A method and system for constructing an index for evaluating breastfeeding

Abstract

The present application relates to the technical field of breast feeding, in particular to a breast feeding evaluation index construction method and system, comprising the following steps: obtaining feeding time and frequency, collecting sucking and swallowing signal sequences, identifying sucking continuous segments and swallowing stable stages, extracting rhythm continuation segments, calculating fluctuation amplitude and time proportion, extracting high-frequency sucking point and swallowing concentrated segment overlap area, measuring matching density and interval fluctuation, screening stable and slowly changing segments, counting behavior type and frequency, and generating feeding evaluation results, in the present application, by collecting time sequence signals of sucking and swallowing, continuous behavior data flow is constructed, rhythm segments and synchronous time intervals are extracted, matching density and rhythm fluctuation level are quantified, stable and slowly changing behavior characteristics are identified, combined with behavior classification and frequency statistics, multi-dimensional evaluation of milk supply, feeding rhythm and matching performance is realized, and the accuracy of breast feeding state judgment and data support ability are improved.

Description

Technical Field

[0001] This invention relates to the field of breastfeeding technology, and in particular to a method and system for constructing breastfeeding evaluation indicators. Background Technology

[0002] The field of breastfeeding technology encompasses technologies for assessing, guiding, and intervening in the entire process of a breastfeeding mother providing nutrition to her infant. The core of this field lies in the collection, recording, and analysis of quantitative and qualitative information on processes such as maternal milk secretion, infant sucking behavior, and physiological responses. This information is used to determine whether feeding behavior meets the infant's growth needs and to guide mothers in adjusting their feeding methods. Overall, the field of breastfeeding technology covers areas such as milk production monitoring, infant sucking behavior analysis, nipple posture detection, physiological signal recognition, recording of vital sign changes, and management of feeding cycle parameters. It often relies on sensor detection, physiological data collection, behavioral observation quantification, and data comparison methods to construct a scientific assessment system.

[0003] The method for constructing breastfeeding evaluation indicators refers to a quantitative analysis system for assessing feeding quality, targeting three aspects: adequacy of feeding, matching of physiological state, and rationality of behavioral performance. This patent covers the formation of evaluation indicator combinations based on specific physiological parameters such as maternal milk production, infant breastfeeding frequency, breastfeeding duration per unit time, infant weight change rate, and comparison of milk intake and excretion data. The indicators are quantified by collecting specific measurements during mother-infant interaction and establishing an indicator scoring method. It employs methods such as setting standard intervals, determining scoring weights, and defining threshold levels, combined with statistical distribution patterns, to construct the indicators, ultimately forming an operable evaluation standard system.

[0004] Current technologies for handling breastfeeding often focus on static physiological data analysis, such as changes in milk volume and weight, failing to delve into the dynamic process of feeding behavior. The lack of real-time monitoring and detailed analysis of sucking and swallowing rhythms makes it difficult to effectively identify and address potential coordination problems and rhythm abnormalities that may arise in infants at different feeding stages. Furthermore, current technologies do not prioritize the continuity of feeding behavior and the interactions between behaviors, limiting a comprehensive assessment of the quality of mother-infant interaction and reducing the timeliness and specificity of interventions. The lack of dynamic and continuous behavioral data analysis means that assessment results may not fully reflect feeding effectiveness, impacting the scientific validity and practicality of breastfeeding guidance. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method and system for constructing breastfeeding evaluation indicators.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for constructing breastfeeding evaluation indicators, comprising the following steps:

[0007] S1: Obtain feeding time and frequency during breastfeeding. Collect the number of times the baby sucks and swallows by the baby through vibration sensors deployed on the baby's chest. Integrate the data in chronological order to obtain feeding record data.

[0008] S2: Based on the feeding record data, extract continuous segments of sucking signals and stable swallowing phases, identify rhythm continuity segments, record start and end times, calculate sucking fluctuation amplitude, and statistically analyze the proportion of time periods to obtain feeding rhythm characteristic segments.

[0009] S3: Based on the feeding rhythm feature segments, extract the high-frequency sucking time points and compare them with the concentrated time periods of swallowing actions, identify overlapping segments and extract synchronous time intervals, calculate the matching density of sucking and swallowing actions within the synchronous intervals, measure the interval fluctuations, and obtain the sucking and swallowing coordination performance.

[0010] S4: Based on the sucking and swallowing coordination, retrieve stable feeding segments, screen gradually changing segments, record start and end times and frequencies, classify and statistically analyze according to feeding behavior types, and obtain milk supply information.

[0011] S5: Based on the milk supply, sucking and swallowing coordination, and feeding rhythm characteristics, classify infant behavior types, count time and frequency, integrate feeding behavior summaries, and generate breastfeeding evaluation results.

[0012] As a further aspect of the present invention, the feeding record data includes feeding time values, feeding frequency values, sucking frequency values, and swallowing frequency values. The feeding rhythm characteristic segments specifically include continuous sucking interval values, stable swallowing time interval values, sucking fluctuation amplitude values, and rhythm period proportion values. The sucking and swallowing coordination performance includes synchronization time intervals, action matching density values, and rhythm interval fluctuation values. The milk supply situation specifically includes stable feeding frequency values, gradual change segment quantity values, and behavior type classification results. The breastfeeding evaluation results specifically refer to infant behavior type classification results, behavior duration values, behavior occurrence frequency values, and behavior summary content.

[0013] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0014] S101: Acquire sucking and swallowing motion signals collected by the infant's chest vibration sensor, record the corresponding time points, pair the motions with the time and integrate them into a time series to generate a sucking and swallowing motion sequence;

[0015] S102: Based on the sucking and swallowing action sequence, identify the time corresponding to the first sucking signal and the last swallowing signal, construct the feeding start and end time interval, and mark the action timestamps during the period to obtain the feeding cycle time interval;

[0016] S103: Based on the sucking and swallowing action sequence and the feeding cycle time interval, filter and sort the events within the interval, count the number of events and the interval, generate the feeding frequency value and the feeding time value, and obtain the feeding record data.

[0017] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0018] S201: Based on the feeding record data, extract segments with continuous signals and intervals not exceeding the sucking interval value, filter time segments that meet the conditions, and obtain the interval value of continuous sucking segments.

[0019] S202: Based on the time range of the continuous sucking segment interval value, extract the corresponding swallowing signal sequence, screen out the time period with stable amplitude and gentle slope, retain the signal segments with consistent rhythm, and generate the stable time period value of swallowing signal.

[0020] S203: Based on the interval values ​​of the continuous sucking segments and the stable time period values ​​of the swallowing signal, extract the time intersection segments, call the amplitude sequence of the sucking segments in the intersection, calculate the fluctuation amplitude of the sucking action in the segment, and at the same time, calculate the proportion of the total duration of the intersection segments in the total duration of the sucking segments to obtain the feeding rhythm feature segments.

