Timeline-based ostomy therapy management system

Abstract

The present application relates to the technical field of medical records, in particular to a stoma treatment management system based on a timeline, which comprises a starting collection module, a content arrangement module, a sequence integration module, an identity comparison module and an intermittent presentation module.In the present application, a time difference identification mechanism is established by extracting the image generation time and the submission time, data binding is completed by combining the image number and the identity information, semantic merging is achieved by integrating the excrement description, product replacement and processing action content, consistency judgment logic of multi-source information is constructed, the sequence structure is constructed according to the time field to guarantee the integrity and accuracy of record connection, the differences of each field content in the same period are identified and a collaborative identifier is generated based on the patient identity, the correlation ability of multi-party data between cross-period and cross-person is improved, and finally the image and nursing content are encapsulated according to the time sequence to form a continuous graphic view with clear structure, consistent content and comparison.

Description

Technical Field

[0001] This invention relates to the field of medical record technology, and in particular to a timeline-based ostomy treatment management system. Background Technology

[0002] The field of medical record technology involves the collection, storage, management, and presentation of patient medical information. Core aspects of this technology include the chronological organization of patient basic information and treatment processes, the archiving of medical records and imaging data, and the transfer and sharing of medical history across different medical institutions. It encompasses a systematic approach to information-based collection, standardized recording, secure storage, and cross-platform sharing. Traditional timeline-based ostomy management methods refer to the chronological organization and management of surgical records, complications, nursing care records, and product usage information generated during the long-term care and treatment of ostomy patients. Traditional methods typically rely on paper medical records and electronic medical records within a single medical institution, manually entering information such as the surgical procedure, basic ostomy condition, nursing care changes, and related medical treatments according to the time of visit.

[0003] Current technologies rely on a single visit time point to linearly organize patients' stoma-related records, lacking means to identify differences between the generation and submission times of image data. This leads to temporal misalignment of information across different stages. Record content mainly relies on manual entry, making it difficult to establish unified content standards for information submitted by different nursing staff. There is a lack of consistency in the judgment of excrement descriptions and treatment information, resulting in conflicts and repetitions in the descriptions of similar records. In addition, there is a lack of a mechanism for cross-time period data comparison using identity tags, affecting the accurate identification of collaborative content. The presentation method fails to combine field structure and time dimension to construct continuous text and image information, resulting in fragmented data browsing, difficulty in tracing, and limited nursing pathway analysis. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a timeline-based ostomy treatment management system. The technical solution is as follows: On the one hand, a timeline-based ostomy treatment management system is provided, which includes: The initial acquisition module collects stoma type image information and equipment information, extracts and compares the image generation time with the submission time, judges the differences, verifies whether the type and height fields are empty, binds the image number and identity, marks inconsistent times, and generates and organizes image and text data. The content organization module is based on the organized graphic data and collects image number, product replacement, excrement description, processing action, location and executor. It compares the excrement description with the fixed word set, merges the images with the replacement and processing content, matches the location and executor source, and generates a unified description item group. The sequential integration module extracts the replacement time, image time, and processing time based on the unified description entry group, sorts them by time, marks the image as lagging if it is later than the replacement time threshold, and marks the processing as corrected if it is earlier than the replacement, thus generating a sorted time series. The identity verification module references the patient's identity in the sorted time series and compares it with the product name, replacement time, and excrement description filled in by the patient during the same period. If the fields are inconsistent, a collaborative identifier is set, a group of fields to be verified is established and the identity label is retained, and a collaborative information group is generated.

[0005] As a further aspect of the present invention, the organized graphic data includes stoma basic information, image information, time status markers, and identity binding relationships; the unified description item group includes product replacement information, excrement characteristics, treatment behavior elements, and executor and location elements; the sorted time series includes replacement time series, image time series, treatment time series, and sequence identifiers; and the collaborative information group includes patient identity elements, product record elements, field difference elements, and verification field group.

[0006] As a further aspect of the present invention, the replacement time threshold is used to determine the time difference boundary between the replacement behavior and the order of image and processing time; If the fields are inconsistent, a collaboration flag is set, indicating that when there are differences between the patient record and the collected information in key fields, it is marked as a state that requires collaborative verification.

