A method and system for individualized analysis of continuous temperature data

Abstract

The application discloses a continuous body temperature data individualized analysis method and system, relates to the technical field of body temperature data analysis, and comprises the following steps: identifying a body temperature curve frequently oscillating in a short time, establishing a time extension sequence of continuous temperature fluctuation and rearranging temperature data; calculating the change amplitude and direction of adjacent intervals based on the time extension sequence, aggregating temperature nodes in the same direction to form a temperature rise segment; calculating the energy difference of the temperature rise segment and determining a latent temperature rise signal according to the energy accumulation rate change; recombining the distribution section of the latent temperature rise signal to form a temperature change sequence; establishing a nursing response logic and dynamically adjusting cooling, fluid replacement and body surface heat dissipation measures according to the temperature rise signal intensity and time distribution. The application reconstructs the body temperature data through time extension, extracts a deep temperature rise trend hidden by fluctuation, establishes a nursing response logic based on the temperature rise intensity and time distribution, dynamically adjusts the cooling and fluid replacement measures, and realizes precise linkage of body temperature monitoring and nursing intervention.

Description

Technical Field

[0001] This invention relates to the field of body temperature data analysis technology, specifically to a method and system for individualized analysis of continuous body temperature data. Background Technology

[0002] Personalized analysis of continuous body temperature data refers to establishing a continuous time series model reflecting the individual's temperature fluctuation characteristics based on long-term dynamic collection of individual body temperature change data, combined with their physiological rhythms, activity patterns, and environmental influencing factors. This model identifies abnormal temperature trends and subtle deviations. By comparing an individual's historical baseline rather than the population average, and analyzing indicators such as circadian rhythm stability, frequency of micro-fluctuations, and recovery delay, the system can accurately determine metabolic status, immune response, and potential disease signs. Based on this analysis, the system can generate differentiated health care methods for different individuals. For example, it can develop timed temperature control and sleep synchronization intervention strategies for those with circadian rhythm disorders, provide temperature recovery rate-guided rehabilitation care plans for postoperative recovery patients, or set dynamic temperature difference threshold-triggered diet and hydration plans for those with frequent temperature fluctuations, achieving end-to-end personalized health management including monitoring, identification, and care.

[0003] The existing technology has the following shortcomings:

[0004] When body temperature curves fluctuate frequently within a short period, the large fluctuations in surface temperature can mask deeper temperature rise signals, leading to distorted identification of temperature trends. At this time, an individual may be in the early stages of infectious fever, but the high-frequency fluctuations interfere with the judgment of continuous trends, causing the body temperature to be mistakenly perceived as still within the normal range, thus failing to trigger timely and targeted nursing interventions. As abnormal fever continues to be ignored, the inflammatory response may further intensify, easily causing a sudden rise in body temperature, increased metabolic load, and delayed nursing intervention, ultimately affecting the timeliness and accuracy of early care.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for individualized analysis of continuous body temperature data to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for individualized analysis of continuous body temperature data, comprising the following steps:

[0008] Identify body temperature curves that oscillate frequently within a short period of time. For the identified body temperature curves, establish a time-extended sequence of continuous temperature fluctuations. Rearrange the collected temperature change data within the extended time window to characterize the slow warming trend masked by surface fluctuations in the time-extended sequence.

[0009] Based on the established time-extended sequence, the change amplitude and direction of adjacent intervals are calculated for each temperature node in the time-extended sequence, and continuous temperature nodes with the same change direction are aggregated to form a heating segment, so as to enhance the identification of deep temperature rise characteristics in the time-extended temperature sequence.

[0010] Based on the obtained warming segments, the energy difference between the start and end points of each warming segment is calculated. The latent warming signal is determined according to the change in the energy accumulation rate between adjacent warming segments, and the risk trend of infectious fever is judged according to the change trend of the energy accumulation rate.

[0011] After identifying the latent warming signal, the distribution segments of the latent warming signal on the time axis are recombined to obtain the recombined temperature change sequence. A micro-delay strategy is introduced between adjacent distribution segments to compress the response interval, thereby restoring the continuity of the body temperature change process at the time series level.

[0012] Based on the recombined temperature change sequence, a nursing response logic is established for the identified latent fever signal. The execution order of cooling, fluid replacement and heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signal, so as to achieve precise nursing intervention in the early stage of infectious fever.

[0013] Preferably, establishing a time-extended sequence of continuous temperature fluctuations includes the following steps:

[0014] During the continuous collection of body temperature data from individuals, the collected temperature change data are arranged sequentially according to the sampling time to form the original body temperature curve. The rate of temperature change between each sampling point is continuously calculated and analyzed point by point. When the temperature change direction of several consecutive sampling points alternates between positive and negative within a preset time window and the alternation frequency exceeds a preset time threshold, the corresponding time segment is marked as a frequent oscillation interval, and high fluctuation segment is generated by merging them according to the continuity of time.

[0015] Based on the calibrated high fluctuation range, the time span of the high fluctuation range is proportionally enlarged on the time axis, the original sampling time interval is extended by a fixed multiple, and the original temperature sampling point positions are redistributed in the extended time window, keeping the temperature value unchanged and forming a time extension sequence that covers the entire fluctuation process.

[0016] The obtained time-extended sequence is rearranged, and the temperature nodes are adjusted sequentially according to the original sampling order. The time interval is evenly divided and the temperature nodes are redistributed within the extended time window to keep the body temperature change curve uniform and continuous on the time axis.

[0017] Trend characterization is performed on the rearranged time-extended sequence to identify the upward trend of temperature over continuous time periods and to characterize the slow warming process masked by surface fluctuations.

[0018] Preferably, aggregating consecutive temperature nodes with the same direction of change to form a heating segment includes the following steps:

[0019] After the time-extended sequence is established, the relationship between each temperature node in the time-extended sequence and its adjacent temperature nodes before and after it is continuously calculated to obtain the magnitude and direction of temperature change, and the start time, end time and direction of change of each interval are recorded on the time axis.

[0020] After identifying the direction of change in adjacent intervals, continuous temperature nodes with the same direction of change are continuously aggregated. Starting from the beginning node of the time extension sequence, the direction of change information is scanned point by point along the time axis. Temperature nodes with the same direction of change and continuous distribution are included in the same aggregate set, and the time span and boundary of each aggregate set are recorded.

[0021] Based on the aggregation of temperature nodes in the same direction, the resulting direction set is filtered and its boundaries are established. A continuous set with an upward changing direction is selected as a candidate set of heating segments. Heating segments are formed based on the candidate set of heating segments, and the start and end points of the heating segments are determined and the time position is recorded. When the time interval between adjacent heating segments is less than a set time interval threshold, they are connected sequentially to form a continuous heating interval.

[0022] Preferably, when connecting adjacent heating segments, the interval between the end time and the start time of the adjacent heating segments is compared. When the interval is less than a set time interval threshold, the temperature node data of the adjacent heating segments are merged in chronological order, and the start time, end time and time span of the new heating segment are recorded in a unified manner to form a heating interval with continuous time distribution in the time extension sequence.

[0023] Preferably, determining the latent warming signal based on the change in energy accumulation rate between adjacent warming segments includes the following steps:

[0024] Based on the obtained heating segments, the energy difference between the start and end of each heating segment is calculated. The start temperature value of the heating segment is taken as the initial temperature and the end temperature value is taken as the final temperature. The temperature nodes between the start and end are extracted and arranged in chronological order. The change amplitude of adjacent temperature nodes is recorded and accumulated to obtain the energy difference value corresponding to each heating segment.

