Infection propagation path sensing method and system based on time-space relationship

Abstract

The invention discloses an infection propagation path sensing method and system based on a time-space relationship. The method comprises the following steps: acquiring and cleaning hospitalization, nursing and inspection data of patients, and constructing a time sequence file of each patient; defining the contact type of the same ward and the medical care medium based on the time sequence file, setting an infection period and an observation window to identify a potential infection transmission path, and forming a contact event table; sorting the time information of the positive patients and the contactors thereof based on the contact event table, and dynamically generating a directed propagation chain; performing cyclic path elimination and credibility grading on the directed propagation chain to obtain a corrected propagation network; and screening out cases without upstream infection sources according to the corrected propagation network, and determining infection sources and properties thereof in combination with hospitalization time and historical hospitalization records. Through the implementation of the method, the adaptive limitation of the prior art to a complex propagation scene can be solved, rapid reconstruction and accurate attribution of a hospital infection propagation chain are realized, and intelligent decision support is provided for reducing the incidence rate of hospital infection.

Description

Technical Field

[0001] This invention relates to medical big data processing methods, and more specifically to a method and system for sensing infection transmission paths based on spatiotemporal relationships. Background Technology

[0002] In modern medical settings, identifying and managing the transmission routes of hospital-acquired infections is crucial for ensuring patient safety. However, current mainstream methods primarily rely on manual epidemiological surveys and static time models to analyze and identify transmission routes. This approach has several significant drawbacks. First, testing delays can lead to misjudgments of the infectious period. Due to the unavoidable time delay between sample collection and result issuance, the actual infection situation may change during this period. For example, some infections may have entered a new stage of transmission while awaiting test results, resulting in inaccurate assessments of the infectious period. Second, the coarse granularity of contact tracing is also a significant issue. Existing contact tracing methods typically record patient activity on a daily basis. This coarse temporal resolution fails to capture close contact events occurring within a short period. For some rapidly spreading pathogens, this coarse tracing method may result in the loss of critical contact information, thereby affecting the accurate identification of the source of infection and the effectiveness of isolation measures.

[0003] Furthermore, the difficulty in attributing infections across hospitalization cycles is a major challenge for existing technologies. Especially in complex scenarios such as drug-resistant bacterial outbreaks, where patients may receive treatment over multiple hospitalization cycles, current methods struggle to accurately attribute a particular infection to a specific hospitalization cycle or procedure. This exacerbates the difficulty of locating the source of infection and makes targeted prevention and control measures difficult to implement. The limitations of static time models further highlight these problems. Traditionally used static time models often assess potential transmission risks based on fixed observation windows. This approach ignores the dynamic nature of infectious diseases, easily leading to missed transmission periods, misjudgments of reverse transmission, and lack of attribution for cross-cycle infections. The limitations of static models are particularly evident in situations requiring highly precise control, such as drug-resistant bacterial outbreaks, resulting in delayed infection control measures and increased prevention and control costs.

[0004] Therefore, it is necessary to design a new method to address the limitations of existing technologies in adapting to complex transmission scenarios, achieve rapid reconstruction and accurate attribution of hospital infection transmission chains, and provide intelligent decision support for reducing the incidence of hospital-acquired infections. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for sensing infection transmission paths based on spatiotemporal relationships.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: an infection transmission path sensing method based on spatiotemporal relationships, comprising:

[0007] Acquire and clean patient hospitalization, nursing, and laboratory data to build a timeline profile for each patient;

[0008] Based on the time-series records, the types of contact with patients in the same ward and with healthcare professionals are defined, and the infectious period and observation window are set to identify potential infection transmission routes and form a contact event table;

[0009] Based on the time information of positive patients and their contacts sorted by the contact event table, a directed transmission chain is dynamically generated.

[0010] The directed propagation chain is subjected to loop path elimination and credibility classification to obtain the corrected propagation network;

[0011] Based on the corrected transmission network, cases without upstream sources of infection were screened out, and the source and nature of the infection were determined by combining the admission time and historical hospitalization records.

[0012] The further technical solution is as follows: The acquisition and cleaning of patient hospitalization, nursing, and laboratory data, and the construction of a time-series profile for each patient, includes:

[0013] Obtain patient hospitalization, nursing, and laboratory data from the hospital information system to obtain key data;

[0014] The key data is standardized to ensure that all timestamp formats are consistent, and missing time data with second-level precision is supplemented. False discharge records are excluded and incomplete ward transfer records are corrected. At the same time, the first positive time of each patient is determined to obtain the cleaning results.

[0015] Based on the cleaning results, a timeline is constructed for each patient, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, to obtain a chronological profile for each patient.

[0016] The further technical solution is as follows: the key data includes the patient's hospitalization records, department transfer information, nursing operation records and microbiological test results, and includes admission and discharge times, ward transfer time points, medical staff service records and their specific operation times and patient identifiers, as well as the sample number and test results.

[0017] The further technical solution is as follows: the contact in the same ward includes the overlap of the stay time of two patients in the same ward area by no less than 1 hour; the contact between medical staff and patients includes the same medical staff serving different patients within 1 hour; the window of the infectious period is 48 hours before and 7 days after the patient's first positive test; the observation window is 7 days after the last contact.

[0018] The further technical solution is as follows: The dynamic generation of a directed transmission chain based on the time information of positive patients and their contacts sorted according to the contact event table includes:

[0019] Based on the contact event table, all positive cases are sorted according to the time of the first positive test, forming a queue arranged in chronological order;

[0020] For each positive patient in the cohort, all contacts within the infectious period window are extracted, and those who meet the criteria are screened out to establish a directed transmission edge, thus obtaining a directed transmission chain.

[0021] The further technical solution is as follows: The process of eliminating loop paths and classifying the credibility of the directed propagation chain to obtain a corrected propagation network includes:

[0022] Eliminate contradictory paths that violate the time sequence in the directed propagation chain, and retain only unidirectional edges that conform to epidemiological patterns;

[0023] The reliability of each transmission link is quantified based on factors such as contact type and the time difference between the last contact and a positive result;

[0024] Eliminate ineligible transmission links and ensure that all transmission events occur during hospitalization to obtain a corrected transmission network.

[0025] The further technical solution is as follows: Based on the corrected transmission network, cases without upstream sources of infection are screened out, and the source and nature of infection are determined by combining the admission time and historical hospitalization records, including:

[0026] Cases without upstream sources of infection were selected as the initial source of infection based on the corrected transmission network;

[0027] The difference between the first positive time and the admission time is calculated for the initial source of infection. The source of infection is determined by combining the historical hospitalization records. The initial source of infection is classified as community infection, current hospitalization infection, or previous hospitalization infection.

[0028] For cases marked as having been hospitalized previously, their historical hospitalization records are traced until the exact source of infection is found;

[0029] When a patient tests positive at different anatomical sites and the time of testing at the new sites meets specific conditions, their infection type is reassessed and updated.

[0030] The further technical solution is as follows: For cases marked as having previously hospitalized infections, tracing their historical hospitalization records until the exact source of infection is found also includes:

[0031] If the historical hospitalized infection type is community infection, the current source is marked as community infection; if the historical hospitalized infection type is current hospitalized infection, the current source is marked as previous hospitalized infection.

[0032] The further technical solution is as follows: the specific conditions include that the time difference between the first positive test time and the admission time is not higher than 48 hours.

[0033] This invention also provides an infection transmission path sensing system based on spatiotemporal relationships, comprising:

[0034] The acquisition unit is used to acquire and clean patient hospitalization, nursing, and laboratory data to build a time-series profile for each patient.

[0035] The contact event table generation unit is used to define the types of contact with patients in the same ward and with medical staff based on the time sequence file, and to set the infectious period and observation window to identify potential infection transmission paths and form a contact event table;

[0036] The transmission chain generation unit is used to dynamically generate a directed transmission chain based on the time information of positive patients and their contacts sorted in the contact event table.