[0007] As a further aspect of the present invention, the initial acquisition module includes: The image acquisition submodule obtains the image source device information and image file generation time, acquires surrounding skin images, extracts the generation time parameter and submission time parameter recorded in the image file, compares the difference between the two values, determines whether the difference exceeds the image submission time difference threshold, and generates a time consistency label. Based on the time consistency label, the field screening submodule performs null value detection on the input stoma type and stoma height fields, determines whether there is any missing content in either of the two fields, classifies and labels the field status, and generates field integrity labels. The image binding submodule calls the field integrity label and image number to obtain the corresponding data record identity parameter, performs one-to-one matching and binding of the image number and data record identity, performs unified labeling on records with abnormal time consistency labels during the binding process, and generates organized image and text data.

[0008] As a further aspect of the present invention, the content organization module includes: The description verification submodule acquires the organized graphic data and collects the excrement description content. It then matches and compares each excrement description with the terms in the fixed descriptive term set, calculates the average matching deviation value of the descriptive terms, and generates an excrement description consistency label. The content merging submodule, based on the excrement description consistency tag, calls the image number, product replacement content and processing action field information, performs a merging judgment on the product replacement content and processing action fields corresponding to the same image number, merges the information of fields with consistent content under the same image number, and generates the content group corresponding to the image number. The information matching submodule retrieves the processing location and executor information based on the content group corresponding to the image number, determines whether the source identifiers of the processing location and executor fields in the same record are consistent, and classifies and integrates items with consistent sources to generate a unified description entry group.

[0009] As a further aspect of the present invention, the sequential integration module includes: The time extraction submodule obtains the image number in the unified description entry group, extracts the corresponding replacement time, image generation time and processing record time, classifies and stores the three time information according to the record number, removes records with missing time fields, and generates a time field set. The sequential judgment submodule judges the sequence of image generation time and replacement time based on the time field set. If the image time is later than the replacement time threshold, an image lag indicator is marked. At the same time, the processing record time is compared with the replacement time order. If the processing time is earlier than the replacement time threshold, an order correction flag is written and a time anomaly indicator set is generated. The sequential structure submodule calls the time anomaly identifier set and time field set to sort all records in ascending order based on replacement time, processing time and image generation time. It then organizes and integrates the time and identifier corresponding to each record to generate a sorted time series.

[0010] As a further aspect of the present invention, the identity verification module includes: The identity extraction submodule obtains the record content in the sorted time series, extracts the patient identity tags in the records, filters the product name, replacement time and excrement description fields from the records filled in by the same patient in the same time period, numbers and archives each group of information according to the patient identity, and generates the same patient field group. The content comparison submodule, based on the same patient field group, calls the corresponding fields in multiple groups, compares the product name, replacement time and excrement description content, and records any inconsistency in any field as a field difference item. The number associated with the field difference item is marked with a collaborative record identifier to generate a field difference identifier group. The difference labeling submodule extracts the record number with collaborative record identifier based on the field difference identifier group, establishes a structure for storing the two field groups in the field difference item respectively, retains the corresponding identity label, and uniformly classifies each group of records into a structured set to generate a collaborative information group.

[0011] As a further aspect of the present invention, the system also includes an intermittent presentation module: The intermittent presentation module, based on the collaborative information group, filters records for the selected time period, generates stoma information cards in chronological order, encapsulates stoma type, replacement, and treatment information with image numbers into a graphic view, and generates a continuous nursing record set.

[0012] As a further aspect of the present invention, the stoma care continuous record set includes stoma information cards, replacement record entries, processing record entries, and a graphic view structure.

[0013] As a further aspect of the present invention, the intermittent presentation module includes: The interval filtering submodule obtains all record nodes in the collaborative information group, filters all record numbers within the time interval selected by the user, sorts the records that meet the conditions in ascending order according to the time of change, retains the corresponding patient identity and image number, and generates a time interval sequence set. The information assembly submodule extracts the stoma type, replacement information, processing content and image number corresponding to the record based on the time segment sequence set, combines the four contents into a structured field group, writes it into the card generation template and adds the record timestamp information to generate a stoma information card group. The image and text encapsulation submodule calls the stoma information card group, merges all image and text information content according to the card generation time, formats and displays the combined card text and image information in a unified manner, establishes the image and text set corresponding to the card group, and generates a stoma care continuous record set.