[0021] As a further aspect of the present invention, the formula for calculating the fluctuation amplitude of the sucking action within the segment is as follows:

[0022]

[0023] Among them, W v V represents the fluctuation range of the sucking action within the segment. si V represents the sampling point of the i-th suction signal amplitude. s The maximum value (V) represents the average value of the sucking signal amplitude sequence. s ) represents the maximum value of the sucking amplitude sequence, min(V s ) represents the minimum value of the sucking amplitude sequence, n represents the total number of samples of the sucking signal within this segment, and δ t It represents the ratio of the variance of the time interval between adjacent sampling points to the average interval.

[0024] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0025] S301: Based on the feeding rhythm feature segment, extract the sucking trajectory data within the corresponding time period, locate the frequent fluctuation points in each sucking signal segment, select the time segments with concentrated peak distribution as the high-frequency point set of behavior, and generate high-frequency sucking time segments.

[0026] S302: Based on the high-frequency sucking time segment, match the vibration data of the corresponding time period in the swallowing record, extract the vibration dense area and identify the time range that overlaps with the high-frequency sucking time, mark the distribution position and total number of occurrences of the overlapping area according to the complete cycle, and generate a sucking and swallowing synchronization segment.

[0027] S303: Based on the sucking and swallowing synchronization segment, calculate the density of behavior within the synchronization segment, measure the difference in time interval changes between adjacent synchronization segments, combine the distribution frequency statistics, integrate the behavioral coordination characteristic indicators, and obtain the sucking and swallowing coordination performance.

[0028] As a further aspect of the present invention, the formula for calculating the density of behavior within the synchronization segment is as follows:

[0029]

[0030] Among them, D i n represents the density of behavior within the i-th synchronization segment. j T represents the number of times the j-th behavior appears in the i-th segment, M represents the total number of behaviors in the segment, and T represents the number of times the j-th behavior appears in the i-th segment. i T represents the duration of the i-th synchronization segment. avg σ represents the average duration of all synchronization segments, and σ represents the standard deviation.

[0031] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0032] S401: Based on the sucking and swallowing coordination performance, obtain the time series of sucking and swallowing signals in the feeding record data, extract the time interval and amplitude difference between adjacent actions, screen out continuous segments whose rhythm fluctuation amplitude matches the stable sucking and swallowing rhythm, and obtain the stable rhythm time interval.

[0033] S402: Based on the signal fluctuation sequence corresponding to the stable rhythm time interval, compare the amplitude changes of adjacent time points, retain continuous segments with stable fluctuation amplitudes, extract time intervals that meet the conditions, and obtain the time zone of the slowly changing action segment.

[0034] S403: Based on the feeding cycle number to which the time zone of the slowly changing action segment belongs, the frequency of occurrence in the cycle is counted, the signal amplitude value and rhythm interval value of the segment are extracted, and the corresponding segments are divided into sucking and swallowing behavior types. The number of segments is accumulated according to the behavior group to obtain the milk supply situation.

[0035] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0036] S501: Based on the milk supply situation, combined with the sucking and swallowing coordination and feeding rhythm characteristics, extract the time segment, action intensity and flow rate type in the segment, split the behavioral data by dimension, and compare each item with the class performance characteristics in the behavioral standard description to generate a behavior comparison mapping table.

[0037] S502: Based on the behavior mapping table, extract the time segments corresponding to each type of standard behavior throughout the process, statistically analyze the start and end intervals, frequency of occurrence and interval distribution characteristics of the behavior in the recording period, construct the performance classification of multiple types of behavior in the cycle, and generate the behavior performance distribution structure.

[0038] S503: Based on the aforementioned behavioral performance distribution structure, multiple behavioral time periods are sequentially spliced ​​together to form a time map. Differentiated time period behaviors are integrated and described according to their performance attribution, and summarized into a complete set of behaviors within a single cycle to generate breastfeeding evaluation results.

[0039] A system for constructing evaluation indicators for breastfeeding, comprising:

[0040] The feeding data acquisition module monitors sucking and swallowing signals in real time by installing a vibration sensor on the infant's chest. It correlates the data with the start and end times of feeding, processes the two action channels on a unified time axis, and integrates them to generate a structured output that includes timing, amplitude changes, and frequency, thus obtaining feeding record data.

[0041] Based on the feeding record data, the rhythm extraction module extracts continuous action segments in the sucking signal with intervals that meet the threshold, records the start and end times and extracts the amplitude sequence, locates segments in the swallowing signal that meet the stability standard, performs intersection processing, calculates the fluctuation amplitude of the sucking action, and counts the duration ratio of the intersection segment in the sucking segment to obtain the feeding rhythm feature segment.

[0042] Based on the rhythmic feature segments, the analysis module locates time points where the amplitude of the sucking signal changes frequently, selects the time intersection with the significant change interval in the swallowing signal, calculates the ratio of the number of suckings to the duration of swallowing within the synchronous segment, generates the synchronous density, and analyzes the time interval fluctuations of adjacent synchronous segments to obtain the coordination performance of sucking and swallowing.

[0043] The behavior classification module, based on the sucking and swallowing coordination, filters out slow-changing action segments with gentle amplitude and slope that meet the standard, extracts the mean amplitude and rhythm interval of the sucking or swallowing signal, classifies them according to the standard, records the frequency and proportion of the behavior in each cycle, and obtains the milk supply situation.

[0044] The indicator generation module, based on the milk supply, sucking and swallowing coordination, and rhythmic characteristics, statistically analyzes the duration and frequency of eligible segments according to standards, organizes the behavioral distribution within the cycle, and generates the breastfeeding evaluation results for a single cycle.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] In this invention, by collecting the temporal signals of sucking and swallowing, a continuous behavioral data stream is constructed, rhythmic segments and synchronization time intervals are extracted, the coordination density and rhythm fluctuation level are quantified, stable and slowly changing behavioral characteristics are identified, and combined with behavioral classification and frequency statistics, a multi-dimensional assessment of milk supply, feeding rhythm and coordination performance is achieved, thereby improving the accuracy of breastfeeding status judgment and data support capabilities. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a schematic diagram of the steps of the present invention;

[0049] Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0050] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0051] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0052] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0053] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0054] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0055] This invention provides a method for constructing breastfeeding evaluation indicators, comprising the following steps:

[0056] S1: Obtain the feeding time and frequency during breastfeeding. By deploying a vibration sensor on the baby's chest, continuously collect the sucking and swallowing signals generated by the baby during feeding. Simultaneously record the start and end times of feeding, construct a data sequence with time correspondence, integrate the data in chronological order, and obtain feeding record data.

[0057] S2: Based on feeding record data, extract continuous segments of sucking signals and steady phases of swallowing movements exhibited by the infant during feeding, then match the time periods of concentrated vibration in the swallowing data, identify overlapping time periods, mark the start and end times, calculate the fluctuation amplitude of sucking movements within the segments, and statistically analyze the proportion of rhythm segments to obtain feeding rhythm feature segments.

[0058] S3: Based on the feeding rhythm feature segments, locate the high-frequency time points of sucking behavior in the corresponding time period, then match the time period of concentrated vibration in the swallowing data, identify the overlapping time period intersection, extract the time period range of synchronous occurrence and sort out the coverage frequency of synchronous behavior in the whole cycle, calculate the matching density of sucking and swallowing actions in the synchronous segment, and measure the time interval fluctuation range between consecutive coordination segments to obtain the sucking and swallowing coordination performance.