[0037] An optimization unit is used to perform loop path elimination and credibility classification on the directed propagation chain to obtain a corrected propagation network.

[0038] The source tracing unit is used to screen out cases without upstream sources of infection based on the corrected transmission network, and to determine the source of infection and its nature by combining the admission time and historical hospitalization records.

[0039] The advantages of this invention compared to existing technologies are as follows: This invention constructs a time-series archive by acquiring and cleaning patient hospitalization, nursing, and laboratory data, defines the types of contact with patients in the same ward and with healthcare providers, and identifies potential infection transmission paths by combining the infectious period and observation window, forming a contact event table. Subsequently, based on this table, it dynamically generates directed transmission chains by sorting the time information of positive patients and their contacts. After eliminating cyclical paths and grading credibility, a corrected transmission network is obtained. Finally, cases without upstream sources of infection are screened out, and the source of infection and its nature are determined by combining admission time and historical hospitalization records, thus solving the problem of insufficient adaptability of existing technologies in handling complex transmission scenarios. This method can quickly reconstruct hospital infection transmission chains and achieve accurate attribution, providing powerful intelligent decision support for reducing the incidence of nosocomial infections. This systematic approach not only improves the understanding and prediction of infection transmission paths but also enhances the response efficiency and control effectiveness of infection outbreaks through the integrated analysis of multi-source data.

[0040] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A schematic diagram illustrating an application scenario of the infection transmission path sensing method based on spatiotemporal relationships provided in this embodiment of the invention;

[0043] Figure 2 A flowchart illustrating the infection transmission path sensing method based on spatiotemporal relationships provided in an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a sub-process of the infection transmission path perception method based on spatiotemporal relationship provided in an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of a sub-process of the infection transmission path perception method based on spatiotemporal relationship provided in an embodiment of the present invention;

[0046] Figure 5 A schematic diagram of a sub-process of the infection transmission path perception method based on spatiotemporal relationship provided in an embodiment of the present invention;

[0047] Figure 6 A schematic diagram of a sub-process of the infection transmission path perception method based on spatiotemporal relationship provided in an embodiment of the present invention;

[0048] Figure 7 A schematic block diagram of an infection transmission path sensing system based on spatiotemporal relationships provided in an embodiment of the present invention;

[0049] Figure 8 A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0052] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0053] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0054] Please see Figure 1 and Figure 2 , Figure 1 This is a schematic diagram illustrating an application scenario of the infection transmission path perception method based on spatiotemporal relationships provided in an embodiment of the present invention. Figure 2 This is a schematic flowchart illustrating the infection transmission path perception method based on spatiotemporal relationships provided in this invention. This method is applied in a server. The server interacts with the terminal, acquiring and cleaning patient hospitalization, nursing, and laboratory data to construct a time-series profile, defining the types of contact with the same ward and healthcare providers, setting infectious periods and observation windows to identify potential transmission paths, dynamically generating directed transmission chains, eliminating cyclical paths, and classifying credibility to form a corrected transmission network. Finally, cases without upstream sources of infection are screened, and the source and nature of infection are determined by combining admission time and historical records. This method overcomes the limitations of existing technologies in handling complex transmission scenarios, achieving rapid reconstruction and accurate attribution of infection transmission chains within hospitals. It provides intelligent decision support for reducing the incidence of hospital-acquired infections. In particular, through precise analysis of patient activity trajectories, healthcare worker service periods, and the time of the first positive test, as well as the time-sequencing of contact event tables and dynamic transmission chain generation, it ensures strict adherence to epidemiological patterns and quantitative assessment of the reliability of transmission links, thereby improving the accuracy and efficiency of infection source determination.

[0055] Figure 2 This is a flowchart illustrating the infection transmission path sensing method based on spatiotemporal relationships provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S150.

[0056] S110. Acquire and clean patient hospitalization, nursing and laboratory data, and build a timeline profile for each patient.

[0057] In this embodiment, a time-series archive refers to a structured data format that records each patient's activity trajectory within the hospital, their contact with medical staff, and related microbiological test results.

[0058] Specifically, the time-series archive contains the following:

[0059] Patient activity trajectory: This includes the patient's admission time, discharge time, transfer times between wards or departments, and the duration of their stay. This information allows for a clear tracking of the patient's movement paths within different areas of the hospital.

[0060] Contact with healthcare personnel and service periods: This records which healthcare personnel provided services to patients and the specific time periods (accurate to the minute). This data is crucial for identifying potential healthcare-associated transmission routes because it helps determine which healthcare personnel may have come into contact with multiple patients within the same time period, thus becoming a medium for infection transmission.

[0061] First positive test time: For each patient, record the exact time of their earliest positive test. This is crucial for establishing the infectious period window and observation window, and subsequently analyzing the time-series relationship of infection transmission.

[0062] The standardized data processing ensures that all timestamps are in the same format and completes any missing time data with second-level precision. Furthermore, it excludes false discharge records and corrects incomplete ward transfer records to improve data accuracy and completeness.

[0063] Time-series archives organize the above information chronologically to form a detailed log of a patient's activities during their hospital stay. This not only helps in understanding the infection risk of individual patients but also has significant implications for the development of infection control strategies, rapid identification of infection sources, and the implementation of effective prevention and control measures throughout the hospital. In short, time-series archives are a fundamental data source for analyzing hospital infection transmission pathways based on spatiotemporal relationships.

[0064] In one embodiment, please refer to Figure 3 The above-mentioned step S110 may include steps S111 to S113.

[0065] S111. Obtain patient hospitalization, nursing, and laboratory data from the hospital information system to obtain key data.

[0066] In this embodiment, the key data includes the patient's hospitalization records, department transfer information, nursing operation records, and microbiological test results, including admission and discharge times, ward transfer times, medical staff service records and their specific operation times and patient identifiers, as well as the submitted sample number and test results.

[0067] First, key data is collected from the hospital information system (HIS), including but not limited to:

[0068] Hospitalization record: admission time, discharge time, and time of ward transfer.

[0069] Departmental transfer information: This includes historical hospitalization records, used to track patient movements between different departments.

[0070] Nursing procedure records: These are service records of healthcare personnel, including the specific time of the procedure and the patient's identifier. This helps to determine the contact between the patient and healthcare personnel.

[0071] Microbiological test results: include the submitted sample number, patient identifier, sample type, submission time, and test results, especially positive test results.

[0072] Specifically, the patient hospitalization record includes the patient identifier (Patient_ID), admission department, admission time, discharge department, and discharge time; the department transfer information includes the patient identifier, department, admission time, and discharge time; the nursing service record includes: medical staff identifier, service patient identifier (Patient_ID), start and end time of nursing operation (accurate to the minute, such as intravenous infusion 09:00-09:30), and operation type (optional, such as dressing change, blood collection, etc.); the microbiological test record includes: sample number, patient identifier (Patient_ID), sample type (such as sputum, blood), submission time (accurate to the minute), and test results (including: pathogen name, positive / negative determination).

[0073] S112. Standardize the key data to ensure that all timestamp formats are consistent, fill in missing time data with second-level precision, exclude false discharge records and correct incomplete ward transfer records, and determine the first positive time for each patient to obtain the cleaning results.

[0074] In this embodiment, the cleaning result refers to obtaining accurate, complete and consistent time-series archive data by standardizing the original data, filling in missing timestamps, excluding false discharge records, and correcting incomplete ward transfer records.

[0075] The next step is to standardize the collected data to ensure consistency and accuracy. This process mainly includes:

[0076] Standardized timestamp format: Convert all time data from different sources into a standard format to facilitate subsequent time calculations and comparisons.

[0077] Complete missing time data with second-level precision: For data that is missing to the second, it is padded with 00 seconds by default to ensure the integrity of the time series.

[0078] Eliminating fake discharge records: By analyzing the time interval between a patient's admission and discharge, fake discharge records that do not reflect the actual situation can be identified and removed.

[0079] Correcting incomplete ward transfer records: For ward transfers without complete start or end times, calculate a reasonable transfer time based on adjacent records.