[0025] After obtaining the energy difference of each heating segment, the energy accumulation rate is determined according to the time span of the heating segment, the start time and end time of the heating segment are extracted, the time interval is calculated and the energy difference is mapped to the time interval to determine the energy change rate per unit time, forming a time-continuous energy accumulation rate sequence.

[0026] Based on the energy accumulation rate sequence, the temporally adjacent heating segments are compared one by one. When the energy accumulation rate of the later heating segment is greater than that of the earlier heating segment, it is determined that there is a latent heating signal in the corresponding time interval, and the time interval corresponding to the continuously increasing energy accumulation rate is marked as the time distribution band of the latent heating signal.

[0027] After identifying the latent fever signal, the risk trend of infectious fever is judged based on the changing trend of the energy accumulation rate. The continuous change of the energy accumulation rate within the extended time range is analyzed. When the energy accumulation rate continues to increase and the latent fever signal is continuously distributed, it is judged that the risk of infectious fever is forming or is in the aggravation stage.

[0028] Preferably, in the process of determining the latent warming signal based on the change in energy accumulation rate between adjacent warming segments, when the energy accumulation rate of three or more consecutive warming segments increases sequentially and the change in energy accumulation rate exceeds a preset amplitude threshold, the corresponding continuous time interval is determined as the continuous distribution interval of the latent warming signal, and the risk of infectious fever is preferentially judged to be in the formation stage within the continuous distribution interval.

[0029] Preferably, recombining the distribution segments of the latent temperature rise signal on the time axis includes the following steps:

[0030] After the latent temperature rise signal is identified, the distribution segment of the latent temperature rise signal on the time axis is accurately extracted. The start and end points of each latent temperature rise signal are marked in chronological order. All temperature nodes and their time labels within the segment are extracted and the duration, start temperature, end temperature and temperature change range are recorded to establish a segment set containing time and temperature information.

[0031] Based on the extraction of the distribution segments of the latent temperature rise signal, the time distribution order of each segment is recombined and sorted from early to late according to the start time. The end time and start time of adjacent segments are compared. When the time interval is greater than the interval duration threshold, the start time of the next segment is adjusted to shorten the time interval, while keeping the temperature node arrangement within the segment unchanged, so as to achieve continuous splicing between segments.

[0032] Based on the time sequence recombination, a micro-delay strategy is introduced between adjacent latent temperature rise signal segments. By determining the time difference between the end time of the previous segment and the start time of the next segment, the starting point of the next segment is shifted forward on the time axis and the temperature node time label is updated, thus compressing the response interval and forming a continuously arranged time structure.

[0033] Based on the recombination and slight delay adjustment of the latent warming signal segments, all latent warming signal segments are integrated to form a recombined temperature change sequence, so that the temperature nodes are continuously distributed on the time axis and reflect the transition process of individual body temperature from the latent warming stage to the sustained warming stage.

[0034] Preferably, when introducing a micro-delay strategy between adjacent latent temperature rise signal segments, the magnitude of the delay adjustment is determined based on the time difference between the end time of the previous segment and the start time of the next segment. When the time difference is less than a preset delay threshold, the segment time position remains unchanged. When the time difference is greater than the preset delay threshold, the start point of the next segment is shifted forward, and the adjusted temperature node time label is updated synchronously to ensure the continuous connection of the time series.

[0035] Preferably, establishing nursing response logic for the identified latent fever signal includes the following steps:

[0036] After obtaining the recombined temperature change sequence, a nursing response logic for the latent temperature rise signal is established. The time location, duration, temperature rise rate, maximum temperature and time interval of the latent temperature rise signal are extracted. The start time and end time are marked on the time axis. Based on the characteristics of continuity and intensity change, it is divided into three nursing response intervals: the initial temperature rise stage, the middle temperature rise stage and the continuous temperature rise stage. A nursing response logic table is established according to the temperature change characteristics of different stages.

[0037] Based on the establishment of the nursing response logic, the execution order of cooling, fluid replacement and heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signal. The measures in the nursing response logic are matched with the time information of the temperature change sequence. The intensity of the latent fever signal is used as the priority basis and the duration is controlled by the time distribution, forming a nursing action execution order that matches the stage of body temperature change.

[0038] Based on the dynamic adjustment of nursing response logic and the execution sequence of nursing actions, precise nursing intervention based on latent fever signals is implemented. The execution time of nursing actions is matched with the time interval of latent fever signals. In the initial fever stage, heat dissipation measures are initiated on the body surface. In the middle fever stage, fluid resuscitation and heat dissipation on the body surface are combined. In the continuous fever stage, the frequency of cooling measures is increased while fluid resuscitation and heat dissipation on the body surface are maintained, forming a nursing intervention behavior that is synchronized with the process of body temperature change.

[0039] A system for individualized analysis of continuous body temperature data includes a time-extended sequence establishment module, a temperature rise segment aggregation module, a latent temperature rise signal identification module, a temperature sequence reconstruction module, and a nursing response logic establishment module.

[0040] Time-extended sequence establishment module: Identifies body temperature curves that oscillate frequently within a short period of time, establishes a time-extended sequence of continuous temperature fluctuations for the identified body temperature curves, and rearranges the collected temperature change data within the extended time window;

[0041] Heating Segment Aggregation Module: Based on the established time-extended sequence, calculate the change amplitude and direction of adjacent intervals for each temperature node in the time-extended sequence, and aggregate consecutive temperature nodes with the same change direction to form a heating segment;

[0042] Latent heating signal identification module: Based on the obtained heating segments, calculate the energy difference between the start and end points of each heating segment, and determine the latent heating signal according to the change in energy accumulation rate between adjacent heating segments;

[0043] Temperature sequence reconstruction module: After determining the latent temperature rise signal, the distribution segment of the latent temperature rise signal on the time axis is recombined to obtain the recombined temperature change sequence.

[0044] Nursing response logic establishment module: Based on the recombined temperature change sequence, nursing response logic is established for the identified latent fever signal, and the execution order of cooling, fluid resuscitation and body surface heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signal.

[0045] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0046] This invention restores the continuity of the body temperature curve in the time dimension by establishing a time-extended sequence of continuous temperature fluctuations and rearranging the temperature change data. This allows for the extraction of masked deep-seated warming trends from frequently oscillating fluctuation curves over a short period. This method fully presents the true evolution of body temperature changes, effectively eliminating the interference of surface temperature fluctuations on trend identification, and enabling the early identification of latent warming signals in the early stages of infectious fever.

[0047] This invention establishes a nursing response logic after identifying latent fever signals and dynamically adjusts the execution sequence of cooling, fluid replacement, and heat dissipation measures based on the intensity and time distribution of the fever signals. This allows nursing interventions to remain synchronized with the body temperature change process, thereby achieving dynamic cyclical control from body temperature monitoring to nursing intervention. This enables nursing measures to respond flexibly to individual body temperature change characteristics, effectively improving the accuracy and continuity of early fever care, and ultimately enhancing the efficiency of body temperature regulation and the quality of nursing care. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0049] Figure 1 This is a flowchart of a method for individualized analysis of continuous body temperature data according to the present invention;

[0050] Figure 2 A flowchart for establishing a time-extended sequence of continuous temperature fluctuations in this invention;

[0051] Figure 3 The present invention determines the latent heating signal based on the change in energy accumulation rate between adjacent heating segments;

[0052] Figure 4 This is a schematic diagram of a module of a continuous body temperature data individualized analysis system according to the present invention. Detailed Implementation

[0053] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0054] This invention provides, for example Figures 1 to 3 The method for individualized analysis of continuous body temperature data, as shown, includes the following steps:

[0055] Step 1: Identify the body temperature curve that fluctuates frequently in a short period of time. For the identified body temperature curve, establish a time-extended sequence of continuous temperature fluctuations. Rearrange the collected temperature change data in the extended time window to characterize the slow warming trend that is masked by surface fluctuations in the time-extended sequence.