[0059] S4: Based on the performance of sucking and swallowing coordination, retrieve complete feeding record data, locate stable feeding segments with stable rhythm and uniform intensity of movements during the feeding process, screen slow-change segments with stable signal fluctuations and good continuity of movements, extract the corresponding start and end times and the frequency of occurrence in the feeding cycle, classify the slow-change segments into behavioral types, and perform group statistics to obtain the milk supply situation.

[0060] S5: Based on the milk supply situation, combined with the analysis of sucking and swallowing coordination and feeding rhythm characteristics, compare them with the behavioral standard descriptions, summarize the corresponding types of the behavioral patterns presented in the content in the standard performance range, sort out the time range and repetition frequency of each type of behavior in the whole process, and integrate them into a behavioral summary of a single feeding period to generate breastfeeding evaluation results.

[0061] Feeding record data includes feeding time, feeding frequency, number of sucks, and number of swallows. Feeding rhythm characteristics specifically include continuous sucking intervals, stable swallowing periods, sucking fluctuation amplitude, and the proportion of rhythmic periods. Sucking and swallowing coordination includes synchronization time intervals, movement matching density, and rhythmic interval fluctuation. Milk supply status includes stable feeding frequency, number of gradual changes, and behavior type classification results. Breastfeeding evaluation results specifically refer to infant behavior type classification results, behavior duration, behavior frequency, and behavior summary content.

[0062] The specific steps of S1 are as follows:

[0063] S101: Acquire sucking and swallowing motion signals collected by the infant's chest vibration sensor, record the corresponding time points, pair the motions with the time and integrate them into a time series to generate a sucking and swallowing motion sequence;

[0064] First, the sampling area of ​​the sensor is determined based on its installation location, and the sampling frequency is set according to the sensor's frequency response parameters. For example, a vibration sensor with a sampling frequency of 1000Hz is used to continuously record weak vibration data on the surface of the infant's chest, generating a raw vibration signal stream. Then, according to the set timestamp synchronization mechanism, each signal sample is labeled with a corresponding millisecond-level time tag, forming a signal sample list with time sequence markings. Next, the raw signal is converted into a dual-channel signal data table with a unified structure through a data format conversion program. One channel corresponds to the vibration characteristics excited by the sucking action, and the other channel corresponds to the swallowing action signal waveform. Subsequently, the data sequences of the sucking signal channel and the swallowing signal channel are retrieved frame by frame from the dual-channel data. In the sucking signal channel, segments with frequent amplitude fluctuations and period intervals close to the average sucking rhythm baseline are extracted. In the swallowing signal channel, high-amplitude segments with concentrated frequency and continuous duration are extracted to determine sucking and swallowing. The central time point of the feature segment is labeled as the sucking point and swallowing point, respectively, based on the time label arrangement. Then, the sucking points and swallowing points are recombined in the order of sampling time from early to late, with each action event corresponding to a specific occurrence time point, forming a list of time-action pairs. If the error between two time points exceeds the set maximum allowable interval threshold (e.g., 200 milliseconds), it is not included in the sequence. Afterward, the sucking points and swallowing points that meet the temporal continuity requirements are uniformly organized into a standard format time series file, such as CSV format, with each record being a combination of "timestamp + action type", such as "12:03:04.123, sucking" or "12:03:06.457, swallowing". An index is used to establish a retrieval table for the action sequence, enabling structured tracking and analysis of subsequent action changes starting from any time point, ultimately generating a sucking and swallowing action sequence.

[0065] S102: Based on the sucking and swallowing action sequence, identify the time corresponding to the first sucking signal and the last swallowing signal, construct the feeding start and end time interval, and mark the action timestamps during the period to obtain the feeding cycle time interval;

[0066] First, the records are read line by line from the structured action sequence, and "sucking" and "swallowing" are extracted according to action type. For the sucking action, the timestamp field is compared item by item, and the earliest occurrence is selected as the first sucking time point. For example, if the time of the first sucking action in the record is "12:03:04.123", then this time is taken as the feeding start time. Then, all action records marked as "swallowing" are traversed, and the timestamp of each record is compared. The record with the largest time value is extracted as the last swallowing time point. For example, if the swallowing time at the end of the record is "12:18:32.456", then this time is set as the feeding end time. Next, the feeding start and end time interval is constructed, that is, the start time is the first sucking time and the end time is the last swallowing time, forming the time interval [12:03:04.123]. [4.123, 12:18:32.456], then the timestamps of all records in the action sequence are judged. All action records falling within this time interval are marked as actions within the feeding cycle, and the time index list is reordered according to the order of the timestamps. If there is an action time interval in the record that exceeds the feeding interruption judgment threshold (e.g., 600 seconds), it is not included in the cycle interval to ensure the accuracy of the feeding continuity judgment. For example, in a certain set of data, the sucking point is recorded as "12:04:11.005" and "12:05:23.019", and the swallowing point is recorded as "12:04:45.701" and "12:06:01.980". Through time series integration and time comparison, the complete start and end action time and action process data are identified and combined, and finally the feeding cycle time interval is obtained.

[0067] S103: Based on the sucking and swallowing action sequence and the feeding cycle time interval, filter and sort the events within the interval, count the number of events and the interval, generate feeding frequency value and feeding time value, and obtain feeding record data;

[0068] First, the generated action sequence is loaded into memory in chronological order. The timestamp of each record is read sequentially, and the time value is converted to a standard time format. The start and end times of the feeding cycle are then used to determine the action for each record. If the action time value is not less than the start time and not greater than the end time, the record is retained; otherwise, it is discarded. For example, if the feeding cycle start time is "08:15:00.000" and the end time is "08:45:00.000", only sucking or swallowing events occurring within this time period are retained. Then, all retained action records are reordered in ascending order of timestamp to construct the complete action flow within the interval. The total number of action events in this flow is then counted. For example, if 80 sucking events and 25 swallowing events are identified within the interval, the total number of events is 105. The time difference between adjacent actions is calculated by calling the timestamp information to form a set of event interval times. If there is a continuous interval in the set that exceeds the feeding rhythm break judgment threshold (e.g., 20 seconds), the corresponding segment will be regarded as a rhythm discontinuity segment and recorded separately. This break threshold is set according to clinical experience as the longest period of time during which the infant continuously loses sucking or swallowing actions. Then, the average frequency of the action events is calculated by dividing the total number of events by the feeding cycle duration. The duration is the difference between the end time and the start time, expressed in seconds or minutes. For example, if the total number of events is 105 and the total cycle time is 30 minutes, then the feeding frequency is 105 ÷ 30 = 3.5 times / minute. At the same time, the duration itself is recorded as the feeding time value. Finally, the number of events, interval distribution, frequency value and feeding duration are summarized to construct a complete record structure and finally obtain the feeding record data.