[0080] Determine the first positive time: For multiple positive test results of the same patient, select the earliest time as the patient's first positive time, which will serve as an important basis for subsequent analysis.

[0081] Specifically, timestamps from different systems will be uniformly converted to a standard format (example: 2023-08-20T14:30:00+08:00).

[0082] For time data lacking second-level precision, the default padding is 00 seconds (e.g., 09:15-09:15:00);

[0083] Excluding patients with fake discharge records:

[0084] Patients who meet the condition of "current admission time - previous discharge time ≤ 48 hours" are judged as having been discharged falsely, and the intermediate admission and discharge process is removed.

[0085] Complete ward transfer records:

[0086] If a transfer lacks a start time: take the departure time from the previous department + 1 minute;

[0087] If a transfer lacks an end time: take the arrival time of the next department - 1 minute;

[0088] Patients with no history of metastasis: will remain in the admitting department by default.

[0089] Positive labeling processing:

[0090] For multiple positive records from the same patient, the earliest submission time shall be taken as the first positive time.

[0091] Invalid test records with sampling times later than the discharge time are excluded.

[0092] S113. Based on the cleaning results, construct a timeline for each patient that includes their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, so as to obtain a timeline file for each patient.

[0093] Based on the cleaned data, a detailed time-series profile is constructed for each patient, including:

[0094] Activity trajectory: Records the time periods patients spend in various departments within the hospital.

[0095] Contact with healthcare staff and service periods: Record which healthcare staff you have been in contact with and the specific time periods of contact.

[0096] First positive test time: Marks the exact time point when a patient first tests positive.

[0097] Specifically, a timeline object is constructed for each patient, and the following events are merged in chronological order:

[0098] Department Transfer Event Table: Patient Unique Identifier, Admission Department, Admission Time, Discharge Department, Discharge Time; The department transfer information includes patient identifier, department, admission time, and discharge time;

[0099] Nursing Service Event Table: Nursing staff identification, patient unique identifier, nursing start time, nursing end time;

[0100] Positive Detection Event Table: Patient Unique Identifier, First Positive Test Submission Time.

[0101] Through these three steps, S111 to S113, a clear, chronological activity trajectory can be effectively established for each patient, accurately reflecting all their contacts during their hospital stay. This is crucial for subsequent spatiotemporal contact modeling, identification of potential transmission routes, and final source determination. Furthermore, this method not only improves data quality but also provides reliable data support for the automated identification of infection transmission pathways.

[0102] S120. Based on the time-series archive, define the types of contact with patients in the same ward and between medical staff and the media, and set the infectious period and observation window to identify potential infection transmission routes and form a contact event table.

[0103] In this embodiment, the contact event table refers to a table that records details of contact events between patients through the same ward or medical care intermediaries, including the type of contact, time, patient ID involved, whether the patient is in the infectious period, and whether the patient is within the observation window.

[0104] The term "contact in the same ward" refers to two patients whose stay in the same ward overlaps by at least one hour; the term "contact between healthcare professionals" refers to the same healthcare professional serving different patients within one hour; the window of the infectious period is from 48 hours before to 7 days after the patient's first positive test; the observation window is 7 days after the last contact.

[0105] The contact in the same ward includes comparing the timeline of the two patients' ward transfer records, calculating the overlap interval of their stay time, and generating a contact event if the overlap time is ≥60 minutes.

[0106] When two patients spend at least one hour (60 minutes) overlapping in their stay in the same ward, it is considered that there has been contact in the same ward.

[0107] Specifically, the departmental records of any two patients are traversed.

[0108] If the two cases overlap by at least 60 minutes in the same ward, a ward contact event is generated.

[0109] The contact event record includes: contact type (marked as same ward), contact start time, and contact end time.

[0110] If the same healthcare worker serves different patients within one hour, it is considered that these patients have had contact through that healthcare worker, i.e., healthcare-mediated contact.

[0111] For each medical staff member, all their nursing procedures are arranged in chronological order.

[0112] Iterate through adjacent operation records. If the above conditions are met and the nursing time overlaps by at least 1 hour, then generate a healthcare-media contact event.

[0113] The contact event log includes: contact type (marked as healthcare intermediary), contact start time, contact end time, and associated patient pair.

[0114] For each positive patient, the infectious period window is defined as 48 hours before and 7 days after the first positive test (a total of 9 days). If the infectious period begins earlier than the admission date, the admission date is used as the starting time.

[0115] For each contact event, the 7-day period following the last contact is defined as the observation window.

[0116] After identifying the various contact types and their time windows, the next step is to construct a contact event table. This table records the unique identifier (UUID) for each contact event, the patient IDs involved (source of infection and exposed person), the contact type, the start and end times of the contact, and the first positive time of the source of infection used for time-series verification.

[0117] In addition, it is necessary to mark whether the contact events meet the infectious period and observation window to ensure that only potential transmission routes that meet the criteria are retained.

[0118] In summary, by defining clear contact types and time windows, and constructing a detailed contact event table based on these criteria, it is possible to effectively identify potential infection transmission routes within hospitals, providing a scientific basis for further analysis and control of hospital infections.

[0119] In this embodiment, a database of potential contact events between patients is constructed through time overlap analysis and association rules among medical staff, providing a data foundation for subsequent transmission chain analysis.

[0120] In this embodiment, firstly, for determining contact within the same ward, for any two patients (P1, P2), all their departmental records are iterated. If a pair of departmental records meets the following conditions: same department > time overlap ≥ 60 minutes; output: generate a contact event within the same ward, recording: contact type (same ward), contact start time, and contact end time. Secondly, for determining medical staff contact, for the same medical staff member (S), all their nursing operations are sorted in chronological order; adjacent operation records (C) are iterated. i -C i+1 If C is satisfied: i+1 Nursing start time - C i Nursing care ending time is greater than or equal to 1 hour and the unique identifiers of the caring patients are different (the caring patients are not the same person); nursing care time overlap is ≥ 1 hour; generate a healthcare contact event and record: contact type (healthcare intermediary), contact start time, contact end time, associated patient pair (C i .Patient_ID-C i+1.Patient_ID). Next, based on the positive test event, for each positive patient P, calculate: Infectious period start time = P. First positive test submission time - 48 hours, Infectious period end time = P. First positive test submission time + 168 hours; if the infectious period start time is earlier than the admission time, then the admission time is taken as the start time; for the end time T of each contact event, set: Observation window end time = T + 168 hours. Then, construct a contact time table and record: Contact event unique identifier (UUID), potential source patient ID (from the infectious period window), potential exposed patient ID, contact type (same ward / medical care intermediary), contact start time, contact end time, and the first positive time of the source of infection (for time sequence verification). Next is the "S204 Time Window Association Marking". For each record in the contact event, the following calculations are performed: 1) Does the contact event occur within the infectious period window of the potential source patient? The contact start time must be ≥ (first positive time of the source patient - 48h) and the contact end time ≤ (first positive time of the source patient + 168h). If these conditions are met, the infectious period window of the potential source patient is marked (Boolean value). 2) The observation window coverage of the exposed patient is calculated: If a positive record exists for the potential exposed patient ID, the first positive time of the potential exposed patient must be ≤ (contact end time + 168h). The observation window of the exposed patient is marked (Boolean value). Finally, for multiple consecutive contact events of the same patient pair (e.g., multiple contacts in the same ward), if the interval is ≤ 24 hours, they are merged into one event. 3) Invalid contact removal: Records meeting the following criteria are deleted: The potential source patient and the potential exposed patient are the same person; the contact duration = 0 (due to time calculation error); the potential exposed patient has been discharged by the end of the contact.