[0056] The specific steps for establishing a time-extended sequence of continuous temperature fluctuations are as follows:

[0057] During continuous body temperature monitoring, the collected temperature change data are arranged sequentially according to sampling time to form a continuous raw body temperature curve. To identify frequent oscillations within a short period, the rate of temperature change between each sampling point in the raw body temperature curve is continuously calculated, and the temperature difference between adjacent sampling points is analyzed point by point. When the direction of temperature change between several consecutive sampling points continuously alternates between positive and negative within a preset time window, and the alternation frequency exceeds a preset time threshold, this time interval is marked as a short-term frequent oscillation interval. Subsequently, adjacent oscillation intervals are merged according to time continuity to generate a complete high-fluctuation segment. Through this continuous comparison method, the frequently oscillating parts of an individual's body temperature curve can be accurately identified in the time series, thus providing a precise input segment for subsequent time-extended processing.

[0058] It should be noted that:

[0059] The time threshold is determined based on the sampling frequency of individual body temperature data and the normal physiological fluctuation cycle. Specifically, the number of sampling points per unit time can be calculated first based on the body temperature sampling time interval. Then, combined with the natural fluctuation cycle range of an individual's body temperature in a resting state, the time length that can cover a complete micro-fluctuation process is selected as the basic threshold range. At the same time, the basic threshold is appropriately reduced to ensure that only high-frequency alternating changes are identified. This allows temperature fluctuations with short duration and frequent alternation in direction to be defined as short-term oscillation ranges, and avoids misjudging diurnal rhythms or slow warming processes as oscillation signals.

[0060] After identifying body temperature curves that oscillate frequently within a short period, a time-extended sequence of continuous temperature fluctuations is constructed based on the identified high-fluctuation segments. This process involves proportionally amplifying the time span of the high-fluctuation segments on the time axis, thus redistributing the position of each sampling node on the time axis. Specifically, the original sampling time interval is extended by a fixed multiple, and the positions of the original temperature sampling points are redistributed within the extended time interval, keeping the temperature values ​​constant but increasing the time distance between sampling points. Through this time extension method, fluctuations that originally occurred densely within a short period are stretched to a longer time interval, allowing the overall trend of body temperature changes to be presented continuously on the new time scale. Simultaneously, the length of the extended time window is dynamically set according to the individual's body temperature fluctuation cycle to ensure that the extended time series covers the complete change process from the beginning to the end of the fluctuation, thereby forming a time-extended sequence with temporal continuity.

[0061] It should be noted that:

[0062] A fixed multiplier refers to a time expansion factor that amplifies the original sampling time interval by a uniform ratio. This multiplier is set based on the sampling frequency of individual body temperature data and the duration of high fluctuation ranges. Specifically, it can be determined by comparing the time density of adjacent sampling points within the high fluctuation range with the overall body temperature change cycle. When the sampling points are densely distributed within a unit of time, a multiplier greater than 1 is selected to amplify the time interval proportionally, so that the spacing between adjacent sampling points on the time axis increases synchronously. This expands the originally concentrated fluctuation data in a short period of time in the time dimension, ensuring that the temperature values ​​before and after the extension remain consistent and the time structure is expanded by a uniform ratio, so that the extended time series can more clearly present the continuous change trend.

[0063] After obtaining the time-extended sequence, the extended temperature change data is rearranged to restore the temporal continuity and uniformity of the body temperature curve. During the rearrangement, the extended temperature nodes are first arranged sequentially according to the original sampling time order. Then, within the extended time window, the relative position of each temperature node on the time axis is readjusted based on the extended sampling interval, ensuring that the time intervals between nodes remain consistent. For cases where there are uneven sampling intervals or missing data in the original body temperature curve, the time intervals are redistributed to uniform interval positions, creating a continuous fluctuating curve of body temperature changes on the time axis. This rearrangement process transforms the originally dense, highly fluctuating segments into a uniformly distributed temperature change curve on the time-extended sequence, eliminating trend breaks caused by uneven sampling density, thus maintaining the overall smoothness and continuity of the body temperature curve within the extended time range.

[0064] After rearrangement, the extended temperature change curves are characterized to reveal the slow warming trend masked by surface fluctuations in the extended time series. During this process, by observing the overall change pattern of the rearranged body temperature curves on the time axis, the trend of continuous temperature increase over a continuous period is identified. Due to the introduction of the extended time window, the temperature curves that previously fluctuated within a short period are stretched to a longer time range, and short-term temperature fluctuations are evenly distributed across a wider time interval, thus showing a gradual increase in body temperature over time in the extended time series. By performing trend analysis on the rearranged extended time series, the continuous temperature rise phase can be identified on the extended body temperature curves, accurately reflecting the continuous process of deep body temperature changes over time. This extended time series of continuous temperature fluctuations not only restores the continuity of body temperature changes in the time dimension but also preserves the true characteristics of individual body temperature fluctuations in the data structure, allowing the slow warming trend masked by surface fluctuations to be clearly presented in the extended time series.

[0065] Through the aforementioned sequential steps, the entire process—from identifying frequently oscillating body temperature curves over a short period to constructing and rearranging a continuous temperature fluctuation time-extended sequence—is completed. This method extends the temporal structure of the original body temperature curve, transforming short-term fluctuation data into an observable, continuous temperature rise over the extended timescale. In this process, the establishment of the time-extended sequence not only ensures the continuity and uniformity of the body temperature data over time but also effectively restores the deeper temperature rise trend masked by frequent oscillations, accurately reflecting the individual's temperature change patterns in the extended time dimension. Through this specific implementation of time extension, the early identification of deeper temperature rise trends can be achieved in continuous individual temperature monitoring, providing reliable evidence of body temperature changes for the early detection of infectious fever and subsequent nursing interventions.

[0066] Step 2: Based on the established time-extended sequence, calculate the change amplitude and direction of adjacent intervals for each temperature node in the time-extended sequence, and aggregate continuous temperature nodes with the same change direction to form a heating segment, so as to enhance the identification of deep temperature rise characteristics in the time-extended temperature sequence.

[0067] The specific steps for aggregating consecutive temperature nodes with the same direction of change to form a heating segment are as follows:

[0068] After the time-extended sequence is established, the relationship between each temperature node in the time-extended sequence and its adjacent temperature nodes is continuously calculated to obtain the amplitude and direction of temperature change. In this process, all temperature nodes in the time-extended sequence are arranged in chronological order, and the temperature data of adjacent temperature nodes are extracted sequentially. The temperature difference between any two consecutive temperature nodes is determined. When the temperature value of the subsequent node is greater than that of the preceding node, the interval is recorded as an increasing interval; when the temperature value of the subsequent node is less than that of the preceding node, the interval is recorded as a decreasing interval; when the two temperature values ​​are the same, the interval is recorded as a stable interval. To ensure the continuity of each change interval in the time-extended sequence, the time axis order is maintained during the calculation process. The start and end times of each interval are marked, and the interval direction information is stored sequentially on the time axis. In this way, a set of change direction markers arranged chronologically is formed in the time-extended sequence, reflecting the continuous directional relationship of temperature change over time throughout the entire body temperature change process.