[0069] The specific steps of S2 are as follows:

[0070] S201: Based on feeding record data, extract segments with continuous signals and intervals not exceeding the sucking interval value, filter time segments that meet the conditions, and obtain the interval value of continuous sucking segments;

[0071] First, the recorded sucking signal channel data is retrieved. Each signal data and its corresponding timestamp information are read sequentially. The time difference between two adjacent signals is used as the basis for calculating the sucking interval. For example, in a certain sampling data, the timestamps are "10:00:00.100", "10:00:00.250", and "10:00:00.400", corresponding to time differences of 150ms and 150ms respectively. After obtaining all time differences, a sucking interval threshold is set. This threshold is set based on the average sucking rhythm of infants. If it is set to 200ms based on clinical observation, then the intervals between the two signals in the above sequence do not exceed this value, and the signal is judged to be continuous. The signal amplitude is then further detected. If the amplitude at each moment is maintained above the baseline value of the sucking action response amplitude, for example, if the baseline value is set to 0.05g and all amplitudes are above 0.06g, then the continuous time period meets the sucking action characteristics. The start and end times of the segment are recorded as candidate segments. Then, the next time period is detected again starting from the next signal record. If the time difference between two adjacent signals is greater than the sucking interval threshold, it is considered that a sucking segment has ended and the next judgment process is restarted. The entire data segment is traversed to filter out all signal segments that meet the interval requirements, forming a preliminary segment set. Then, it is judged whether the length of each segment meets the minimum duration requirement. For example, the minimum sucking segment length is set to 1 second. If the length of a segment is only 0.6 seconds, it is discarded. Finally, the set of segments that meet the sucking interval threshold, have stable amplitude, and have segment length that meets the base duration is retained, and the start and end times of each segment are recorded, such as "10:00:00.100~10:00:04.300", "10:00:05.500~10:00:09.000", etc. Finally, the sucking continuous segment interval value is obtained.

[0072] S202: Based on the time range of the continuous sucking segment interval, extract the corresponding swallowing signal sequence, screen out the time periods with stable amplitude and gentle slope, retain the signal segments with consistent rhythm, and generate stable time period values ​​of swallowing signal;

[0073] First, the interval values ​​are parsed chronologically, and the start and end times of each interval are used as query conditions. The corresponding swallowing signal channel data from the feeding record data is retrieved, and the signal amplitude sequence is read using millisecond-level timestamps. During the reading process, the amplitude change of each signal is recorded item by item, and the signal slope between adjacent points is calculated, i.e., the difference in amplitude change per unit time. All slope values ​​are compared with a set swallowing action slope threshold. If the slope of consecutive signal points is within a stable range (e.g., a threshold of 0.005 g / ms), only signal segments with slope fluctuations within this threshold are retained. Further statistical analysis is performed on the amplitudes within the retained segments, retrieving the maximum, minimum, and average amplitudes within the segment to calculate the amplitude fluctuation value, which is then compared with the stable swallowing amplitude baseline. For example, if the maximum amplitude in a data segment is 0.084 g, the minimum amplitude is 0.079 g, and the average is 0.081 g, then the amplitude fluctuation is 0.005 g. If the fluctuation is less than the swallowing amplitude... If the stability threshold is 0.008g, the segment is determined to be a stable amplitude segment. Then, time segments that simultaneously meet both the slope and amplitude criteria are recorded as candidate swallowing signal segments, and the start and end times of the segment are used as labels. Subsequently, rhythm consistency is judged for multiple segments, the average time interval between signal points within each segment is calculated, and the difference between the average interval values ​​of all segments is judged. If the difference is within the rhythm consistency judgment threshold, for example, set to 20ms, the segment is retained as a rhythm consistent signal segment. If the average interval of a segment is 105ms and the average interval of other segments is 100ms, the difference is 5ms, which meets the rhythm consistency requirement. Finally, all signal segments that meet the requirements are integrated into a unified signal segment data structure, and the start and end times of each segment are recorded, such as "10:00:08.050~10:00:09.500", "10:00:10.200~10:00:11.700", etc., and finally, the stable swallowing signal time period value is generated.

[0074] S203: Based on the interval values ​​of continuous sucking segments and the stable time period values ​​of swallowing signals, extract the time intersection segments, call the amplitude sequence of sucking segments in the intersection, calculate the fluctuation amplitude of sucking action within the segment, and at the same time, calculate the proportion of the total duration of the intersection segment in the total duration of the sucking segment to obtain the feeding rhythm feature segment.

[0075] The formula for calculating the fluctuation range of the sucking action within the segment is as follows:

[0076]

[0077] Among them, W v V represents the fluctuation range of the sucking action within the segment. si V represents the sampling point of the i-th suction signal amplitude. sThe maximum value (V) represents the average value of the sucking signal amplitude sequence. s ) represents the maximum value of the sucking amplitude sequence, min(V s ) represents the minimum value of the sucking amplitude sequence, n represents the total number of samples of the sucking signal within this segment, and δ t It represents the ratio of the variance of the time interval between adjacent sampling points to the average interval;

[0078] Suppose that in a specific feeding monitoring session, the following sucking amplitude sequence (unit: g) was recorded.

[0079] V s = [0.075, 0.080, 0.085, 0.078, 0.082];

[0080] Based on these values, we perform the following calculations:

[0081] max(V s ) = 0.085;

[0082] min(V s ) = 0.075;

[0083]

[0084] Assuming the variability of the time interval is 0.1, this value is derived from statistical analysis of actual time interval data. For example, the variance of the time interval is 0.002 seconds squared, and the average time interval is 0.02 seconds.

[0085] Substitute into the formula:

[0086]

[0087] The results indicate that the amplitude fluctuations were very small within the selected sucking segments, accounting for only 0.0386% of the average amplitude, suggesting that the sucking action was relatively stable within this range. This amplitude fluctuation is an important indicator that helps identify and analyze the infant's sucking stability during feeding, and can be used to optimize feeding strategies to ensure the infant's comfort and efficiency during feeding.

[0088] The specific steps for S3 are as follows:

[0089] S301: Based on feeding rhythm feature segments, extract sucking trajectory data within the corresponding time period, locate frequent fluctuation points in each sucking signal segment, select time segments with concentrated peak distribution as a set of high-frequency behavioral points, and generate high-frequency sucking time segments.

[0090] First, the start and end time information recorded in each rhythm segment is extracted. Using this information as an index, the corresponding pressure sequence interval is located from the original suction pressure trajectory. Within each interval, a curve structure is constructed with time as the horizontal axis and pressure value as the vertical axis. Peak extraction is performed on this curve. By comparing the pressure values ​​of each sampling point with the two time points before and after it, the time nodes that represent local maxima are identified. If the pressure value at this node is higher than the points on either side, and the difference is not lower than the suction fluctuation difference benchmark (set to 0.3 kPa), then this node is marked as a peak point. In the example, if a pressure sequence contains consecutive records of 17.6 kPa, 18.0 kPa, and 17.4 kPa, then the 18.0 kPa point is considered a peak. Peak points are identified, and the time locations of all peak points are collected and sorted by occurrence time. In adjacent time segments, if the number of peaks occurring within a unit of time exceeds the peak density judgment threshold (e.g., more than 3), the time segment is marked as a peak concentration segment. The same screening action is performed in all rhythm segments, and the identified peak density time segments are recorded. At the same time, a table record structure is established with parameters such as segment number, start and end time, number of peaks, and average time interval as fields. For example: number T03, start time 42.5 seconds, end time 48.0 seconds, number of peaks 7, average interval 0.8 seconds. Finally, all high-frequency segments of the records are numbered and archived, and arranged in the feeding order. The output is the sucking high-frequency time segment.