[0121] For example, Patient A: Departmental trajectory: Respiratory Department (August 1, 2023, 8:00 AM to August 5, 2023, 10:00 AM); First positive time: August 3, 2023, 2:00 PM. Patient B: Departmental trajectory: Respiratory Department (August 2, 2023, 12:00 PM to August 4, 2023, 9:00 AM); First positive time: August 5, 2023, 9:00 AM. Contact event calculation: Contact time in the same ward: August 2, 2023, 12:00 PM to August 4, 2023, 9:00 AM (duration = 45 hours). Infectious window of A: August 1, 2023, 2:00 PM to August 10, 2023, 2:00 PM. Contact occurred within A's infectious window → Infectious window of the source patient = TRUE. B's positive time falls within the observation window (exposure end time + 168h = 2023-08-11 09:00) - the observation window of the exposed patient = TRUE. Because A's first positive time is earlier than B's, an effective transmission link AB is formed.

[0122] S130. Based on the time information of positive patients and their contacts sorted by the contact event table, a directed transmission chain is dynamically generated.

[0123] In this embodiment, the directed transmission chain refers to the sorting of patients based on the time of their first positive test, and the screening of eligible contacts based on the infectious period window and the observation window, thereby establishing a directed transmission path from the source patient to the infected person, so as to dynamically display the potential infection transmission relationship.

[0124] In one embodiment, please refer to Figure 4 The above-mentioned step S130 may include steps S131 to S132.

[0125] S131. Based on the contact event table, sort all positive cases according to the time of the first positive test, forming a queue arranged in chronological order.

[0126] In this embodiment, firstly, based on the data in the contact event table, all positive cases are sorted in ascending order according to the time of each patient's first positive test. The purpose of this step is to create a chronologically ordered queue for subsequent analysis of the infectious window and potential transmission routes for each positive patient. In this way, it can be ensured that the construction of the transmission chain follows the "first-infected, first-spread" principle.

[0127] Specifically, extract all positive test records from the contact event table; sort them according to the time of each patient's first positive test (take the earliest time); create an ordered list or queue to arrange these positive patients in order of their first positive test time from earliest to latest.

[0128] S132. For each positive patient in the cohort, extract all contacts within the infectious period window, screen out contacts that meet the criteria, establish a directed transmission edge, and obtain a directed transmission chain.

[0129] In this embodiment, for each positive patient Pi (i.e., the source of infection) in the queue, the following steps are performed to determine its possible transmission path and generate a directed transmission chain:

[0130] Specifically, based on the time window rules set in step S202, the infectious period window of positive patient Pi (from 48 hours before to 168 hours after the first positive test result) is calculated. Within this time period, all other patients (C) who have had contact with Pi are identified. j As potential exposed individuals.

[0131] Screening for eligible contacts: For all contacts C identified above j Further screening will be conducted to retain only contacts who meet the following criteria:

[0132] The contact occurred within the infectious window of the positive patient Pi;

[0133] The first positive time of the contact was within the observation window after the contact event (i.e., the time of end of contact + 168 hours), and this time was later than the first positive time of the source of infection.

[0134] Establish a directed propagation edge: for each pair of positive patients P that meet the above conditions i Its corresponding contact person C j Establish a path from P i Pointing to C j A directed propagation edge indicates the possible existence of a path from P. i To C j The infection transmission route.

[0135] Suppose there are two positive patients, A and B:

[0136] Patient A's first positive test was at 14:00 on August 3, 2023. Therefore, the infectious window for this patient is from 14:00 on August 1, 2023 to 14:00 on August 10, 2023.

[0137] Patient B's first positive test was at 9:00 AM on August 5, 2023, and the time spent in the same ward as Patient A overlapped by more than 60 minutes, which meets the definition of contact in the same ward.

[0138] If patient B's first positive test occurs within the infectious window of patient A, and the positive result occurs within 168 hours after patient A's last contact with patient B, then an effective transmission link from A to B is considered to exist.

[0139] Specifically, a directed propagation network is constructed by combining temporal constraints with cross-validation of contact events to ensure that all propagation directions follow the epidemiological principle of "first-to-infect, first-to-spread." First, three main datasets were collected: a positive patient cohort, a contact event database, and a patient timeline. The positive patient cohort was obtained from the positive test event table and sorted in ascending order according to the time of the first positive test. The contact event database contains the infectious period window of potential source patients and the observation window of exposed patients. For each positive patient, their infectious period window is pre-calculated as the period from the time of the first positive test minus 48 hours to the time of the first positive test plus 168 hours, and a hash index is built to map the patient ID to its infectious period window. When multiple possible sources of infection exist, the source of infection from the last contact is prioritized, as the most recent contact is more likely to cause infection; if the last contact times are the same, the source of infection from the earliest positive result is selected. Next, an initial transmission network was established. Events in this network include a unique identifier for the transmission side (UUID), the source patient ID, the infected patient ID, the contact type (e.g., ward or healthcare provider), the time of first contact, the time of last contact, the time difference between the last contact and positive exposure, and whether it is a primary transmission route. To accommodate newly added positive cases, the system employs an incremental update mechanism. New cases are simply added sequentially to the sorted queue, and their association as a source or infected individual is reassessed. This not only effectively tracks disease transmission routes but also reflects the latest transmission dynamics in a timely manner.

[0140] This method allows for the dynamic generation of a directed transmission network throughout the hospital, helping to identify potential infection transmission pathways and support decision-making. This process requires strict adherence to temporal and spatial constraints to ensure the accuracy and reliability of the transmission chain.

[0141] S140. Perform loop path elimination and credibility classification on the directed propagation chain to obtain the corrected propagation network.

[0142] In this embodiment, the corrected transmission network refers to an optimized transmission path structure that conforms to epidemiological logic and has clear reliability for each transmission link, generated by eliminating cyclical paths that violate the time sequence, determining unidirectional transmission relationships based on the patient's first positive time, and excluding interference from out-of-hospital transmission.

[0143] In one embodiment, please refer to Figure 5 The above-mentioned step S140 may include steps S141 to S143.

[0144] S141. Eliminate contradictory paths that violate the time order in the directed propagation chain, and retain only unidirectional edges that conform to epidemiological laws.

[0145] In this embodiment, firstly, it is necessary to eliminate all contradictory paths in the directed propagation chain that violate the temporal order, ensuring that each edge conforms to the epidemiological law of "first-infect, first-spread." This step mainly includes:

[0146] Circular path identification: to find whether there is a mutual transmission path between the same pair of patients (e.g., AB and BA coexist) or a circular structure (e.g., ABCA).

[0147] Conflict resolution: Compare the order of the first positive time of the two patients, retain only the one-way edge from the earlier to the later first positive time, and delete the reverse edge. For the closed-loop case, locate the patient with the earliest first positive time (denoted as X), and delete all incoming edges pointing to X, retaining only X and its downstream propagation path.

[0148] S142. Quantify the reliability of each transmission link based on the contact type and the time difference between the last contact and the positive result.

[0149] In this embodiment, the reliability of each transmission link is then quantitatively assessed based on factors such as contact type and the time difference between the last contact and a positive result. The specific grading rules are as follows:

[0150] Level I (high confidence): The contact type is ward contact; the infected person has no other potential sources of infection (only one source of infection); the time difference between the last contact and the positive result does not exceed the average incubation period of the pathogen (e.g., 7 days for COVID-19).

[0151] Level II (requires verification): The contact type is healthcare-mediated contact; the infected person has multiple potential sources of infection; the time difference between the last contact and the positive result significantly exceeds the average incubation period.

[0152] Level III (Suspected): Meets the time window criteria (overlapping infectious period and observation window), but lacks clear evidence of contact with patients in the same ward or with medical staff.

[0153] S143. Eliminate ineligible transmission links to ensure that all transmission events occur during hospitalization, thus obtaining a corrected transmission network.

[0154] In this embodiment, finally, those transmission chains that do not conform to the pattern of occurring during hospitalization are eliminated to ensure that all transmission events occur during the patient's hospitalization. Specific operations include:

[0155] Time range constraint: Remove any contact time of any source of infection that is earlier than the time of admission, or any contact time of the infected person that is later than the time of discharge.