[0014] The beneficial effects of the technical solution provided by this invention include at least the following: (1) In this invention, a time difference recognition mechanism is established by extracting the image generation time and submission time, data binding is completed by combining image number and identity information, semantic merging is achieved by integrating excrement description, product replacement and processing action content, consistency judgment logic of multi-source information is constructed, and the sequential structure is constructed based on the time field to ensure the integrity and accuracy of record connection. Based on the patient identity, the differences in the content of each field within the same time period are identified and collaborative identifiers are generated, which improves the correlation ability of multi-party data across time periods and across personnel. Finally, the images and nursing content are encapsulated according to the time sequence to form a continuous graphic and text view with clear structure, consistent content and comparability.

[0015] (2) By forming a continuous graphic view with clear structure, consistent content and comparability, this invention can better analyze the stoma care path, facilitate better stoma care, reduce the probability of stoma deterioration, avoid delaying treatment, reduce the workload of stoma nurses, detect problems in the early stage of stoma complications, provide early warning for patients, obtain timely and effective medical professional support, and greatly reduce the workload of clinical nursing staff. Attached Figure Description

[0016] 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.

[0017] Figure 1 This is a schematic diagram of the timeline-based ostomy treatment management system provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the system framework of the present invention; Figure 3 This is a diagram of the initial acquisition module in this invention; Figure 4 This is a diagram of the content organization module in this invention; Figure 5 This is a diagram of the sequential integration modules in this invention; Figure 6 This is a diagram of the identity verification module in this invention; Figure 7 This is a diagram of the discontinuous presentation module in this invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] 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.

[0020] 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, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] 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.

[0022] 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.

[0023] This invention provides a timeline-based stoma treatment management system, such as... Figure 1-2 The diagram shown illustrates a timeline-based ostomy care management system, which includes: The initial acquisition module obtains the stoma type, stoma location, stoma height, surrounding skin images, and image source device information input during the stoma patient care process. It extracts the image file generation time, calls the submission time for difference judgment, screens whether the stoma type and height fields are empty, matches and binds the image number with the data record identity, marks the time inconsistency status for time fields with differences, and generates and organizes graphic and textual data. The content organization module calls and organizes graphic and textual data, and collects image number, product replacement content, excrement description, treatment action, treatment location and executor information. It compares the excrement description with a fixed set of descriptive words to determine content consistency, merges the image number with the corresponding replacement content and treatment action fields, performs source consistency matching operation on the treatment location and executor fields, and generates a unified description entry group. The sequential integration module extracts the replacement time, image generation time, and processing record time from the unified description entry group, arranges them in chronological order, writes an image lag flag for records whose image time appears after the replacement time, writes a sequence correction flag for records whose processing record time is earlier than the replacement record time, organizes all time fields to form a complete sequential structure, and generates a sorted time series. The identity verification module references patient identity tag records in a sorted time series, compares the product name, replacement time and excrement description filled in by the patient in the same time period, judges the content differences between the fields, sets a collaborative record identifier when any field is inconsistent, creates a field group to be verified for the two groups of content and retains the identity tag, and generates a collaborative information group. The intermittent presentation module filters all record nodes within the user-selected time interval from the collaborative information group, generates stoma information cards in chronological order, and combines and encapsulates the stoma type, replacement information, treatment content and image number in the card content into a continuous graphic page view structure to generate a continuous stoma care record set.

[0024] The data organization includes basic stoma information, image information, time status markers, and identity binding relationships. The unified description item group includes product replacement information, excrement characteristics, treatment behavior elements, and performer and location elements. The sorted time series includes replacement time series, image time series, treatment time series, and sequence identifiers. The collaborative information group includes patient identity elements, product record elements, field difference elements, and verification field groups. The stoma care continuous record set includes stoma information cards, replacement record items, treatment record items, and graphic view structure.