[0069] After identifying the direction of change for each adjacent interval in the extended time sequence, consecutive temperature nodes with the same direction of change are continuously aggregated to form a set of temperature changes with consistent direction. During aggregation, starting from the beginning node of the extended time sequence, the direction of change information is scanned point by point along the time axis. When multiple adjacent intervals are detected to have the same direction identification and be continuously distributed, the temperature nodes corresponding to these intervals are included in the same aggregation set. For temperature nodes continuously in an upward direction, they are integrated into a set of increasing temperature directions; for temperature nodes continuously in a downward direction, they are integrated into a set of decreasing temperature directions; and for temperature nodes in a stable direction, they are retained separately for subsequent reference. During this process, the time span of each direction set is recorded, using the start and end times of consecutive nodes as the time boundary of the aggregation range. Simultaneously, the values ​​of all temperature nodes participating in the aggregation are stored in the aggregation unit in chronological order. To ensure the integrity of the aggregation results, the time intervals between aggregation intervals are assessed for connectivity. When the time interval between adjacent aggregation sets is less than the average interval of the extended time window, the two adjacent aggregation sets are merged into a larger set of consecutive directions to ensure the continuous distribution of the aggregation results on the time axis. Through this continuous aggregation process, the originally scattered temperature nodes in the time-extended sequence are reorganized into several continuous temperature sets with unified directional attributes, so that the trend of body temperature change forms a continuous directional segment in the time structure.

[0070] After aggregating temperature nodes in the same direction, the resulting direction sets are filtered and their boundaries are established to form heating segments in the time-extended temperature sequence and enhance the identification of deep temperature rise characteristics. Specifically, continuous sets with an upward changing direction are selected from all direction sets as candidate sets for heating segments. In each candidate set, the first temperature node of the set is determined as the starting point of the heating segment, and the last temperature node of the set is determined as the ending point of the heating segment. The time positions of the starting and ending points in the time-extended sequence are recorded. By recording the time span between the starting and ending points, the duration of the heating segment on the time axis can be obtained. For the relationship between multiple heating segments, the continuity of their time distribution is further judged. When the interval between the ending time of one heating segment and the starting time of the next heating segment in the time-extended sequence is less than a set time interval threshold, the two heating segments are connected sequentially, so that they form a longer continuous heating interval in the time-extended sequence. During the connection process, the time order of each segment is maintained, and the connected temperature node data is merged into a new set of heating segments. In the final warming segments, each segment consists of temperature nodes that are continuous in time and aligned in direction, with a stable upward direction and definite time boundaries. Through this process, the upward trends that were originally scattered in the time-extended sequence are integrated into warming segments with temporal coherence. These warming segments exhibit a continuous upward characteristic in the time-extended temperature sequence, accurately reflecting the process of the individual's body temperature rising at a deeper level.

[0071] It should be noted that:

[0072] The time interval threshold is determined based on the average time interval between adjacent temperature nodes in the time-extended sequence and the actual distribution characteristics of the heating segments. Specifically, the time interval between all adjacent nodes in the time-extended sequence can be statistically analyzed and its average value can be calculated. This average time interval is used as a basic reference. Then, combined with the actual interval distribution between heating segments, the reference value is amplified by a certain proportion to obtain the time interval threshold used to judge the continuity of segments. When the time interval between adjacent heating segments is within the range of this threshold, it can be determined as a segmented performance in the same continuous heating process, and thus be processed sequentially. At the same time, it avoids the mistaken merging of independent heating processes with large time intervals.

[0073] Through the above steps, the entire process of extracting the relationship between adjacent temperature nodes from the time-extended sequence, calculating the amplitude and direction of changes in adjacent intervals, and aggregating consecutive temperature nodes in the same direction to form a warming segment is completed. Throughout the process, the time-extended sequence is maintained, ensuring the continuity of the temperature data structure and the consistency of its direction in the time dimension. By analyzing the amplitude and direction of changes, aggregating nodes in the same direction, and forming warming segments, the identification of deep temperature rise characteristics in the time-extended temperature sequence can be strengthened, allowing the continuous warming trend, which was originally masked by high-frequency oscillations, to be re-expressed in the time structure. The resulting warming segment provides a data foundation for subsequent energy change calculations and the capture of latent warming signals.

[0074] Step 3: Based on the obtained warming segments, calculate the energy difference between the start and end points of each warming segment, determine the latent warming signal based on the change in energy accumulation rate between adjacent warming segments, and judge the risk trend of infectious fever based on the trend of energy accumulation rate.

[0075] The specific steps for determining the latent heating signal based on the change in energy accumulation rate between adjacent heating segments are as follows:

[0076] Based on the obtained heating segments, the energy difference between the start and end points of each segment is calculated. In this process, the starting temperature of the heating segment is used as the initial temperature, and the ending temperature as the final temperature. All temperature nodes between them are extracted and arranged in chronological order to form a continuous temperature change path. Then, the time position of each temperature node in the time-extended sequence is recorded, the temperature change amplitude between adjacent temperature nodes is calculated, and these amplitudes are accumulated sequentially on the time axis. Through this accumulation method, the overall temperature change range of the heating segment from the start to the end point can be determined, serving as a direct representation of energy change. During the calculation, the chronological order is maintained to ensure that the change of each temperature node is fully incorporated into the time-extended structure, thus ensuring that the energy difference calculation result includes both the total temperature change and the continuity of temperature change over time. Ultimately, each heating segment corresponds to an independent energy difference value, which represents the degree of heat accumulation within that time interval, providing a concrete quantitative basis for subsequent calculations of the energy accumulation rate.

[0077] After obtaining the energy difference for each warming segment, the energy accumulation rate is determined based on the time span of the warming segment. This process begins by extracting the start and end times of the warming segment and calculating the time interval between them as the duration of the warming segment. Then, the energy difference of the warming segments is correlated with the time interval, and the rate of energy change per unit time within that time period is determined by processing the proportional relationship between the two. Each warming segment corresponds to an energy accumulation rate, which reflects the rate of increase in body temperature and the intensity of heat accumulation within the body during that time period. To maintain the temporal continuity of the energy accumulation rate, the energy accumulation rates of all warming segments are arranged chronologically to form an energy accumulation rate sequence. This energy accumulation rate sequence is continuously distributed along the time axis, allowing for longitudinal comparison of the energy accumulation characteristics of different warming segments. When the energy accumulation rate between adjacent warming segments shows a continuous increase over time, it indicates that the heat accumulation process within the body is intensifying, and the warming trend is in a continuously advancing stage within the time-stretched sequence. Through this process, the energy accumulation rate transforms the numerical change in body temperature into heat growth information within a time-stretched structure, laying a continuous temporal foundation for identifying latent warming signals.

[0078] After establishing the energy accumulation rate sequence, latent warming signals are determined based on the changes in energy accumulation rates between adjacent warming segments. In this process, continuous warming segments are compared pairwise using the time-stretched sequence as a benchmark. For two temporally adjacent warming segments, the energy accumulation rates of the preceding and following segments are extracted. When the energy accumulation rate of the following segment is greater than that of the preceding segment, it indicates that the rate of body temperature rise increases over time, the process of heat accumulation in the body accelerates, and a latent warming signal exists in that interval. When identifying latent warming signals, the start and end times of consecutive warming segments are correlated to ensure that the comparison of energy accumulation rates is conducted within a continuous time period. If the energy accumulation rates of three or more consecutive warming segments show a continuously increasing characteristic, it indicates that the individual's body temperature continues to accumulate heat in multiple adjacent time intervals, and the latent warming signal exists continuously in the time dimension. These continuously existing latent warming signals are marked in chronological order to form a temporal distribution band of latent warming signals, which reflects the process of gradual heat accumulation at deeper levels of the individual's body. In this way, the latent warming signal is extracted from the continuous changes in the rate of energy accumulation, providing a time- and energy-based reference for further judging the trend of infectious fever risk.