[0156] Status verification: If any patient was not hospitalized when the contact occurred (e.g., visiting an outpatient clinic or being treated in the emergency room but not admitted), then the transmission link is eliminated.

[0157] By following these three steps, the transmission network can be effectively corrected, illogical loops eliminated, the reliability of each transmission link quantified, and unreasonable transmission events removed, resulting in a more accurate and reliable corrected transmission network. This corrected network not only helps to accurately determine the source of infection but also supports hospitals in taking targeted prevention and control measures to effectively control the spread of hospital-acquired infections.

[0158] In this embodiment, the correction and classification of the transmission network is a crucial step, aiming to ensure that the transmission path conforms to epidemiological patterns and quantify the reliability of each link. First, in step S401, the goal of loop link detection is to eliminate contradictory paths in the transmission network that violate temporal order. This includes three main operations: First, loop path identification is performed to check whether there are mutual transmission paths between the same pair of patients (e.g., AB and BA coexisting) or forming a ring structure (e.g., ABCA). Next is conflict resolution, which compares the order of the first positive time of two patients, retaining only the one-way edge pointing from the earlier first positive time to the later one, and deleting the reverse edges. Finally, for loop closure processing, the patient with the earliest first positive time (denoted as X) needs to be located in the loop, all incoming edges pointing to X (e.g., the CX edge in the loop) are deleted, and only X and its downstream transmission paths are retained.

[0159] Subsequently, in the confidence grading, the reliability of each transmission link will be quantified to support prioritization analysis. The grading rules are as follows: Level I (high confidence) indicates that the contact type is ward contact, the infected person has no other potential sources of infection (only one source of infection), and the time difference from the last contact to the positive result does not exceed the average incubation period of the pathogen (e.g., 7 days for COVID-19). Level II (requires verification) refers to the contact type being healthcare worker contact, the infected person having multiple potential sources of infection, and the time difference from the last contact to the positive result significantly exceeding the average incubation period. Level III (suspected) only meets the time window condition (the infectious period overlaps with the observation window), but lacks clear evidence of ward or healthcare worker contact.

[0160] Next, in eliminating interference from outside the hospital, it is essential to ensure that all transmission events occur during hospitalization. Specific measures include: based on time constraints, removing any links where the contact time with any source of infection is earlier than the patient's admission time, or where the contact time with the infected person is later than the patient's discharge time; and status verification, i.e., if any patient was not hospitalized when the contact occurred (e.g., outpatient visit, emergency room visit without hospitalization), then that transmission link should be eliminated.

[0161] Furthermore, in the generation of propagation subnets, the complex network is split into independent propagation subtrees to facilitate source tracing analysis. This process includes three parts: root node identification, subtree expansion, and independent subnet labeling. Root node identification selects all nodes with an in-degree of 0 (i.e., cases without upstream sources of infection) as the root nodes of the subtrees; subtree expansion starts from each root node and traverses downstream layer by layer according to the propagation direction, recording the maximum depth (propagation generations) and the total number of cases for each subtree; if two subtrees have overlapping nodes, they are merged into the same subnet; if there is no propagation association between the root nodes, they are marked as independent subnets.

[0162] Finally, in the post-correction network verification, it is necessary to ensure that the corrected network conforms to epidemiological logic. This includes temporal consistency checks (the first positive time of the source of infection for all edges must be earlier than that of the infected person), spatial validity checks (the contact event must occur during the hospitalization of both parties and in the same department), integrity checks (each infected person has at least one valid source of infection, unless it is the root node), and the absence of unprocessed cycles or contradictory paths. For example, in a scenario, Patient A (first positive: September 1st) - Patient B (September 3rd) - Patient C (September 5th) - Patient A (illegal cycle). During the correction process, an illegal cycle edge C→A is detected. Comparing the first positive times, it is found that A (September 1st) is earlier than C (September 5th), so the CA edge is deleted, and the ABC link is retained. This series of operations ensures the accuracy and reliability of the transmission network.

[0163] S150. Based on the corrected transmission network, cases without upstream sources of infection are screened out, and the source of infection and its nature are determined by combining the admission time and historical hospitalization records.

[0164] In one embodiment, please refer to Figure 6 The above-mentioned step S150 may include steps S151 to S154.

[0165] S151. Based on the corrected transmission network, cases without upstream sources of infection are selected as the initial sources of infection.

[0166] In this embodiment, firstly, based on the corrected transmission network, nodes or cases without any arrows pointing to them (i.e., without an upstream source of infection) are identified. These nodes represent independent sources of infection, also known as initial sources of infection.

[0167] S152. Calculate the difference between the first positive time and the admission time for the initial source of infection, and determine the source of infection by combining it with historical hospitalization records. Classify the initial source of infection as community infection, current hospitalization infection, or previous hospitalization infection.

[0168] In this embodiment, for each case identified as the initial source of infection, the time difference (ΔT) between the time of the first positive test and the time of admission is calculated. Then, the source of infection is preliminarily determined by combining the patient's historical hospitalization records.

[0169] If ΔT is greater than 48 hours, it is marked as "infection during this hospitalization".

[0170] If ΔT is less than or equal to 48 hours and there is no hospitalization record in this hospital in the past 14 days, it is initially marked as "community infection or infection in other healthcare facilities".

[0171] If there is a hospitalization record within the past 14 days, further investigation is needed to confirm whether it is a "previous hospitalized infection".

[0172] S153. For cases marked as having been hospitalized in the past, trace their historical hospitalization records until the exact source of infection is found.

[0173] In this embodiment, for cases marked as "previous hospitalized infections," their historical hospitalization records need to be traced until the exact source of infection is found. This process may involve multiple iterations, each time checking the patient's infection status from their most recent hospitalization and updating the current infection source classification according to the following rules:

[0174] If the previous hospitalization was marked as a community infection, then the current source should also be marked as a community infection.

[0175] If the previous hospitalization was marked as the current hospitalization infection, then the current source should be marked as a previous hospitalization infection.

[0176] If the source of infection cannot be determined, continue tracing back until clear evidence is found or the earliest hospitalization record is reached.

[0177] S154. When a patient tests positive at different anatomical sites and the new site test time meets specific conditions, reassess and update the infection type.

[0178] If the historical hospitalized infection type is community infection, the current source is marked as community infection; if the historical hospitalized infection type is current hospitalized infection, the current source is marked as previous hospitalized infection.

[0179] The specific conditions include that the time difference between the first positive test and the admission time is no more than 48 hours.

[0180] When a patient tests positive for infection at different anatomical sites, their infection type needs to be reassessed and updated if the timing of the new site test meets certain criteria (e.g., the time interval between the new site test and the previous site test exceeds 48 hours). In particular, if the new positive test result occurs during the current hospitalization, it may be reclassified as "current hospitalization infection," even if there were previous positive tests.

[0181] By following the steps described above, the true source of infection can be accurately identified from the complex internal infection transmission network of the hospital, and its nature can be scientifically and rationally determined, providing an important basis for effective infection control.

[0182] In this embodiment, after locating the transmission chain of this infection, the nature of the infection is determined for the source patient at the top of the transmission chain. The nature of the source patient's infection is divided into three categories: community infection or infection in other medical institutions, current hospitalization, and previous hospitalization infection.

[0183] The source patient was not infected within two calendar days of admission, but was diagnosed with infection after two calendar days, which is considered to be an infection during this hospitalization.

[0184] If the source patient is diagnosed with infection within two calendar days of admission, and the system shows that there is hospitalization data in this hospital within 14 days (the patient had been hospitalized before), then it is a previous hospitalization infection.

[0185] If the source patient is diagnosed with infection within two calendar days of admission, and no previous data within the past 14 days can be found, it is considered an infection from another healthcare facility or the community. However, if infection occurs in different sites after two calendar days, it is considered an infection during this hospitalization.

[0186] For cases where the source of infection was a previous hospitalized patient, the nature of the previous infection needs to be confirmed:

[0187] If the previous hospitalization was caused by infection in the community or other medical institutions, update the previous hospitalization infection record to indicate that it was caused by infection in the community or other medical institutions.