[0091] S302: Based on the high-frequency sucking time segment, match the vibration data of the corresponding time period in the swallowing record, extract the vibration dense area and identify the time range that overlaps with the high-frequency sucking time, mark the distribution position and total number of occurrences of the overlapping area according to the complete cycle, and generate the sucking and swallowing synchronization segment.

[0092] First, it's necessary to define the high-frequency sucking time period, typically by identifying high-frequency components in the signal. For example, a threshold can be set to filter out the higher-frequency portions of the sucking signal. Specifically, when the signal frequency exceeds a certain set value, that time period is marked as a high-frequency sucking time period. For instance, suppose the high-frequency sucking time period is identified as [10s, 15s], where the signal frequency is significantly higher. Next, extract the portion overlapping with the sucking signal time period from the swallowing record. This can be achieved by comparing the timestamps of the swallowing and sucking signals. Suppose the swallowing signal time period is [12s, 18s], which partially overlaps with the sucking signal time period [10s, 15s], then the overlapping interval is [12s, 15s]. After obtaining the overlapping interval, the next step is to extract the vibration intensity interval. This is done by evaluating the intensity of the vibration signal changes within the time period. Typically, the intensity of the vibration signal can be determined by the amplitude change. Suppose the vibration signal shows significant amplitude changes within the [12.5s, 14s] interval, then this interval is identified as the vibration intensity interval. Next, by analyzing the intersection of the high-frequency time period of the sucking signal and the dense vibration interval, their overlapping interval can be determined. For example, if the dense vibration interval is [12.5s, 14s], the overlap with the high-frequency sucking time period [10s, 15s] is [12.5s, 14s]. Finally, the distribution location and total occurrence frequency of the overlapping interval are marked according to the complete cycle. This step is accomplished by calculating the number of vibration signal cycles within the overlapping time period. Assuming the vibration cycle within the overlapping interval is 2 seconds, then the vibration cycle occurs 0.75 times within the [12.5s, 14s] interval, thus marking the overlapping interval with a cycle. Ultimately, the generated sucking and swallowing synchronization segment is [12.5s, 14s], which is the time period during which sucking and swallowing occur synchronously.

[0093] S303: Based on the sucking and swallowing synchronization segment, calculate the density of behavior within the synchronization segment, measure the difference in time interval changes between adjacent synchronization segments, combine the distribution frequency statistics, integrate behavioral coordination characteristic indicators, and obtain the sucking and swallowing coordination performance.

[0094] The specific formula for calculating the density of behavior within the synchronization segment is as follows:

[0095]

[0096] Among them, D i n represents the density of behavior within the i-th synchronization segment. j T represents the number of times the j-th behavior appears in the i-th segment, M represents the total number of behaviors in the segment, and T represents the number of times the j-th behavior appears in the i-th segment. i T represents the duration of the i-th synchronization segment. avgσ represents the average duration of all synchronization segments, and σ represents the standard deviation.

[0097] Assume the following synchronization segment of data is collected:

[0098] Segment 1: T1 = 10000 (10 seconds), Behavioral statistics: Sucking behavior occurred 5 times, swallowing behavior occurred 2 times;

[0099] Segment 2: T2 = 12000 (12 seconds), behavioral statistics: sucking behavior occurred 6 times, swallowing behavior occurred 3 times;

[0100] Segment 3: T3 = 8000 (8 seconds), behavioral statistics: sucking behavior occurred 4 times, swallowing behavior occurred 1 time;

[0101] Furthermore, assuming the average duration T of all segments avg =10000 (10 seconds), and the standard deviation σ is calculated to be 1500 (1.5 seconds);

[0102] Calculate the density of each segment:

[0103] Calculate the density D1 of segment 1:

[0104] n1 = 5 (number of sucking behaviors), n2 = 2 (number of swallowing behaviors);

[0105] M = 5 + 2 = 7;

[0106] T1=10000 (10 seconds);

[0107] Calculate using the formula:

[0108]

[0109]

[0110] Calculate the density D2 of segment 2:

[0111] n1 = 6 (number of sucking behaviors), n2 = 3 (number of swallowing behaviors);

[0112] M = 6 + 3 = 9;

[0113] T2 = 12000 (12 seconds);

[0114] Calculate using the formula:

[0115]

[0116] Calculate the density D3 of segment 3:

[0117] n1 = 4 (number of sucking behaviors), n2 = 1 (number of swallowing behaviors);

[0118] M = 4 + 1 = 5;

[0119] T3 = 8000 (8 seconds);

[0120] Calculate using the formula:

[0121]

[0122] The results indicate that the density of behavior within each synchronization segment was calculated as follows: segment 1 had a density of 0.3889%, segment 2 had a density of 0.4373%, and segment 3 had a density of 0.2857%. These results represent the density of behavior within the differentiated synchronization segments. Higher density (segment 2) indicates more frequent sucking and swallowing behaviors within the segment, while lower density (segment 3) indicates fewer behaviors occurring in the segment. This calculation allows for analysis of the coordination of sucking and swallowing behaviors during differentiated time periods and provides data support for feeding rhythm analysis.

[0123] The specific steps of S4 are as follows:

[0124] S401: Based on the performance of sucking and swallowing coordination, obtain the time series of sucking and swallowing signals in the feeding record data, extract the time interval and amplitude difference between adjacent actions, screen the continuous segments whose rhythm fluctuation amplitude matches the stable sucking and swallowing rhythm, and obtain the stable rhythm time interval.

[0125] First, signal recording sequences from two channels are extracted from the feeding record data, corresponding to sucking and swallowing signals respectively. Each record includes a timestamp and signal amplitude. The data from both channels are uniformly sorted according to time labels to construct a complete time series data stream. Then, the times of sucking and swallowing event points are paired to obtain the time difference between each pair of adjacent events and calculate their interval duration. If the time interval between a pair of actions is greater than the rhythm continuity judgment threshold (e.g., 5 seconds), then that pair of records is not considered for further rhythm analysis. For event point pairs that meet this condition, the corresponding sucking or swallowing amplitude is retrieved, and the difference between the two amplitudes is calculated. Further screening is performed to determine if the amplitude difference is within the allowable range of rhythm fluctuation. The sucking amplitude difference threshold is set to 0.02g, and the swallowing amplitude difference threshold is set to 0.015g. If the amplitudes of two adjacent sucking events are 0.0g and 0.015g respectively, then the difference is considered acceptable. The difference between 80g and 0.095g is 0.015g, which meets the condition for stable sucking rhythm. Similarly, it is determined whether the difference in amplitude between adjacent swallowing is within the threshold. If it is, the event segment is retained. Multiple consecutive action segments that meet the rhythm stability are merged to form stable rhythm candidate segments. Then, the time intervals of all actions in the segment are calculated again to obtain the difference between the maximum and minimum values ​​of the rhythm fluctuation range, and it is determined whether this value is less than the upper limit of rhythm consistency judgment. For example, the rhythm fluctuation is set to not exceed 0.5 seconds. If the maximum interval in a segment is 1.6 seconds and the minimum interval is 1.2 seconds, the difference is 0.4 seconds, which is less than the set threshold. Then the segment is identified as having stable rhythm fluctuation. The start time and end time of the segment are recorded to form a time interval output in a unified format, such as "10:02:01.200~10:02:07.800", and finally the stable rhythm time interval is obtained.