[0079] After identifying the latent fever signal, the risk trend of infectious fever is judged based on the changing trend of the energy accumulation rate. In this process, the time interval of the latent fever signal is combined with the energy accumulation rate sequence to analyze the trajectory of the energy accumulation rate throughout the entire extended time range. When the energy accumulation rate shows a continuous upward trend over a continuous period, and the latent fever signal is continuously distributed along the time axis, it indicates that the heat accumulation process in the body is in an enhanced phase, and the body temperature change trend is developing towards a continuous rise; at this point, it is judged that the risk of infectious fever is forming. When the energy accumulation rate remains high in multiple adjacent warming segments without a decreasing phase, it indicates that the body temperature rise process has lasted for some time, and the individual's heat release mechanism has not returned to equilibrium; it is judged that the risk of infectious fever is in an aggravated phase. This judgment method based on the changing trend of the energy accumulation rate allows for a dual analysis of body temperature changes from both temporal continuity and energy accumulation intensity. In the temporal dimension, the continuous increase in the energy accumulation rate represents the extension of the warming process; in the energy dimension, the continuous distribution of the latent fever signal represents the persistence of heat accumulation. By combining the two, we can construct the evolutionary path of infectious fever risk trends, thereby identifying latent fever trends in advance before body temperature rises significantly.

[0080] It should be noted that:

[0081] The higher value is defined as a range in which the energy accumulation rate continuously exceeds twice the average energy accumulation rate under steady-state conditions.

[0082] In determining the latent warming signal based on the change in energy accumulation rate between adjacent warming segments, the warming segments are arranged continuously in chronological order, and the energy accumulation rate and its change amplitude corresponding to each warming segment are extracted one by one. When the energy accumulation rate of three or more consecutive warming segments shows a progressively increasing relationship, the difference in energy accumulation rate between adjacent warming segments is calculated. When the increment of energy accumulation rate between each adjacent segment is greater than the preset amplitude threshold, the time range corresponding to the consecutive warming segments is determined as the continuous distribution interval of the latent warming signal, and the risk of infectious fever is preferentially determined to be in the formation stage within this continuous distribution interval.

[0083] In setting the amplitude threshold, the individual's historical body temperature data is used as a reference baseline. The fluctuation range of energy accumulation rate between adjacent time intervals under normal physiological conditions is statistically analyzed, and the upper limit of this fluctuation range is selected as the basic reference value. Combined with the overall fluctuation level in the current time extension sequence, this reference value is proportionally amplified to obtain the amplitude threshold used to distinguish between normal fluctuations and continuous warming trends. This amplitude threshold can reflect the inherent characteristics of individual body temperature changes and effectively distinguish abnormal energy accumulation changes during continuous warming, thereby ensuring the stability and specificity of latent warming signal identification.

[0084] Through the above process, focusing on the temperature rise segment and using a time-extended sequence as the temporal basis, the analysis of body temperature changes is expanded from simple numerical fluctuations to the level of energy accumulation pattern analysis, achieving a fusion of temporal and thermal characteristics. In this implementation, the calculation of energy difference reflects the degree of heat accumulation within the temperature rise segment, the determination of the energy accumulation rate reveals the rate of change in body temperature rise, the identification of latent temperature rise signals characterizes the continuity of deep heat increase, and the judgment of the trend of energy accumulation rate changes enables early assessment of the risk of infectious fever. Through cross-analysis of time and energy dimensions, the deep temperature rise trend of an individual and the prediction of risk trends can be achieved during continuous body temperature monitoring, providing a basis for nursing response based on body temperature changes.

[0085] Step 4: After determining the latent warming signal, the distribution segments of the latent warming signal on the time axis are recombined to obtain the recombined temperature change sequence. A micro-delay strategy is introduced between adjacent distribution segments to compress the response interval, thereby restoring the continuity of the body temperature change process at the time series level.

[0086] After identifying the latent temperature rise signal, the specific steps for recombinating the distribution segments of the latent temperature rise signal on the time axis are as follows:

[0087] After the latent temperature rise signal is identified, its distribution segments on the time axis are precisely extracted. Specifically, in the time-extended sequence, the start and end points of each latent temperature rise signal are marked chronologically, clearly defining the time boundaries of each latent temperature rise signal interval. For each latent temperature rise signal segment, all temperature nodes and their time labels within that segment are extracted completely, maintaining consistency in time order during extraction to ensure that the temperature change structure within the segment matches the original time series. Simultaneously, the duration, starting temperature, ending temperature, and temperature change amplitude of each latent temperature rise signal segment are recorded, establishing a set of segments containing both time and temperature information. In this set, each segment represents a complete latent temperature rise process, including the specific manifestation of a continuous temperature increase over time. Through this extraction method, latent temperature rise signals originally scattered at different locations in the time-extended sequence are integrated into time-distributed units with clearly defined boundaries.

[0088] After extracting the latent warming signal segments, their distribution order on the time axis is recombined to form a new time series structure. In this process, the segments are first sorted from morning to night according to their start time to ensure the recombined time sequence matches the actual rhythm of individual body temperature changes. Next, the end time of each segment is compared with the start time of the next segment to calculate the time interval. When a time interval between adjacent segments is found to be greater than a threshold, the start time of the next segment is adjusted to shorten the time interval, making the time connection between segments tighter. To maintain the natural continuity of the body temperature change process, the arrangement of temperature nodes within a segment is not changed during the adjustment process; only the overall time position of the segment is moved, thus achieving continuous splicing between segments. After this recombination step, the previously discontinuous latent warming signals on the time axis are rearranged into a sequential time structure, transforming the time distribution of body temperature changes from a scattered state to a continuous extended state, forming a temporally ordered and interconnected latent warming signal sequence.

[0089] It should be noted that:

[0090] The interval duration threshold is determined based on the actual distribution density of latent warming signal segments in the time-extended sequence and the statistical results of the time intervals between adjacent segments. Specifically, the time intervals between all latent warming signal segments can be statistically analyzed first, and their average interval value can be calculated as a basic reference. Combined with the continuous characteristics of body temperature changes, the average interval can be appropriately amplified to obtain the interval duration threshold used to determine whether a segment belongs to the same continuous warming process. When the time interval between adjacent segments is less than the threshold, it is considered as part of a continuous change process, while when the time interval exceeds the threshold, it is considered as an independent change segment. This ensures temporal continuity while avoiding incorrect splicing between different warming processes.

[0091] After reordering the time sequence, a micro-delay strategy is introduced between adjacent latent temperature rise signal segments to compress the response interval, thereby enhancing the continuity of the time series. Specifically, for two adjacent latent temperature rise signal segments, the time difference between the end time of the preceding segment and the start time of the following segment is determined. When the time difference is less than a preset delay threshold, the segment's time position remains unchanged; when the time difference is greater than the preset delay threshold, a micro-delay adjustment is introduced within this time difference. When introducing the micro-delay, the start point of the following segment is slightly shifted forward on the time axis, making its time position closer to the end point of the preceding segment, thus compressing the time interval between them in physical time. To ensure the consistency of the adjusted time structure, the adjusted temperature node time labels are updated to maintain a continuous arrangement on the new time axis. In this process, the time distribution of all latent temperature rise signal segments is uniformly adjusted, making the time connection between adjacent segments closer and avoiding discontinuities in body temperature changes caused by differences in sampling intervals or uneven measurement times. After all segments have completed the slight delay adjustment, the latent warming signal presents a continuous arrangement on the time axis. The response interval of body temperature change is fully compressed, and the originally scattered warming process is integrated into a coherent time evolution process, forming a smoothly connected temperature rise trajectory.