[0188] If the previous hospitalization resulted in infection during this hospitalization, then it's considered a previous hospitalization infection and doesn't need updating. If the previous hospitalization was without infection, then it was infected in the community or other healthcare facilities.

[0189] If the previous hospitalization was due to an infection contracted during a previous hospitalization, then continue searching for the category of the previous hospitalization.

[0190] Identify the initial source of infection in the transmission chain and determine its nature (community infection, current hospitalization infection, previous hospitalization infection).

[0191] Screen all cases in the transmission network that have no upstream source of infection (i.e., no arrow pointing to that case).

[0192] Each candidate source generates an independent transmission subtree. If multiple subtrees have overlapping cases, the possibility of cross-infection needs to be manually verified.

[0193] Where ΔT = first positive time - admission time. If ΔT > 48 hours, it is marked as this hospitalized infection; otherwise, it is entered into the historical tracing process.

[0194] Check the patient's hospitalization records within 14 days prior to admission: if there is no history of hospitalization, mark it as community / other institution infection; if there is a history of hospitalization, trigger recursive source tracing.

[0195] Tracing stops when the nature of the infection in the historical hospitalization record is different from previous infections; if the historical records are exhausted, it is marked as a community / other institution infection (insufficient evidence).

[0196] For example, the current patient's hospitalization ΔT = 36 hours, and they have a history of hospitalization within the past 14 days;

[0197] Historical Hospitalization 1: Infection type is current hospitalization infection; final classification is previous hospitalization infection;

[0198] Positive results from different anatomical sites in the same patient (e.g., sputum-blood).

[0199] If the ΔT time for a new positive site is greater than 48 hours, it will be forcibly marked as an infection during this hospitalization.

[0200] For example, patient A has a positive sputum test (ΔT = 24 hours - preliminary classification: community infection); 3 days later, a positive blood test (ΔT = 96 hours - reclassification: infection during this hospitalization);

[0201] The final output results are shown in Table 1 below.

[0202] Table 1. Output Results

[0203]

[0204] The timeline propagation diagram is as follows:

[0205] X-axis: Cases are sorted by the time of first positive result;

[0206] Arrow colors: red (high confidence), orange (medium confidence), grayish-white (low confidence);

[0207] Distribution map of infection sources:

[0208] Coloring by ward: The shade of color indicates the number of initial sources of infection generated in that area.

[0209] In this embodiment, the generation of the propagation link must simultaneously meet the following conditions:

[0210] The contact incident occurred within the infectious window of the infected patient;

[0211] The first positive test of the patient who was in contact with the infected person occurred within the observation window and later than the first positive test of the source of infection.

[0212] The propagation network correction includes a loop path processing rule: when a closed loop of patient A-patient B-patient A is detected, the single path pointing from the patient with the earlier first positive time to the patient with the later time is retained.

[0213] Recursive tracing includes:

[0214] If the current hospitalization ΔT is less than or equal to 48 hours and there is a historical hospitalization record, trace the infection type of that historical hospitalization.

[0215] If the historical hospitalized infection type is community infection, then the current source is marked as community infection;

[0216] If the historical hospitalized infection type is the current hospitalized infection, then the current source is marked as a previous hospitalized infection.

[0217] Multi-site infection coverage rule: When a patient tests positive at different anatomical sites and the time to the new site test meets ΔT>48 hours, the infection type is forcibly updated to the current hospitalization infection.

[0218] The generated spatiotemporal propagation network diagram includes: nodes arranged from left to right according to the time of the first positive result; the direction of the edge arrows indicates the propagation path.

[0219] The method in this embodiment sets an infectious period window (48 hours before to 7 days after the first positive result) and an observation window (7 days after the last contact) for hospital-acquired infections, overcoming the limitations of traditional symmetrical time models. This asymmetric time window design can effectively cover both the incubation period and the viral shedding period, thereby improving the accuracy of identifying the infection transmission period.

[0220] The time-driven propagation direction correction mechanism ensures that the propagation direction conforms to the actual sequence of events by requiring the first positive result of the propagation source node to occur earlier than that of the propagated node, and eliminates loop paths based on this principle. This method solves the problem of misjudgment in backpropagation caused by detection delay, and improves the accuracy and reliability of the propagation network.

[0221] For cases infected within 48 hours of admission, the algorithm traces the patient's historical hospitalization records over the past 14 days, establishing a three-tiered termination rule (community infection marker, three recursion depths, and a 21-day time threshold) to achieve penetrating attribution of nested infection sources. This helps address the problem of missing attribution for trans-period infections.

[0222] When a patient tests positive across anatomical sites (such as the respiratory, bloodstream, or urinary tract) and the new site's ΔT is greater than 48 hours, the system will automatically update the infection type to the current hospitalization infection. This strategy addresses the potential for missed diagnoses that can occur with single-site testing.

[0223] The aforementioned methods are systematically supported by a three-dimensional spatiotemporal relationship database (including patient-bed-time minute-level trajectories) and interactive visualization modules (such as temporal propagation network diagrams and ward heat maps). These tools not only provide a complete technical system from minute-level spatiotemporal modeling to dynamic propagation network generation and intelligent source tracing decision-making, but also offer a precise and efficient closed-loop solution for hospital infection outbreaks such as those caused by drug-resistant bacteria. By integrating spatiotemporal contact logic and a dynamic time window mechanism, the methods solve the problem of misjudgment of propagation direction due to detection delays in traditional methods, and achieve cross-period infection attribution through multi-level source tracing algorithms. This provides strong technical support for the rapid location and precise control of hospital infection outbreaks.

[0224] The aforementioned method for perceiving infection transmission paths based on spatiotemporal relationships constructs a time-series profile by acquiring and cleaning patient hospitalization, nursing, and laboratory data. It defines the types of contact within the same ward and between healthcare providers and patients, and identifies potential transmission paths by combining the infectious period and observation window, forming a contact event table. Subsequently, based on this table, it dynamically generates directed transmission chains by sorting the time information of positive patients and their contacts. After loop path elimination and credibility grading, a corrected transmission network is obtained. Finally, cases without upstream sources of infection are screened out, and the source and nature of infection are determined by combining admission time and historical hospitalization records. This addresses the problem of insufficient adaptability of existing technologies in handling complex transmission scenarios. This method can quickly reconstruct hospital infection transmission chains and achieve accurate attribution, providing powerful intelligent decision support for reducing the incidence of nosocomial infections. This systematic approach not only improves the understanding and prediction of infection transmission paths but also enhances the response efficiency and control effectiveness of infection outbreaks through the integrated analysis of multi-source data.

[0225] Figure 7 This is a schematic block diagram of an infection transmission path sensing system 300 based on spatiotemporal relationships provided in an embodiment of the present invention. Figure 7 As shown, corresponding to the above-described infection transmission path sensing method based on spatiotemporal relationships, the present invention also provides an infection transmission path sensing system 300 based on spatiotemporal relationships. This infection transmission path sensing system 300 includes a unit for executing the above-described infection transmission path sensing method based on spatiotemporal relationships, and the system can be configured in a server. Specifically, please refer to... Figure 7The infection transmission path perception system 300 based on spatiotemporal relationships includes an acquisition unit 301, a contact event table generation unit 302, a transmission chain generation unit 303, an optimization unit 304, and a source tracing unit 305.

[0226] The acquisition unit 301 is used to acquire and clean patient hospitalization, nursing, and testing data to construct a time-series profile for each patient; the contact event table generation unit 302 is used to define the types of contact with patients in the same ward and with healthcare providers based on the time-series profile, and to set the infectious period and observation window to identify potential infection transmission paths and form a contact event table; the transmission chain generation unit 303 is used to dynamically generate a directed transmission chain based on the time information of positive patients and their contacts sorted by the contact event table; the optimization unit 304 is used to perform loop path elimination and credibility grading on the directed transmission chain to obtain a corrected transmission network; the source tracing unit 305 is used to screen out cases without upstream sources of infection based on the corrected transmission network, and to determine the source of infection and its nature by combining the admission time and historical hospitalization records.