[0126] S402: Based on the signal fluctuation sequence corresponding to the stable rhythm time interval, compare the amplitude changes of adjacent time points, retain the continuous segments with stable fluctuation amplitude, extract the time intervals that meet the conditions, and obtain the time zone of the slowly changing action segment;

[0127] First, the sucking and swallowing signals in the feeding record data are retrieved according to the start and end times of the stable rhythm period. The original signal amplitudes corresponding to each segment are extracted one by one and sorted in ascending order by timestamp to construct a complete and continuous fluctuation value sequence. Based on this, the difference between the amplitudes of any two adjacent sampling points is calculated, and the instantaneous amplitude change in each continuous sequence is recorded and compared with a set fluctuation stability threshold. This threshold is set based on the sensor noise level and the range of normal physiological rhythm variation of the infant. For example, if the fluctuation stability threshold is 0.004g, then the fluctuation amplitude is judged to be stable only when the amplitude change does not exceed this value. All sampling point pairs are traversed and points that meet the condition are continuously marked. Then, adjacent stable points are merged into continuous segments to form a candidate set of slowly changing action segments. For each candidate segment, further judgment is made. The total duration of each segment is checked against the minimum duration requirement for the gradual change segment. For example, if the minimum duration is set to 1.5 seconds, a segment that lasts from "10:05:30.200" to "10:05:31.400" would have a duration of 1.2 seconds, which would not meet the requirement and would be discarded. However, a segment with a duration of 2.0 seconds would be retained. The maximum instantaneous change amplitude within each segment is then summarized and verified to ensure that it is all below the upper limit of the gradual change tolerance. If it is, the segment is confirmed as a valid gradual change action segment. Finally, all continuous time periods that meet the conditions are summarized and a unified output structure is constructed according to the start and end times, such as "10:03:05.100~10:03:07.200" and "10:06:12.600~10:06:14.800", thus obtaining the time zone of the gradual change action segment.

[0128] S403: Based on the feeding cycle number to which the time zone of the slow-change action segment belongs, the frequency of occurrence in the cycle is counted, the signal amplitude value and rhythm interval value of the segment are extracted, and the corresponding segments are divided into sucking and swallowing behavior types. The number of segments is accumulated according to the behavior group to obtain the milk supply situation.

[0129] First, the start and end times of each slowly varying motion segment are compared with existing feeding cycle intervals to determine if it falls within a specific feeding cycle. If the start and end times are completely contained within a cycle, the segment is assigned to the corresponding feeding cycle number, and a correspondence between cycle number and segment number is established. For example, if a segment's time is "10:04:15.000~10:04:18.500" and its feeding cycle number is cycle 3, then the segment is labeled "Segment ID003 - Cycle 3". Next, the segments within each cycle number are counted to determine the frequency of slowly varying motion segments in each feeding cycle. Simultaneously, the corresponding signal amplitude sequence within each segment is retrieved, and the maximum, minimum, and average values ​​within that segment are calculated to form a segment amplitude parameter set. Then, the time interval sequence of continuous signal points is extracted, and its average rhythm interval value is calculated to construct the rhythm of each segment. The rhythmic feature labels, for example, if the average amplitude of a certain segment is 0.081g and the average rhythm interval is 0.95 seconds, then this set of values ​​will participate in the behavior type judgment. The amplitude range of sucking behavior is set to 0.07g~0.09g, and the rhythm interval is 0.8~1.2 seconds. The amplitude range of swallowing behavior is 0.09g~0.12g, and the rhythm interval is 1.3~2.0 seconds. If the amplitude and interval of a segment both fall within the threshold range of sucking behavior, then the segment is marked as sucking type. Conversely, if it falls within the swallowing range, then it is marked as swallowing type. After classification, the number of each behavior type is accumulated according to the feeding cycle. The number of sucking type segments and the number of swallowing type segments in each cycle are counted. For example, if 5 sucking segments and 2 swallowing segments are identified in cycle 3, then the behavior composition feature dominated by sucking in this cycle can be obtained. Finally, the distribution of behavior types in each cycle is output as a behavior statistics table, and the milk supply situation is finally obtained.

[0130] The specific steps of S5 are as follows:

[0131] S501: Based on milk supply, combined with sucking and swallowing coordination and feeding rhythm characteristics, extract time segments, action intensity and flow rate type from the segments, split the behavioral data by dimension, and compare each item with the class performance characteristics in the behavioral standard description to generate a behavior comparison mapping table.

[0132] First, it is necessary to combine the characteristics of sucking and swallowing coordination with feeding rhythm segments to extract the time intervals, intensity of actions, and flow rate type within each segment. Initially, by analyzing milk flow rate and behavioral data, the occurrence time and intensity of sucking and swallowing behaviors within each time interval are determined. For example, in the [12s, 13s] segment, sucking behavior occurs more frequently, with an intensity of 1.5 times per second, while swallowing behavior is sparser, with an intensity of 0.5 times per second. The flow rate type in this segment shows a gradual increase, assuming the flow rate increases from 0.5 mL / s to 1.0 mL / s. Next, these behavioral data are split by dimension: by time interval into different segments such as [12s, 13s], [14s, 15s], etc.; by action intensity into high-intensity segments (e.g., [12s, 13s]) and low-intensity segments (e.g., [14s, 15s]); and by flow rate type into categories such as gradually increasing, stable, and gradually decreasing. After segmentation, each behavioral characteristic in the behavioral standard description is compared. This step uses pre-defined behavioral standards, such as whether the sucking intensity exceeds 1.0 times per second and whether the swallowing behavior is less than 0.3 times per second. For example, the [12s, 13s] segment meets the standard for high-intensity sucking behavior, while [14s, 15s] meets the standard for stable flow rate and low-intensity swallowing behavior. Through these comparisons, a behavioral mapping table is generated, matching each time segment with its corresponding behavioral type, intensity, flow rate type, etc., forming a complete behavioral mapping table. For example, the mapping table records behavioral patterns such as [12s, 13s] corresponding to high-intensity sucking and gradually increasing flow rate, and [14s, 15s] corresponding to low-intensity swallowing and stable flow rate, thus achieving a precise mapping between behavior and milk flow rate.

[0133] S502: Based on the behavior mapping table, extract the time segments corresponding to each type of standard behavior in the whole process, statistically analyze the start and end intervals, frequency of occurrence and interval distribution characteristics of the behavior in the recording period, construct the performance classification of multiple types of behaviors in the cycle, and generate the behavior performance distribution structure.