[0092] It should be noted that:

[0093] The preset delay threshold is determined based on the time interval distribution between adjacent latent temperature rise signal segments in the time-extended sequence. Specifically, the time difference between all adjacent segments can be statistically analyzed and its average or median value can be calculated as a basic reference. Then, combined with the continuous characteristics of body temperature change, the reference value is reduced by a certain proportion to obtain the threshold used to determine whether time compression processing should be performed. When the time difference between adjacent segments exceeds the threshold, it is determined that there is a large response interval, thereby triggering a micro-delay adjustment. When the time difference is within the threshold range, the original time structure remains unchanged to avoid unnecessary disturbance to the original continuous change process.

[0094] When making a slight forward shift, the end time of the previous segment is used as a reference, and the start time of the next segment is adjusted forward according to a preset shift ratio. This creates a small and continuous time interval between the adjusted start time and the end time of the previous segment, while keeping the time interval between each temperature node within the next segment unchanged. In other words, the time position of the segment is shifted forward synchronously as a whole, thereby compressing the time gap between segments without changing the temperature change structure, so that adjacent segments can be continuously connected on the time axis.

[0095] After recombination and slight delay adjustment of the time-extended sequence, a new temperature change sequence is obtained, restoring the continuity of the body temperature change process at the time series level. In this process, all latent warming signal segments after time recombination and delay adjustment are integrated into a complete temperature change curve. The new temperature change curve maintains a strict sequential relationship on the time axis, with the time position of each temperature node being redistributed according to the extended structure, ensuring the continuous extension of the temperature change process in time. With the continuous splicing of latent warming signals, the body temperature change curve exhibits a continuously rising shape in the time-extended sequence, and the time process of heat accumulation is completely restored. Through this reconstruction of the time series, body temperature change is transformed from discrete data points into a dynamic process with continuous warming characteristics. In the new temperature change sequence, the time interval between temperature nodes remains stable, and the trend of body temperature rise is smoothly connected on the time axis, thus restoring the continuity of body temperature change. The resulting temperature change sequence can fully reflect the transition process of an individual's body temperature from the latent warming signal stage to the sustained warming stage in the time dimension, providing a temperature time series basis for the establishment of nursing response logic.

[0096] By combining time reconstruction with micro-delay adjustments, the temporal continuity of the body temperature change process was fully restored, transforming the latent warming signal into a continuous heat accumulation process on the time axis, reflecting the sustained warming pattern during the latent phase of body temperature change. The resulting temperature change sequence not only preserved the true trend of individual body temperature rise but also achieved a coherent expression in its temporal structure, providing a basis for subsequent temperature-based nursing intervention strategies.

[0097] Step 5: Based on the recombined temperature change sequence, establish a nursing response logic for the identified latent fever signal, and dynamically adjust the execution order of cooling, fluid replacement and heat dissipation measures according to the intensity and time distribution of the latent fever signal, so as to achieve precise nursing intervention in the early stage of infectious fever.

[0098] The specific steps for establishing nursing response logic based on the identified latent fever signals are as follows:

[0099] After obtaining the recombined temperature change sequence, a nursing response logic for latent fever signals is established. In this process, the recombined temperature change sequence is used as the basic input, and the temporal location, duration, rate of increase, maximum temperature, and time interval of the latent fever signal are extracted item by item. Then, the start and end times of each latent fever signal are marked on the time axis, and the corresponding temperature nodes and time spans are recorded. Based on the continuity and intensity variation characteristics of the fever signal on the time axis, it is divided into three nursing response intervals: the initial fever stage, the intermediate fever stage, and the sustained fever stage. Each interval represents a different heat accumulation process: the initial fever stage reflects the beginning of heat accumulation, the intermediate fever stage reflects the continuous rise in body temperature, and the sustained fever stage reflects the stable high body temperature. Based on this, a nursing response logic table is established according to the temperature change characteristics of different stages, corresponding each temperature change to a specific nursing intervention. For example, when the temperature rise signal is in the initial stage and the temperature change is small and short-lived, the nursing response logic is set to initiate heat dissipation measures on the body surface; when the temperature rise signal is in the middle stage and the temperature continues to rise, the nursing response logic is set to perform a combined intervention of fluid resuscitation and heat dissipation measures on the body surface; when the temperature rise signal enters the sustained stage and the temperature remains in the high range, the nursing response logic is set to initiate cooling measures and maintain fluid resuscitation. In this way, the nursing response logic corresponds to the body temperature change process, so that each temperature rise signal can trigger the corresponding nursing action, ensuring that the nursing process is synchronized with the trend of body temperature change.

[0100] In practice, by comparing the duration and magnitude of the latent temperature rise signal on the time axis with thresholds, the process is divided into three nursing response intervals: the initial temperature rise stage, the intermediate temperature rise stage, and the continuous temperature rise stage.

[0101] First, the duration threshold and temperature threshold for body temperature changes are determined. The segment where the duration of the latent warming signal is shorter than the set duration threshold and the temperature increase is lower than the temperature threshold is defined as the initial warming stage, which reflects the process of heat accumulation in the body. Second, when the duration of the latent warming signal exceeds the duration threshold and the temperature increase is greater than the temperature threshold but less than the upper limit of the temperature threshold, this segment is divided into the intermediate warming stage, which characterizes the process of sustained body temperature increase. Finally, when the duration of the latent warming signal far exceeds the duration threshold and the temperature increase reaches or exceeds the upper limit of the temperature threshold, this segment is divided into the sustained warming stage, which describes the heat accumulation process of maintaining a high body temperature.

[0102] It should be noted that:

[0103] The duration and temperature thresholds are determined based on the individual's historical body temperature data and its normal fluctuation range. Specifically, the individual's body temperature changes in a stable state can be statistically analyzed to obtain the average duration and typical temperature rise of the continuous temperature rise process, which are then used as the basic reference values ​​for the duration and temperature thresholds, respectively. Then, the reference values ​​are adjusted appropriately based on the overall fluctuation level in the current time-extended sequence, so that the duration threshold can distinguish between short-term fluctuations and continuous temperature rise processes, and the temperature threshold can distinguish between minute fluctuations and effective temperature rise changes, thereby achieving coordinated division of the time and temperature dimensions in different stages of body temperature change.

[0104] This dual-threshold comparison method based on duration and temperature amplitude can accurately distinguish different stages of body temperature rise in both time and temperature dimensions, thus providing a clear stage basis for the formulation of subsequent nursing response logic.

[0105] After the nursing response logic is established, the execution order of cooling, fluid resuscitation, and heat dissipation measures is dynamically adjusted based on the intensity and time distribution of the latent fever signal. In this process, each measure in the nursing response logic is correlated with the time information of the temperature change sequence, using the intensity of the latent fever signal as the execution priority and the time distribution as the control basis for execution duration. First, the time intervals of each latent fever signal in the temperature change sequence are read, and the intensity of the fever signals is sorted. When the rate of temperature rise exceeds the temperature rise threshold or the duration exceeds the upper limit of the duration, cooling measures are implemented first, using physical cooling methods such as cold compresses, air circulation, and ambient temperature regulation to reduce body surface heat. Second, when the temperature rise is accompanied by water loss or changes in fluid balance, fluid resuscitation measures are implemented simultaneously with cooling measures to promote heat conduction and dissipation. In the mid-stage of fever rise, when the body temperature continues to rise and the rate of heat release slows down, the nursing sequence is adjusted, and fluid resuscitation and heat dissipation measures are implemented simultaneously to achieve thermal balance by maintaining water metabolism and skin heat dissipation. During the sustained warming phase, when the warming signal intensity stabilizes and the temperature remains high, the nursing sequence is adjusted, increasing the frequency of cooling measures and using surface heat dissipation as an auxiliary action to maintain a stable decrease in body temperature. The entire dynamic adjustment process maintains consistency with the distribution of latent warming signals over time, ensuring that the timing of each nursing action matches the stage of body temperature change, thus making the nursing intervention continuous and targeted.