[0025] Specifically, such as Figure 2 , 3 As shown, the initial acquisition module includes: The image acquisition submodule obtains the image source device information and image file generation time, acquires surrounding skin images, extracts the generation time parameter and submission time parameter recorded in the image file, compares the difference between the two values, determines whether the difference exceeds the image submission time difference threshold, and generates a time consistency label. First, the unique identification number of the acquisition device is read. This number identifies the device. Second, the generation time parameter is retrieved from the image file's metadata. This time records the precise moment the image was captured by the device. Simultaneously, the submission time parameter is read from the submitting system. This time represents the actual moment the image was uploaded or submitted. Both are then converted to hours, minutes, and seconds format for second-level difference calculation. For example, if the generation time is 10:30:25 and the submission time is 10:35:40, the difference is 315 seconds. This time difference is then compared to a preset time difference threshold. The threshold can be set... The time was determined by repeatedly testing network latency and the time required for manual verification during actual data collection. For example, after 100 consecutive tests, the average latency was 180 seconds and the maximum latency was 280 seconds. To avoid misjudgment due to individual fluctuations, a certain safety range could be added to the maximum value, with 300 seconds as the threshold. When performing comparisons, if the time difference is less than or equal to 300 seconds, it is judged as time consistent; otherwise, it is judged as abnormal. In this example, 315 seconds is greater than 300 seconds, so it is judged as abnormal. The device number, generation time, submission time, time difference, and final judgment result are recorded, and a corresponding time consistency label is generated.

[0026] The field screening submodule performs null value detection on the input stoma type and stoma height fields based on time consistency labels, determines whether there is missing content in either of the two fields, classifies and labels the field status, and generates field integrity labels. Based on time consistency labels, the stoma type and stoma height fields are checked one by one. First, the content of the stoma type field is read. If the content is empty, only a blank space, or not filled, the field status is marked as missing and the mark value is set to 1. If there is valid content, it is marked as 0. Then, the stoma height field is read and checked and marked in the same way. For example, if the stoma type is filled in as "circular" in a record, the status of the field is marked as 0. If the stoma height is empty, the status is marked as 1, and finally a status combination (0, 1) is formed. Then, a field integrity label is generated according to a pre-set encoding rule. For example, in the encoding rule, 00 represents both items are complete, 01 represents missing height, 10 represents missing type, and 11 represents both items are missing. Then the combination corresponds to label 01. After the label is generated, it is also saved together with the time consistency label of the record. For example, if a record has a normal time consistency label and a field integrity label of 01, it will be recorded as a normal time but missing height status, which is convenient for subsequent classification processing.

[0027] The image binding submodule calls the field integrity label and image number to obtain the corresponding data record identity parameter, performs a one-to-one matching and binding of the image number and data record identity, performs unified labeling on records with abnormal time consistency labels during the binding process, and generates organized image and text data. The process involves retrieving field integrity labels and image ID information. First, the data record table searches for a record matching the image ID, retrieving the corresponding identity information, such as patient ID, user ID, or other identity parameters. Then, a one-to-one binding relationship is established between the image ID and the identity information. During this process, the time consistency label of the record is also read. If the time consistency label value is abnormal, an abnormal flag is added to the binding result and set to 1; otherwise, it is set to 0. For example, if a record has an image ID of "Image 103025", identity information of "ID 12345", and a time consistency label of abnormality, then the binding result would be "Image 103025, ID 12345, Abnormal Flag 1". After binding, the field integrity label and abnormal flag are also recorded, ultimately forming a well-organized image and text data set. Each record in this set contains an image ID, identity information, time consistency label, field integrity label, and abnormal flag, facilitating subsequent data management and retrieval.