[0227] In one embodiment, the acquisition unit 301 includes:

[0228] The data acquisition subunit is used to acquire patient hospitalization, nursing, and testing data from the hospital information system to obtain key data. The cleaning subunit is used to standardize the key data, ensure that all timestamp formats are consistent, fill in missing time data with second-level precision, exclude false discharge records, correct incomplete ward transfer records, and determine the first positive time for each patient to obtain the cleaning results. The construction subunit is used to construct a timeline for each patient based on the cleaning results, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of the first positive test, to obtain a chronological profile for each patient.

[0229] In one embodiment, the propagation chain generation unit 303 includes:

[0230] The sorting subunit is used to sort all positive cases according to the first positive test time based on the contact event table, forming a queue arranged in chronological order; the establishment subunit is used to extract all contacts within the infectious period window for each positive patient in the queue, screen out contacts that meet the conditions, and establish a directed transmission edge to obtain a directed transmission chain.

[0231] In one embodiment, the optimization unit 304 includes:

[0232] The elimination subunit is used to eliminate contradictory paths that violate the time sequence in the directed transmission chain, retaining only unidirectional edges that conform to epidemiological laws; the quantification subunit is used to quantify the reliability of each transmission link based on the contact type and the time difference between the last contact and the positive result; the elimination subunit is used to eliminate transmission links that do not meet the conditions, ensuring that all transmission events occur during hospitalization to obtain the corrected transmission network.

[0233] In one embodiment, the traceability unit 305 includes:

[0234] The screening subunit is used to screen out cases without upstream sources of infection as the initial source of infection based on the corrected transmission network;

[0235] The classification subunit is used to calculate the difference between the first positive time and the admission time for the initial source of infection, and to determine the source of infection by combining it with historical hospitalization records, classifying the initial source of infection as community infection, current hospitalization infection, or previous hospitalization infection. The source determination subunit is used to trace the historical hospitalization records of cases marked as previous hospitalization infections until the exact source of infection is found; if the historical hospitalization infection type is community infection, the current source is marked as community infection; if the historical hospitalization infection type is current hospitalization infection, the current source is marked as previous hospitalization infection. The evaluation subunit is used to re-evaluate and update the infection type when a patient shows positive tests at different anatomical sites and the detection time at the new site meets specific conditions.

[0236] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned infection transmission path perception system 300 based on spatiotemporal relationships and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0237] The aforementioned infection transmission path sensing system 300 based on spatiotemporal relationships can be implemented as a computer program, which can, for example... Figure 8 It runs on the computer device shown.

[0238] Please see Figure 8 , Figure 8 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0239] See Figure 8 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0240] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a spatiotemporal infection propagation path awareness method.

[0241] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0242] The internal memory 504 provides an environment for the execution of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute an infection propagation path perception method based on spatiotemporal relationships.

[0243] This network interface 505 is used for network communication with other devices. Those skilled in the art will understand that... Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0244] The processor 502 is used to run a computer program 5032 stored in the memory to perform the following steps:

[0245] Acquire and clean patient hospitalization, nursing, and laboratory data to construct a time-series profile for each patient; define the types of contact with patients in the same ward and with healthcare providers based on the time-series profile, and set the infectious period and observation window to identify potential infection transmission paths, forming a contact event table; sort the time information of positive patients and their contacts based on the contact event table to dynamically generate directed transmission chains; perform loop path elimination and credibility grading on the directed transmission chains to obtain a corrected transmission network; screen out cases without upstream sources of infection based on the corrected transmission network, and determine the source of infection and its nature by combining the admission time and historical hospitalization records.

[0246] In one embodiment, when implementing the step of acquiring and cleaning patient hospitalization, nursing, and laboratory data, and constructing a time-series profile for each patient, the processor 502 specifically implements the following steps:

[0247] Patient hospitalization, nursing, and laboratory data from the hospital information system are obtained to acquire key data. This key data is then standardized to ensure that all timestamps are in the same format, and missing time data with second-level precision is supplemented. False discharge records are excluded, and incomplete ward transfer records are corrected. Simultaneously, the first positive time for each patient is determined to obtain the cleansing results. Based on the cleansing results, a timeline is constructed for each patient, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, to obtain a chronological profile for each patient.

[0248] The key data includes the patient's hospitalization records, department transfer information, nursing operation records, and microbiological test results, including admission and discharge times, ward transfer times, medical staff service records and their specific operation times and patient identifiers, as well as the sample numbers and test results.

[0249] In one embodiment, when the processor 502 implements the step of dynamically generating a directed propagation chain based on the time information of positive patients and their contacts sorted from the contact event table, the processor 502 specifically implements the following steps:

[0250] Based on the contact event table, all positive cases are sorted according to the time of the first positive test, forming a queue arranged in chronological order; for each positive patient in the queue, all contacts within the infectious period window are extracted, and contacts that meet the criteria are screened out to establish a directed transmission edge to obtain a directed transmission chain.

[0251] In one embodiment, when the processor 502 performs the step of eliminating loop paths and classifying the credibility of the directed propagation chain to obtain the corrected propagation network, it specifically implements the following steps:

[0252] Eliminate contradictory paths that violate the time sequence in the directed transmission chain, retaining only unidirectional edges that conform to epidemiological patterns; quantify the reliability of each transmission link based on contact type and the time difference between the last contact and the positive result; remove transmission links that do not meet the criteria, ensuring that all transmission events occur during hospitalization, in order to obtain the corrected transmission network.

[0253] In one embodiment, when implementing the step of screening out cases without upstream sources of infection based on the corrected transmission network and determining the source of infection and its nature by combining the admission time and historical hospitalization records, the processor 502 specifically implements the following steps:

[0254] Based on the corrected transmission network, cases without upstream sources of infection are selected as initial sources of infection. The difference between the first positive time and the admission time is calculated for the initial source of infection, and the source of infection is determined by combining historical hospitalization records. The initial source of infection is classified as community infection, current hospitalization infection, or previous hospitalization infection. For cases marked as previous hospitalization infection, their historical hospitalization records are traced until the exact source of infection is found. When a patient shows positive results at different anatomical sites and the detection time at the new site meets specific conditions, the infection type is reassessed and updated.

[0255] The specific conditions include that the time difference between the first positive test and the admission time is no more than 48 hours.

[0256] In one embodiment, after implementing the step of tracing the historical hospitalization records of cases marked as having previously hospitalized infections until the exact source of infection is found, the processor 502 further implements the following steps:

[0257] If the historical hospitalized infection type is community infection, the current source is marked as community infection; if the historical hospitalized infection type is current hospitalized infection, the current source is marked as previous hospitalized infection.

[0258] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0259] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0260] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein when executed by a processor, the computer program causes the processor to perform the following steps:

[0261] Patient hospitalization, nursing, and laboratory data from the hospital information system are obtained to acquire key data. This key data is then standardized to ensure that all timestamps are in the same format, and missing time data with second-level precision is supplemented. False discharge records are excluded, and incomplete ward transfer records are corrected. Simultaneously, the first positive time for each patient is determined to obtain the cleansing results. Based on the cleansing results, a timeline is constructed for each patient, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, to obtain a chronological profile for each patient.

[0262] The key data includes the patient's hospitalization records, department transfer information, nursing operation records, and microbiological test results, including admission and discharge times, ward transfer times, medical staff service records and their specific operation times and patient identifiers, as well as the sample numbers and test results.

[0263] In one embodiment, when the processor executes the computer program to acquire and clean patient hospitalization, nursing, and laboratory data, and construct a timeline profile for each patient, it specifically implements the following steps:

[0264] Patient hospitalization, nursing, and laboratory data from the hospital information system are obtained to acquire key data. This key data is then standardized to ensure that all timestamps are in the same format, and missing time data with second-level precision is supplemented. False discharge records are excluded, and incomplete ward transfer records are corrected. Simultaneously, the first positive time for each patient is determined to obtain the cleansing results. Based on the cleansing results, a timeline is constructed for each patient, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, to obtain a chronological profile for each patient.