[0134] First, it is necessary to extract the corresponding time segments for each type of standard behavior throughout the entire process. This is done by looking up the time segments for each type of behavior in the behavior mapping table. For example, in the mapping table, sucking behavior occurs in the segments [12s, 13s], [17s, 18s], and [21s, 22s], and swallowing behavior occurs in the segments [14s, 15s], [19s, 20s], and [23s, 24s]. These time segments can then be extracted. Next, the start and end intervals, frequency of occurrence, and interval distribution characteristics of the behavior are statistically analyzed within the recorded time period. First, the start and end intervals of each behavior are recorded. For example, the intervals for sucking behavior are [12s, 13s], [17s, 18s], and [21s, 22s], and the intervals for swallowing behavior are [14s, 15s], [19s, 20s], and [23s, 24s]. Then, the frequency of each behavior is 3 times. The interval distribution of each behavior can be obtained by calculating the time interval between adjacent segments. Assuming the intervals for sucking behavior are [13s, 14s], [18s, 19s], and [22s, 23s], with each interval being 1 second, the behavior is then categorized within the cycle. This step involves classifying the behavior of each type in different time periods. For example, sucking behavior is characterized by high frequency and intensity in the [12s, 13s] period, while swallowing behavior is characterized by lower frequency in the [14s, 15s] period. Based on these characteristics, a structured behavioral performance distribution is constructed by analyzing the frequency, interval distribution, and intensity of the behaviors. It is assumed that the high frequency of sucking behavior is concentrated in the early stage, while the low frequency is concentrated in the later stage, and swallowing behavior occurs relatively evenly throughout the process. The final behavioral performance distribution structure clearly reflects the temporal characteristics, frequency, and interval characteristics of different behaviors.

[0135] S503: Based on the distribution structure of behavioral performance, multiple behavioral time periods are sequentially spliced ​​to form a time map. Differentiated time period behaviors are integrated and described according to their performance attribution, and summarized into a complete set of behaviors within a single cycle to generate breastfeeding evaluation results.

[0136] First, the time periods of various behaviors need to be sequentially pieced together to form a time map. This process involves connecting the time periods of different behaviors in sequence to create a complete time series. For example, in the aforementioned behavior mapping table, the time periods for sucking behavior are [12s, 13s], [17s, 18s], and [21s, 22s], while the time periods for swallowing behavior are [14s, 15s], [19s, 20s], and [23s, 24s]. These time periods are sequentially pieced together in the time map to form a complete behavioral time series from [12s] to [24s]. Next, the behaviors in the differentiated time periods are integrated and described according to their performance attributes. This process involves classifying the different behaviors within each time period based on their performance characteristics. For example, if sucking behavior is characterized by high frequency and intensity in the [12s, 13s] segment, while swallowing behavior is characterized by low frequency and intensity in the [14s, 15s] segment, then the behaviors in these two segments can be described and classified as high-intensity sucking and low-frequency swallowing. Furthermore, these categorized time periods are combined into a complete set of behaviors within a single cycle. For example, assuming the entire time cycle is [12s, 24s], the alternation of sucking and swallowing behaviors and their performance characteristics constitute the complete behavioral pattern of this cycle. Finally, breastfeeding evaluation results are generated. By integrating and analyzing these behavioral time periods and their performance characteristics, the evaluation results of breastfeeding can be obtained. For example, this process may yield an evaluation of "good coordination between high-frequency sucking and low-frequency swallowing, with stable flow rate changes," indicating that the coordination of sucking and swallowing is good throughout the cycle and meets the standards for healthy feeding.

[0137] Please see Figure 2 A system for constructing evaluation indicators for breastfeeding includes:

[0138] The feeding data acquisition module monitors sucking and swallowing signals in real time by installing a vibration sensor on the infant's chest. It correlates the data with the start and end times of feeding, processes the two action channels on a unified time axis, and integrates them to generate a structured output that includes timing, amplitude changes, and frequency, thus obtaining feeding record data.

[0139] The rhythm extraction module extracts continuous action segments with intervals that meet the threshold from the sucking signal based on feeding record data. It records the start and end times and extracts the amplitude sequence. It locates segments with stability that meet the standard in the swallowing signal, performs intersection processing, calculates the fluctuation amplitude of the sucking action, and counts the duration ratio of the intersection segment in the sucking segment to obtain the feeding rhythm feature segment.

[0140] The coordination analysis module, based on rhythmic feature segments, locates time points where the amplitude of the sucking signal changes frequently, selects the time intersection with the significant change interval in the swallowing signal, calculates the ratio of the number of suckings to the duration of swallowing within the synchronous segment, generates the synchronization density, and analyzes the time interval fluctuation of adjacent synchronous segments to obtain the coordination performance of sucking and swallowing.

[0141] The behavior classification module is based on the performance of sucking and swallowing coordination. It filters slow-changing action segments with gentle amplitude and slope that meet the standard, extracts the mean amplitude and rhythm interval of sucking or swallowing signals, classifies them according to the standard, and records the frequency and proportion of the behavior in each cycle to obtain the milk supply situation.

[0142] The indicator generation module, based on milk supply, sucking and swallowing coordination, and rhythmic characteristics, statistically analyzes the duration and frequency of eligible segments according to standards, organizes the behavioral distribution within the cycle, and generates the breastfeeding evaluation results for a single cycle.

[0143] 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 variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for constructing an index of breastfeeding evaluation, characterized by, The method comprises the following steps: S1: obtaining feeding time and feeding frequency in the breastfeeding process, collecting the number of sucking and swallowing of the infant through a vibration sensor arranged on the chest of the infant, integrating data in time sequence to obtain feeding record data; S2: based on the feeding record data, extracting continuous segments of sucking signals and stable stages of swallowing, identifying rhythm continuation segments, recording start and end times, calculating sucking fluctuation amplitude, and counting time period proportion to obtain feeding rhythm characteristic segments; S3: based on the feeding rhythm characteristic segments, extracting sucking high-frequency time points and comparing them with swallowing action concentrated time periods, identifying overlapping segments and extracting synchronous time intervals, calculating the matching density of sucking and swallowing actions in the synchronous segments, and measuring interval fluctuation to obtain sucking and swallowing coordination performance; S4: based on the sucking and swallowing coordination performance, searching for stable feeding segments, screening slow-changing segments, recording start and end times and frequency, classifying and counting according to feeding behavior types to obtain milk supply conditions; S5: based on the milk supply conditions, sucking and swallowing coordination performance, and feeding rhythm characteristic segments, classifying infant behavior types, counting time and frequency, integrating feeding behavior summary, and generating breastfeeding evaluation results; The specific steps of S2 are: S201: based on the feeding record data, extracting segments with continuous signals and interval time not exceeding the sucking interval value, screening time segments meeting the conditions to obtain sucking continuous segment interval value; S202: based on the time range in the sucking continuous segment interval value, extracting corresponding swallowing signal sequences, screening time periods with stable amplitude and gentle slope, retaining signal segments with consistent rhythm to generate swallowing signal stable time period value; S203: based on the sucking continuous segment interval value and the swallowing signal stable time period value, extracting time intersection segments, calling the amplitude sequence of the sucking segments in the intersection, calculating the fluctuation amplitude of the sucking action in the segment, and simultaneously counting the proportion of the total duration of the intersection segment in the total duration of the sucking segment to obtain the feeding rhythm characteristic segment; The calculation formula of the fluctuation amplitude of the sucking action in the segment is specifically: ； wherein, represents the fluctuation amplitude of the sucking action within the segment, represents the amplitude of the i-th sucking signal sample point, represents the average value of the sucking signal amplitude sequence, represents the maximum value of the sucking amplitude sequence, represents the minimum value of the sucking amplitude sequence, represents the total number of samples of the sucking signal within the segment, represents the ratio of the variance of the time interval of adjacent sample points to the average interval.