[0106] It should be noted that:

[0107] The setting of the temperature rise threshold and the upper limit of the duration are determined based on the individual's historical body temperature change characteristics and the dynamic change level in the current time-extended sequence. Specifically, the body temperature change data of the individual in a stable state and a non-feverish state can be statistically analyzed to obtain the typical temperature rise rate range per unit time as the basic reference for the temperature rise threshold. Then, the threshold is appropriately increased by combining the overall change amplitude of the temperature rise segment in the current temperature change sequence to distinguish between normal fluctuations and continuous temperature rise trends. At the same time, the duration of the continuous temperature rise process is statistically analyzed, and its maximum duration is extracted as the basis for the upper limit of the duration. This is then corrected by combining the individual's body temperature rhythm characteristics so that the upper limit can reflect the time boundary of the abnormal temperature rise process, thereby effectively distinguishing between rapid temperature rise and continuous temperature rise states in the judgment process.

[0108] After dynamically adjusting the nursing response logic and execution sequence, precise nursing intervention based on latent fever signals is implemented. In this process, the execution time of nursing actions is matched with the time interval of the latent fever signal, relying on the recombined temperature change sequence. For latent fever signals in the initial fever stage, when the body temperature just begins to rise, surface heat dissipation measures are initiated first, corresponding to the early stage of the signal in the time sequence. This is achieved through cold compresses on the skin surface or adjustment of the ambient temperature to help dissipate heat and reduce heat accumulation on the body surface. When the temperature change curve enters the mid-fever stage, the nursing intervention simultaneously transitions to a combined execution state. Surface heat dissipation measures continue, while fluid resuscitation is increased to maintain body fluid balance and promote heat conduction, thus mitigating the fever trend within the body. When the temperature change curve extends into the sustained fever stage, the focus of the nursing intervention shifts to cooling measures. The frequency of cooling is increased, and the rise in body temperature is gradually narrowed through temperature regulation. Surface heat dissipation and fluid resuscitation are maintained throughout the cooling process to ensure coordination between heat release and fluid regulation. During the implementation of nursing interventions, the initiation and termination times of all nursing measures were adjusted according to the time nodes in the temperature change sequence, ensuring that the body temperature change curve and the nursing intervention behavior remained synchronized. When the latent fever signal ended, the frequency of intervention measures was gradually reduced to coordinate nursing behavior with the body temperature recovery process.

[0109] This implementation method dynamically controls the nursing response over time, ensuring that nursing actions are consistent with the process of body temperature changes. This allows for targeted intervention in the early stages of infectious fever, effectively inhibiting further increases in body temperature.

[0110] This invention restores the continuity of the body temperature curve in the time dimension by establishing a time-extended sequence of continuous temperature fluctuations and rearranging the temperature change data. This allows for the extraction of masked deep-seated warming trends from frequently oscillating fluctuation curves over a short period. This method fully presents the true evolution of body temperature changes, effectively eliminating the interference of surface temperature fluctuations on trend identification, and enabling the early identification of latent warming signals in the early stages of infectious fever.

[0111] This invention establishes a nursing response logic after identifying latent fever signals and dynamically adjusts the execution sequence of cooling, fluid replacement, and heat dissipation measures based on the intensity and time distribution of the fever signals. This allows nursing interventions to remain synchronized with the body temperature change process, thereby achieving dynamic cyclical control from body temperature monitoring to nursing intervention. This enables nursing measures to respond flexibly to individual body temperature change characteristics, effectively improving the accuracy and continuity of early fever care, and ultimately enhancing the efficiency of body temperature regulation and the quality of nursing care.

[0112] This invention provides, for example Figure 4 The system shown is a personalized analysis system for continuous body temperature data, including a time-extended sequence establishment module, a temperature rise segment aggregation module, a latent temperature rise signal identification module, a temperature sequence reconstruction module, and a nursing response logic establishment module.

[0113] Time-extended sequence establishment module: Identifies body temperature curves that oscillate frequently within a short period of time, establishes a time-extended sequence of continuous temperature fluctuations for the identified body temperature curves, and rearranges the collected temperature change data within the extended time window;

[0114] Heating Segment Aggregation Module: Based on the established time-extended sequence, calculate the change amplitude and direction of adjacent intervals for each temperature node in the time-extended sequence, and aggregate consecutive temperature nodes with the same change direction to form a heating segment;

[0115] Latent heating signal identification module: Based on the obtained heating segments, calculate the energy difference between the start and end points of each heating segment, and determine the latent heating signal according to the change in energy accumulation rate between adjacent heating segments;

[0116] Temperature sequence reconstruction module: After determining the latent temperature rise signal, the distribution segment of the latent temperature rise signal on the time axis is recombined to obtain the recombined temperature change sequence.

[0117] Nursing response logic establishment module: Based on the recombined temperature change sequence, nursing response logic is established for the identified latent fever signal, and the execution order of cooling, fluid resuscitation and heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the fever signal.

[0118] The present invention provides a method for individualized analysis of continuous body temperature data, which is implemented through the aforementioned system for individualized analysis of continuous body temperature data. For details of the specific method and process of the system for individualized analysis of continuous body temperature data, please refer to the embodiment of the method for individualized analysis of continuous body temperature data, which will not be repeated here.

[0119] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for individualized analysis of continuous body temperature data, characterized in that, Includes the following steps: For the identified body temperature curve, a time-extended sequence of continuous temperature fluctuations is established, and the collected temperature change data is rearranged within the extended time window. Based on the established time-extended sequence, the change amplitude and direction of adjacent intervals are calculated for each temperature node in the time-extended sequence, and continuous temperature nodes with the same change direction are aggregated to form a heating segment. Based on the obtained heating segments, the energy difference between the start and end points of each heating segment is calculated, and the latent heating signal is determined according to the change in the energy accumulation rate between adjacent heating segments. After identifying the latent temperature rise signal, the distribution segments of the latent temperature rise signal on the time axis are recombined to obtain the recombined temperature change sequence. Based on the recombined temperature change sequence, a nursing response logic is established for the identified latent fever signals, and the execution order of cooling, fluid replacement and heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signals.

2. The method for individualized analysis of continuous body temperature data according to claim 1, characterized in that, Establishing a time-stretched sequence of continuous temperature fluctuations includes the following steps: The collected temperature change data are arranged in chronological order of sampling time to form the original body temperature curve. The rate of temperature change between each sampling point is continuously calculated and analyzed point by point. When the temperature change direction of several consecutive sampling points alternates between positive and negative within a preset time window and the alternation frequency exceeds the preset time threshold, the corresponding time segment is marked as a frequent oscillation interval and is merged to generate a high fluctuation segment based on time continuity. Based on the calibrated high fluctuation range, the time span of the high fluctuation range is proportionally enlarged on the time axis, the original sampling time interval is extended by a fixed multiple, and the original temperature sampling point positions are redistributed in the extended time window, keeping the temperature value unchanged and forming a time extension sequence that covers the entire fluctuation process. The obtained time-extended sequence is rearranged, and the temperature nodes are adjusted sequentially according to the original sampling order. The time intervals are evenly divided and the temperature nodes are redistributed within the extended time window. Trend characterization is performed on the rearranged time-extended sequence to identify the upward trend of temperature over continuous time periods and to characterize the warming process masked by surface fluctuations.