[0265] The key data includes the patient's hospitalization records, department transfer information, nursing operation records, and microbiological test results, including admission and discharge times, ward transfer times, medical staff service records and their specific operation times and patient identifiers, as well as the sample numbers and test results.

[0266] In one embodiment, when the processor executes the computer program to dynamically generate a directed propagation chain based on the time information of positive patients and their contacts in the contact event table, the processor specifically implements the following steps:

[0267] Based on the contact event table, all positive cases are sorted according to the time of the first positive test, forming a queue arranged in chronological order; for each positive patient in the queue, all contacts within the infectious period window are extracted, and contacts that meet the criteria are screened out to establish a directed transmission edge to obtain a directed transmission chain.

[0268] In one embodiment, when the processor executes the computer program to perform loop path elimination and confidence grading on the directed propagation chain to obtain a corrected propagation network, the specific steps are as follows:

[0269] Eliminate contradictory paths that violate the time sequence in the directed transmission chain, retaining only unidirectional edges that conform to epidemiological patterns; quantify the reliability of each transmission link based on contact type and the time difference between the last contact and the positive result; remove transmission links that do not meet the criteria, ensuring that all transmission events occur during hospitalization, in order to obtain the corrected transmission network.

[0270] In one embodiment, when the processor executes the computer program to implement the steps of screening cases without upstream sources of infection based on the corrected transmission network and determining the source of infection and its nature by combining admission time and historical hospitalization records, the processor specifically implements the following steps:

[0271] Based on the corrected transmission network, cases without upstream sources of infection are selected as initial sources of infection. The difference between the first positive time and the admission time is calculated for the initial source of infection, and the source of infection is determined by combining historical hospitalization records. The initial source of infection is classified as community infection, current hospitalization infection, or previous hospitalization infection. For cases marked as previous hospitalization infection, their historical hospitalization records are traced until the exact source of infection is found. When a patient shows positive results at different anatomical sites and the detection time at the new site meets specific conditions, the infection type is reassessed and updated.

[0272] The specific conditions include that the time difference between the first positive test and the admission time is no more than 48 hours.

[0273] In one embodiment, after executing the computer program to trace the historical hospitalization records of cases marked as having previous hospitalized infections until the exact source of infection is found, the processor further performs the following steps:

[0274] If the historical hospitalized infection type is community infection, the current source is marked as community infection; if the historical hospitalized infection type is current hospitalized infection, the current source is marked as previous hospitalized infection.

[0275] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0276] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0277] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of each unit is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0278] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the system of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0279] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0280] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for sensing infection transmission paths based on spatiotemporal relationships, characterized in that, include: Acquire and clean patient hospitalization, nursing, and laboratory data to build a timeline profile for each patient; Based on the time-series records, the types of contact with patients in the same ward and with healthcare professionals are defined, and the infectious period and observation window are set to identify potential infection transmission routes and form a contact event table; Based on the time information of positive patients and their contacts sorted by the contact event table, a directed transmission chain is dynamically generated. The directed propagation chain is subjected to loop path elimination and credibility classification to obtain the corrected propagation network; Based on the corrected transmission network, cases without upstream sources of infection were screened out, and the source and nature of the infection were determined by combining the admission time and historical hospitalization records.

2. The infection transmission path perception method based on spatiotemporal relationships according to claim 1, characterized in that, The process of acquiring and cleaning patient hospitalization, nursing, and laboratory data to construct a time-series profile for each patient includes: Obtain patient hospitalization, nursing, and laboratory data from the hospital information system to obtain key data; The key data is standardized to ensure that all timestamp formats are consistent, and missing time data with second-level precision is supplemented. False discharge records are excluded and incomplete ward transfer records are corrected. At the same time, the first positive time of each patient is determined to obtain the cleaning results. Based on the cleaning results, a timeline is constructed for each patient, including their activity trajectory within the hospital, contact with medical staff and service periods, and the time of their first positive test, to obtain a chronological profile for each patient.

3. The infection transmission path perception method based on spatiotemporal relationships according to claim 2, characterized in that, The key data includes the patient's hospitalization records, department transfer information, nursing operation records, and microbiological test results, including admission and discharge times, ward transfer times, medical staff service records and their specific operation times and patient identifiers, as well as sample numbers and test results.

4. The infection transmission path perception method based on spatiotemporal relationships according to claim 1, characterized in that, The term "contact in the same ward" refers to two patients whose stay in the same ward overlaps by at least one hour; the term "contact between healthcare professionals" refers to the same healthcare professional serving different patients within one hour; the window of the infectious period is from 48 hours before to 7 days after the patient's first positive test; the observation window is 7 days after the last contact.

5. The infection transmission path perception method based on spatiotemporal relationships according to claim 1, characterized in that, The step of dynamically generating a directed transmission chain based on the time information of positive patients and their contacts sorted according to the contact event table includes: Based on the contact event table, all positive cases are sorted according to the time of the first positive test, forming a queue arranged in chronological order; For each positive patient in the cohort, all contacts within the infectious period window are extracted, and those who meet the criteria are screened out to establish a directed transmission edge, thus obtaining a directed transmission chain.

6. The infection transmission path perception method based on spatiotemporal relationships according to claim 1, characterized in that, The process of eliminating loop paths and assigning credibility levels to the directed propagation chain to obtain the corrected propagation network includes: Eliminate contradictory paths that violate the time sequence in the directed propagation chain, and retain only unidirectional edges that conform to epidemiological patterns; The reliability of each transmission link is quantified based on factors such as contact type and the time difference between the last contact and a positive result; Eliminate ineligible transmission links and ensure that all transmission events occur during hospitalization to obtain a corrected transmission network.

7. The infection transmission path perception method based on spatiotemporal relationships according to claim 1, characterized in that, The process of screening out cases without upstream sources of infection based on the corrected transmission network, and determining the source and nature of infection by combining admission time and historical hospitalization records, includes: Cases without upstream sources of infection were selected as the initial source of infection based on the corrected transmission network; The difference between the first positive time and the admission time is calculated for the initial source of infection. The source of infection is determined by combining the historical hospitalization records. The initial source of infection is classified as community infection, current hospitalization infection, or previous hospitalization infection. For cases marked as having been hospitalized previously, their historical hospitalization records are traced until the exact source of infection is found; When a patient tests positive at different anatomical sites and the time of testing at the new sites meets specific conditions, their infection type is reassessed and updated.

8. The infection transmission path perception method based on spatiotemporal relationships according to claim 7, characterized in that, The requirement to trace the historical hospitalization records of cases marked as having previous hospitalized infections until the exact source of infection is found also includes: If the historical hospitalized infection type is community infection, the current source is marked as community infection; if the historical hospitalized infection type is current hospitalized infection, the current source is marked as previous hospitalized infection.

9. The infection transmission path perception method based on spatiotemporal relationships according to claim 8, characterized in that, The specific conditions include that the time difference between the first positive test and the admission time is no more than 48 hours.

10. An infection transmission path sensing system based on spatiotemporal relationships, characterized in that, include: The acquisition unit is used to acquire and clean patient hospitalization, nursing, and laboratory data to build a time-series profile for each patient. The contact event table generation unit is used to define the types of contact with patients in the same ward and with medical staff based on the time sequence file, and to set the infectious period and observation window to identify potential infection transmission paths and form a contact event table; The transmission chain generation unit is used to dynamically generate a directed transmission chain based on the time information of positive patients and their contacts sorted in the contact event table. An optimization unit is used to perform loop path elimination and credibility classification on the directed propagation chain to obtain a corrected propagation network. The source tracing unit is used to screen out cases without upstream sources of infection based on the corrected transmission network, and to determine the source of infection and its nature by combining the admission time and historical hospitalization records.

Images