Speed ​​<0.6 km / h;

[0040] The speed is within the interval [0.6,5.0] km / h and the stay time is >1200 seconds;

[0041] The speed is within the interval (5.0,10.0] km / h and the stay time is >500 seconds;

[0042] The method of calculating speed based on mobile phone signaling is to arrange the mobile phone signaling of the same user in ascending order in chronological order within a day. For each mobile phone signaling (except the first mobile phone signaling), find the distance between it and the previous mobile phone signaling, calculate the ratio of the distance to the dwell time recorded by the mobile phone signaling, and get the speed of the mobile phone signaling in km / h.

[0043] (4) When the second user's stay time period is within the working hours of the hospital's day clinic and the stay time meets the preset stay time range, the second user is identified as a day clinic patient, and the sum of the day clinic patients is the number of day clinic patients.

[0044] Preferably, this embodiment assigns the attributes of tertiary hospitals and non-tertiary hospitals to the medical area based on the hospital grade classification data provided by the Jiangsu Provincial Health Commission. If the signaling coordinates of a second user fall within multiple hospital identification areas, the hospital to which the user is heading is preferentially marked as a tertiary hospital;

[0045] Preferably, in this embodiment, the working hours of the hospital day clinic are set to 9:00-18:00 on the same day;

[0046] Taking into account that under normal circumstances, the number of people going to hospitals on weekends is more than on weekdays, and the number of people going to tertiary hospitals is more than non-tertiary hospitals, preferably, this embodiment stipulates the time range conditions for going to tertiary hospitals and non-tertiary hospitals on weekdays and weekends respectively: the preset stay time on weekdays is longer than the preset stay time on weekends; the preset stay time in tertiary hospitals is longer than the preset stay time in non-tertiary hospitals.

[0047] (5) Find the length of stay range parameter that keeps the number of hospital daytime outpatient visits based on mobile phone signaling identification statistics consistent with the actual average daily number of visits in the undisturbed medical area, and use this to correct the preset length of stay range.

[0048] The non-interference medical area refers to a medical area that does not overlap with other functional areas or only overlaps with residential areas and work areas. Preferably, the following three non-interference medical areas are selected in Nanjing in this embodiment:

[0049] Table 1 List of selected non-interference medical areas

[0050] Hospital Name level Level Administrative region coordinate Nanjing Tongren Hospital Level 3 B Jiangning District 31.90，118.82 Taikang Xianlin Gulou Hospital Level 3 Not rated Qixia District 32.10，118.93 Nanjing Meishan Hospital Level 2 Grade A Yuhuatai District 31.90，118.62

[0051] Adjust the second user's stay time range parameter in the hospital and calculate the number of hospital daytime outpatient visits based on mobile phone signaling recognition statistics, so that the calculation result is close to the actual average daily number of visits in the undisturbed medical area:

[0052] Table 2 Comparison of the number of day clinic patients in the non-interference medical area under the specified threshold

[0053]

[0054] The parameters for the duration of stay are determined as follows:

[0055] During working days:

[0056] Grade 3A hospitals: 1.9 hours ≤ length of stay ≤ 6 hours;

[0057] Non-Grade 3A hospitals: 1.3 hours ≤ length of stay ≤ 6 hours;

[0058] During rest days:

[0059] Grade 3A hospitals: 2.0 hours ≤ length of stay ≤ 8 hours;

[0060] Non-Grade 3A hospitals: 1.6 hours ≤ length of stay ≤ 8 hours;

[0061] (6) Calculate the ratio of the number of daytime clinic patients identified by mobile phone signaling in the interfered medical area to the actual average daily number of patients in the corresponding hospital. The average of the ratios is called the error coefficient. The number of daytime clinic patients identified by mobile phone signaling in the target area is corrected according to the error coefficient:

[0062] The interfered medical area refers to the medical area overlapping with the rest area and the commercial service area. Preferably, the following representative interfered medical areas are selected in this embodiment:

[0063] Table 3 List of selected disturbed medical areas

[0064] Hospital Name level Level Administrative region coordinate Nanjing Jiangbei Hospital Level 3 B Pukou District 32.24，118.75 Nanjing Liuhe District People's Hospital Level 3 No Comments Liuhe District 32.35，118.84 Nanjing Hospital of Integrated Traditional Chinese and Western Medicine Level 2 Grade A Xuanwu District 32.04，118.85 Nanjing Red Cross Hospital Level 2 B Qinhuai District 32.03，118.79

[0065] Calculate the ratio of the number of daytime outpatient visits in the affected medical area based on mobile phone signaling identification to the actual average daily number of visits in the corresponding hospital. The average of the ratio is called the error coefficient.