3. The method for individualized analysis of continuous body temperature data according to claim 2, characterized in that, Aggregating consecutive temperature nodes with the same direction of change to form a heating segment includes the following steps: After the time-extended sequence is established, the relationship between each temperature node in the time-extended sequence and its adjacent temperature nodes before and after it is continuously calculated to obtain the magnitude and direction of temperature change, and the start time, end time and direction of change of each interval are recorded on the time axis. After identifying the direction of change in adjacent intervals, continuous temperature nodes with the same direction of change are continuously aggregated. Starting from the beginning node of the time extension sequence, the direction of change information is scanned point by point along the time axis. Temperature nodes with the same direction of change and continuous distribution are included in the same aggregate set, and the time span and boundary of each aggregate set are recorded. Based on the aggregation of temperature nodes in the same direction, the resulting direction set is screened and its boundaries are established. From this set, a continuous set with an upward changing direction is selected as a candidate set of heating segments, and heating segments are formed based on the candidate set of heating segments.

4. The method for individualized analysis of continuous body temperature data according to claim 3, characterized in that, When connecting adjacent heating segments, the interval between the end time and the start time of the adjacent heating segments is compared. When the interval is less than the set time interval threshold, the temperature node data of the adjacent heating segments are merged in chronological order, and the start time, end time and time span of the new heating segment are recorded in a unified manner to form a heating interval with continuous time distribution in the time extension sequence.

5. The method for individualized analysis of continuous body temperature data according to claim 3, characterized in that, Determining the latent temperature rise signal based on the change in energy accumulation rate between adjacent temperature rise segments includes the following steps: Based on the obtained heating segments, the energy difference between the start and end of each heating segment is calculated. The start temperature value of the heating segment is taken as the initial temperature and the end temperature value is taken as the final temperature. The temperature nodes between the start and end are extracted and arranged in chronological order. The change amplitude of adjacent temperature nodes is recorded and accumulated to obtain the energy difference value corresponding to each heating segment. After obtaining the energy difference of each heating segment, the energy accumulation rate is determined according to the time span of the heating segment, the start time and end time of the heating segment are extracted, the time interval is calculated and the energy difference is mapped to the time interval to determine the energy change rate per unit time, forming a time-continuous energy accumulation rate sequence. Based on the energy accumulation rate sequence, the temporally adjacent heating segments are compared one by one. When the energy accumulation rate of the later heating segment is greater than that of the earlier heating segment, it is determined that there is a latent heating signal in the corresponding time interval, and the time interval corresponding to the continuously increasing energy accumulation rate is marked as the time distribution band of the latent heating signal. After identifying the latent fever signal, the risk trend of infectious fever is judged based on the changing trend of the energy accumulation rate. When the energy accumulation rate continues to increase and the latent fever signal is continuously distributed, it is judged that the risk of infectious fever is forming or is in the aggravation stage.

6. The method for individualized analysis of continuous body temperature data according to claim 5, characterized in that, In the process of determining the latent warming signal based on the change in energy accumulation rate between adjacent warming segments, when the energy accumulation rate of three or more consecutive warming segments increases sequentially and the change in energy accumulation rate exceeds the preset amplitude threshold, the corresponding continuous time interval is determined as the continuous distribution interval of the latent warming signal, and the risk of infectious fever is preferentially judged to be in the formation stage within the continuous distribution interval.

7. The method for individualized analysis of continuous body temperature data according to claim 5, characterized in that, Recombining the distribution segments of the latent temperature rise signal on the time axis includes the following steps: After the latent temperature rise signal is identified, the distribution segment of the latent temperature rise signal on the time axis is extracted, the start and end points of each latent temperature rise signal are marked in chronological order, all temperature nodes and their time labels are extracted in the segment, and the duration, start temperature, end temperature and temperature change amplitude are recorded to establish a segment set containing time and temperature information. After the distribution segments of the latent temperature rise signal are extracted, the time distribution order of each segment is recombined and sorted from early to late according to the start time. The end time and start time of adjacent segments are compared. When the time interval is greater than the interval duration threshold, the start time of the next segment is adjusted to shorten the time interval, while keeping the arrangement of temperature nodes within the segment unchanged. A micro-delay strategy is introduced between adjacent latent temperature rise signal segments. By determining the time difference between the end time of the previous segment and the start time of the next segment, the starting point of the next segment is shifted forward on the time axis and the temperature node time label is updated, thus compressing the response interval and forming a continuously arranged time structure. Based on the recombination and slight delay adjustment of the latent temperature rise signal segments, all latent temperature rise signal segments are integrated to form a recombined temperature change sequence.

8. The method for individualized analysis of continuous body temperature data according to claim 7, characterized in that, When a micro-delay strategy is introduced between adjacent latent temperature rise signal segments, the magnitude of the delay adjustment is determined based on the time difference between the end time of the previous segment and the start time of the next segment. When the time difference is less than the preset delay threshold, the segment time position remains unchanged. When the time difference is greater than the preset delay threshold, the start point of the next segment is shifted forward, and the adjusted temperature node time tag is updated synchronously.

9. The method for individualized analysis of continuous body temperature data according to claim 7, characterized in that, Establishing a nursing response logic based on the identified latent fever signal includes the following steps: After obtaining the recombined temperature change sequence, a nursing response logic for the latent temperature rise signal is established. The time location, duration, temperature rise rate, maximum temperature and time interval of the latent temperature rise signal are extracted. The start time and end time are marked on the time axis. Based on the characteristics of continuity and intensity change, it is divided into three nursing response intervals: the initial temperature rise stage, the middle temperature rise stage and the continuous temperature rise stage. A nursing response logic table is established according to the temperature change characteristics of different stages. After the nursing response logic is established, the execution order of cooling, fluid replacement and heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signal. The measures in the nursing response logic are matched with the time information of the temperature change sequence. The intensity of the latent fever signal is used as the priority basis and the duration is controlled by the time distribution, forming a nursing action execution order that matches the stage of body temperature change. Based on the dynamic adjustment of nursing response logic and the execution sequence of nursing actions, the execution time of nursing actions is matched with the time interval of latent fever signal. In the initial fever stage, heat dissipation measures are initiated, in the middle fever stage, fluid resuscitation and heat dissipation are combined, and in the continuous fever stage, the frequency of cooling measures is increased while maintaining fluid resuscitation and heat dissipation.

10. A system for individualized analysis of continuous body temperature data, used to implement the method for individualized analysis of continuous body temperature data as described in any one of claims 1-9, characterized in that, It includes a time-extended sequence establishment module, a temperature rise fragment aggregation module, a latent temperature rise signal identification module, a temperature sequence recombination module, and a nursing response logic establishment module; Time-extended sequence establishment module: For the identified body temperature curve, a time-extended sequence of continuous temperature fluctuations is established, and the collected temperature change data is rearranged in the extended time window. Heating Segment Aggregation Module: Based on the established time-extended sequence, calculate the change amplitude and direction of adjacent intervals for each temperature node in the time-extended sequence, and aggregate consecutive temperature nodes with the same change direction to form a heating segment; Latent heating signal identification module: Based on the obtained heating segments, calculate the energy difference between the start and end points of each heating segment, and determine the latent heating signal according to the change in energy accumulation rate between adjacent heating segments; Temperature sequence reconstruction module: After determining the latent temperature rise signal, the distribution segments of the latent temperature rise signal on the time axis are recombined to obtain the recombined temperature change sequence; Nursing response logic establishment module: Based on the recombined temperature change sequence, nursing response logic is established for the identified latent fever signal, and the execution order of cooling, fluid resuscitation and body surface heat dissipation measures is dynamically adjusted according to the intensity and time distribution of the latent fever signal.

Images