[0028] Specifically, such as Figure 2 , 4 As shown, the content organization module includes: The description verification submodule acquires and organizes image and text data, collects excrement description content, matches and compares each excrement description with each term in a fixed set of descriptive terms, calculates the average matching deviation value of the descriptive terms, and generates excrement description consistency tags. The formula for calculating the average matching deviation of descriptive terms is as follows: ; in, This represents the average matching deviation value for all descriptive terms. Representing the The difference in character position between the excrement description term and the corresponding fixed description term, r i Representing the The difference in character length between the excrement description entry and the corresponding fixed description entry. Representing the The character difference weight value between the excrement description term and the corresponding fixed description term. Representing the The total character encoding of the excrement description entries. Representing the The sum of character codes for a fixed descriptive term. Representing the The importance weighting coefficient of matching descriptive terms. To avoid division by zero for the stable term in the denominator, The total number of descriptive terms that were matched; In this embodiment, the introduced symbols and operational logic are explained as follows: "in, Indicates the first Item description: accumulation of entries; parameters (Character position difference) and r i (Character length difference) is expressed by the plus sign " "Combination" is used to reflect the combined effect of different sources of deviation, followed by a minus sign. "with weight" The difference is compared to measure the actual deviation magnitude; the absolute value symbol "" is used. "Obtain a positive value for the deviation to ensure that the direction of the deviation does not affect the calculation; multiplication sign" "This deviation amount is compared with the matching importance weight" Multiplication is used to reflect the importance and contribution of different items; the denominator is first subtracted by a minus sign. "Calculate the difference between the sums of character encodings" Then through the flat method (" "" (that is, exponentiation) means finding the square or sum of squares, which implies the Euclidean norm, using the square root sign " "Calculate the magnitude of the total coding differences; add the final item" Minimal Stable Term To avoid division by zero, the ratio of numerator deviation to denominator coding difference is calculated using the division sign " / ". This quantifies the matching deviation; the smaller the relative coding difference and the larger the deviation, the larger the ratio, indicating a decrease in matching accuracy. Example of parameter acquisition and quantization process, and data assignment and calculation example: Assuming it is used for consistency assessment of excrement description, for The project was demonstrated, and it strictly followed the actual monitoring and quantification process.

[0029] Non-numerical data quantization process: The calculation method is the absolute value of the difference between the indices of the input word and its index in the standard description list. For example, "Sample A" has a standard index of 8, and the input description index is 10. .

[0030] : The absolute value of the difference in string length. For example, the standard "sample A" has a length of 3, while the input "sample AB" has a length of 4. .

[0031] Edit distance normalization value, calculated as "Levenshtein distance ÷ maximum possible distance". For example, edit distance 1, maximum distance 5 → .

[0032] , : The sum of character encodings, which is the sum of the Unicode values ​​of each character. For example, the character encoding of "sample A" is 30000 + 23567 + 65 = 535632.

[0033] The importance weight of each item is determined by domain experts; for example, key items are assigned a value of 1.4, and ordinary items are assigned a value of 1.0.

[0034] Setting the minimum stable term to 0.01 is sufficient to avoid a zero denominator, but does not significantly change the result. Table 1: Parameter Values ​​and Sources Explanation. As shown in Table 1, all parameters have clear sources. The index and length are obtained from the text position and character count; the edit distance standardization is derived from character edit comparison; the encoding sum is obtained by accumulating the character encodings; and the weights are evaluated by experts. The default value is extremely small.

[0035] Substitute into the formula to calculate The formula is: ; molecular: Item 1: ; Item 2: ; Total numerator = ; Denominator: Difference between the two codes: , →Sum of squares=0; If the square root is followed by 0, add... →Denominator = 0.01; Final calculation: ; The result indicates that the calculated average matching deviation The error is much higher than the preset "matching deviation acceptable threshold". For example, the system standard threshold is set to 50 (the actual experience range is set to 30-100). This indicates that the matching deviation of the current description item is too large, and further adjustment of data input or rematching strategy is required.

[0036] The advantage of the formula lies in introducing matching importance weights. The impact of deviations in key items is significantly amplified, making the system more sensitive to the accuracy of key descriptions, thereby improving the recognition accuracy and practicality in consistency evaluation.