2. The method of claim 1, wherein: The feeding record data includes feeding time value, feeding frequency value, sucking frequency value, and swallowing frequency value, the feeding rhythm characteristic segment is specifically sucking continuous interval value, swallowing stable time period value, sucking fluctuation amplitude value, and rhythm time period proportion value, the sucking and swallowing coordination performance includes synchronous time interval, action matching density value, and rhythm interval fluctuation value, the milk supply conditions are specifically stable feeding frequency value, slow-changing segment quantity value, and behavior type classification result, and the breastfeeding evaluation result specifically refers to infant behavior type classification result, behavior duration value, behavior occurrence frequency value, and behavior summary content.

3. The method of claim 1, wherein: The specific steps of S1 are: S101: obtaining sucking and swallowing action signals collected by the infant chest vibration sensor, recording corresponding time points, pairing the actions and time points, and integrating them into time sequence in sequence to generate sucking and swallowing action sequence; S102: Based on the sequence of sucking and swallowing actions, the time corresponding to the first sucking signal and the last swallowing signal is identified, the feeding start and end time interval is constructed, and the action time stamp during the interval is marked to obtain the feeding cycle time interval; S103: Based on the sequence of sucking and swallowing actions and the feeding cycle time interval, the interval events are screened and sorted, the number and interval of events are counted, the feeding frequency value and feeding time value are generated, and the feeding record data is obtained.

4. The method of claim 1, wherein the breast feeding evaluation index is constructed by the steps of: The specific steps of S3 are: S301: Based on the feeding rhythm feature segment, the sucking trajectory data in the corresponding time period is extracted, the frequent fluctuation points in each sucking signal are located, the time segments with concentrated peak values are screened as the behavior high-frequency point set, and the sucking high-frequency time segment is generated; S302: Based on the sucking high-frequency time segment, the vibration data in the corresponding time period in the swallowing record is matched, the vibration intensive interval is extracted, and the time range overlapping with the sucking high-frequency time is identified, the distribution position and total number of occurrences of the overlapping interval are marked according to the complete cycle, and the sucking and swallowing synchronization section is generated; S303: Based on the sucking and swallowing synchronization section, the intensive degree of behavior in the synchronization section is calculated, the time interval change difference between adjacent synchronization sections is measured, the behavior coordination feature index is integrated combined with the distribution frequency statistical value, and the sucking and swallowing coordination performance is obtained.

5. The method of claim 4, wherein: The calculation formula of the intensive degree of behavior in the synchronization section is: ； in, Representing the The density of behavior within each synchronization segment Representing the The behavior in the first The number of times it appears within the segment This represents the total number of all actions within this segment. Representing the The duration of each synchronization segment, This represents the average duration of all synchronization segments. Represents standard deviation.

6. The method of claim 1, wherein the breast feeding evaluation index is constructed by the steps of: The specific steps of S4 are: S401: Based on the sucking and swallowing coordination performance, the time sequence of sucking and swallowing signals in the feeding record data is obtained, the time interval and action amplitude difference between adjacent actions are extracted, the continuous paragraphs with stable sucking and swallowing rhythm fluctuation amplitude are screened, and the stable rhythm time interval is obtained; S402: Based on the signal fluctuation sequence corresponding to the stable rhythm time interval, the amplitude change of adjacent time points is compared, the continuous segments with stable fluctuation amplitude are retained, the time segments meeting the conditions are extracted, and the slowly changing action segment time zone is obtained; S403: Based on the feeding cycle sequence number to which the slowly changing action segment time zone belongs, the occurrence frequency in the cycle is counted, the signal amplitude value and rhythm interval value of the segment are extracted, the corresponding is divided into sucking and swallowing behavior types, the segment number is accumulated according to the behavior group, and the milk supply situation is obtained.

7. The method for constructing breastfeeding evaluation indicators according to claim 1, characterized in that, The specific steps of S5 are: S501: Based on the milk supply situation, the sucking and swallowing coordination performance and the feeding rhythm feature segment are combined, the time interval, action intensive degree and flow type in the paragraph are extracted, the behavior data is split according to the dimension, and the class performance characteristics in the behavior standard description are compared item by item, and the behavior comparison mapping table is generated; S502: Based on the behavior comparison mapping table, the time segment corresponding to each type of standard behavior in the whole process is extracted, the start and end interval, occurrence frequency and interval distribution characteristics of the behavior in the record period are counted, the performance classification of multiple behaviors in the cycle is constructed, and the behavior performance distribution structure is generated; S503: Based on the behavior performance distribution structure, sequentially splice the multi-class behavior time period and form a time atlas, integrate and describe the differentiated time period behavior according to the performance attribution, induce a complete behavior set in a one-time period, and generate a breastfeeding evaluation result.

8. A system for constructing an index of breastfeeding evaluation, characterized by The system is used to implement the construction method of the breastfeeding evaluation index according to any one of claims 1-7, and the system comprises: The feeding collection module monitors the sucking and swallowing action signals in real time by installing a vibration sensor on the chest of the infant, associates the data with the feeding start and end time, processes the two action channels on a unified time axis, integrates and generates a structured output including timing, amplitude change and frequency, and obtains feeding record data; The rhythm extraction module extracts the action section in the sucking signal that is continuous and has an interval time meeting a threshold value based on the feeding record data, records the start and end time and extracts the amplitude sequence, locates the section meeting the stability standard in the swallowing signal, performs intersection processing, calculates the fluctuation amplitude of the sucking action, and calculates the time length proportion of the intersection fragment in the sucking fragment, to obtain a feeding rhythm characteristic fragment; The coordination analysis module locuses the time node with frequent amplitude value change in the sucking signal based on the rhythm characteristic fragment, selects the time intersection of the significant change interval in the swallowing signal, calculates the proportion of the sucking frequency and the swallowing duration in the synchronization section, generates the synchronization intensity, and analyzes the time interval fluctuation of adjacent synchronization fragments, to obtain the coordination performance of sucking and swallowing; The behavior typing module filters the slowly changing action fragments with flat amplitude and slope meeting the standard based on the sucking and swallowing coordination performance, extracts the sucking or swallowing signal amplitude mean value and rhythm interval value, classifies according to the standard, records the occurrence frequency and proportion of the behavior in each period, and obtains the milk supply situation; The index generation module generates the breastfeeding evaluation result of a single period based on the milk supply situation, the sucking and swallowing coordination performance and the rhythm characteristic fragment, according to the standard, and calculates the fragment length and number meeting the conditions, and arranges the behavior distribution in the period to generate the breastfeeding evaluation result of a single period.

Images