[0066] Table 4 Comparison of the number of patients visiting the day clinic in the disturbed medical areas

[0067]

[0068] The number of day clinic patients identified based on mobile phone signaling in Nanjing area during the period from November 7 to November 13, 2022 is corrected according to the error coefficient. Preferably, the correction method of this embodiment is to divide the number of day clinic patients in all hospitals calculated based on mobile phone signaling in the target area by the error coefficient.

[0069] The calculated results are compared with the actual average daily number of patients in Nanjing announced by the Nanjing Municipal Health Commission, as shown in Table 5. The results show that they can meet the overall judgment of the changes in the number of patients visiting hospitals.

[0070] Table 5 Verification of the number of patients visiting the day clinics in Nanjing hospitals

[0071]

Claims

1. A method for counting the number of patients visiting a hospital day clinic based on mobile phone signaling, characterized in that: The following steps are involved: (1) The target area is divided into medical area, residential area, recreation area, commercial service area, office area and industrial park according to urban functional areas, among which office area and industrial park are collectively referred to as work area; (2) Using mobile phone signaling to find mobile phone users whose workplaces or residences are located in the medical area, the mobile phone users in the target area excluding those whose workplaces or residences are located in the medical area are called first users; (3) The movement speed and stay time of the first user are calculated based on the mobile phone signaling. If the first user meets one of the following conditions, the first user is considered to be in the stay state of the medical area and is referred to as the second user: Speed ​​<V1; The speed is in the interval [V1, V2], and the dwell time is > T1; The speed is in the interval (V2, V3], and the dwell time is > T2; Among them, V1<V2<V3; V1, V2, V3, T1, T2 are all preset values; (4) When the second user's stay time period is within the working hours of the hospital's day clinic and the stay time meets the preset stay time range, the second user is identified as a day clinic patient, and the sum of the day clinic patients is the number of day clinic patients.

2. A method for counting the number of patients visiting a hospital day clinic based on mobile phone signaling as claimed in claim 1, characterized in that: A medical area that does not overlap with other functional areas or only overlaps with residential areas and work areas is defined as an interference-free medical area. The stay time range parameters are found such that the number of hospital daytime outpatient visits based on mobile phone signaling recognition statistics is consistent with the actual average daily number of visits in the interference-free medical area, so as to correct the preset stay time range.

3. The method for counting the number of patients visiting the daytime clinic of a hospital based on mobile phone signaling as claimed in claim 1, characterized in that: The medical area overlapping with the rest area and commercial service area is defined as the interfered medical area. The ratio of the number of daytime outpatient clinic patients identified based on mobile phone signaling in the interfered medical area to the actual average daily number of patients in the corresponding hospital is calculated. The average of the ratios is called the error coefficient. The number of daytime outpatient clinic patients identified based on mobile phone signaling in the target area is corrected according to the error coefficient.

4. The method for counting the number of patients visiting a hospital day clinic based on mobile phone signaling as claimed in claim 2, characterized in that: Different preset stay time ranges are set for weekdays and weekends, and the preset stay time for weekdays is greater than the preset stay time for weekends.

5. The method for counting the number of patients visiting the daytime clinic of a hospital based on mobile phone signaling as claimed in claim 2, characterized in that: The medical area is assigned the attributes of a tertiary hospital or a non-tertiary hospital according to the hospital grade. If the signaling coordinates of the second user's mobile phone fall within multiple medical areas, the hospital he is heading to is marked as a tertiary hospital. Different preset stay time ranges are set for tertiary hospitals and non-tertiary hospitals, and the preset stay time of a tertiary hospital is greater than the preset stay time of a non-tertiary hospital.

6. The method for counting the number of patients visiting a hospital day clinic based on mobile phone signaling as claimed in claim 1, characterized in that: Before counting the first user, the mobile phone signaling is subjected to noise reduction, and the noise reduction method includes: eliminating the mobile phone signaling that cannot form a complete record time chain; eliminating data drift; and eliminating the ping-pong effect.

Images