[0037] The content merging submodule, based on the excrement description consistency tag, calls the image number, product replacement content and processing action field information, performs a merging judgment on the product replacement content and processing action fields corresponding to the same image number, merges the information of fields with consistent content under the same image number, and generates the content group corresponding to the image number. After obtaining the excrement description consistency label, the system retrieves the image number, product replacement content, and treatment action fields corresponding to each record. Then, the records are grouped according to the image number. For example, if three records have the same image number, "Image A," they are grouped together. Next, within the same image number group, the text content of the product replacement content field is compared for complete consistency. If they are identical, they are considered consistent. Then, the text content of the treatment action field is compared for complete consistency. If they are identical, they are also considered consistent. When two fields are both considered consistent within the same group, the corresponding field content of these records is merged. For example, if three records all have the product replacement content "replace ostomy bag" and the treatment action "clean surrounding skin," the values ​​of these two fields in these three records are merged into the value of the corresponding field in one record. If a record has a different product replacement content, such as "replace dressing," then that field is not merged; only the difference is retained. The merging criterion is based on completely identical text matching, not fuzzy matching. The merged field results for each image number form the content group corresponding to that image number.

[0038] The information matching submodule calls the processing location and executor information according to the content group corresponding to the image number, determines whether the source identifier of the processing location and executor fields in the same record is consistent, classifies and integrates items with consistent sources, and generates a unified description entry group. After the image number corresponds to a content group, the processing location and executor information related to that content group are read. First, the source identifier of the processing location field in the same record is determined. For example, the source identifier could be the data input terminal number, collection form number, etc. Then, the source identifier of the executor field is determined. The source identifier format is consistent with the processing location field. Next, the source identifiers of the two are compared. If the source identifiers are the same, it is determined that the source is consistent and recorded. For example, if the processing location source identifier of a record is "Terminal 05" and the executor source identifier is "Terminal 05", then the source is consistent. If the processing location source identifier is "Terminal 05", then the source is consistent. If the executor source identifier is "Terminal 05", then the source is inconsistent. All records with consistent sources are grouped into the same set to form a consistent source set, while records with inconsistent sources are grouped into another set. For each record in the consistent source set, the text content of its processing location and executor fields is extracted. These contents are integrated into a unified description entry group according to the record order. For example, if multiple records have the processing location as "Ward Group 1" and the executor as "Nurse A", they are integrated into "Ward Group 1 - Nurse A". These unified entry groups are then bound to the corresponding image numbers to generate a unified description entry group.

[0039] Specifically, such as Figure 2 , 5 As shown, the sequential integration module includes: The time extraction submodule obtains the image number from the unified description entry group, extracts the corresponding replacement time, image generation time and processing record time, classifies and stores the three time information according to the record number, removes records with missing time fields, and generates a time field set. After obtaining the image number from the unified description entry group, the system first reads the replacement time field corresponding to that number in the data record and converts it to a unified hour, minute, and second format. Then, it reads the image generation time field, which is directly extracted from the image file's metadata and also uniformly formatted as hour, minute, and second. Next, it reads the corresponding processing record time field, which comes from the submission time of the operation record form. After ensuring that all three times are stored in the same format, the three times are grouped into the same data according to the record number. For example, if the replacement time of record number 001 is 10:05:10, the image generation time is 10:06:20, and the processing record time is 10:07:00, then these three times are grouped together. Then, a missing value check is performed on all records. If any of the three times is empty, only a space, or NULL, then the record is determined to have a missing time field and is removed from the set. For example, if record number 003 is missing the processing record time, then the entire record will not be included in the subsequent set. Finally, the remaining records store their three time information in numerical order to form a time field set.

[0040] The sequence judgment submodule judges the sequence of image generation time and replacement time based on the time field set. If the image time is later than the replacement time, it marks the image as lagging. At the same time, it compares the processing record time with the replacement time. If the processing time is earlier than the replacement time, it writes the sequence correction flag and generates a time anomaly flag set. Based on the time field set, the image generation time and replacement time of each record are first compared sequentially. The image time and replacement time are converted into total seconds. For example, the replacement time of 10:05:10 is converted into 36310 seconds, and the image generation time of 10:06:20 is converted into 38180 seconds. If the total number of seconds for the image generation time is greater than the total number of seconds for the replacement time, an image lag flag is written for the record, with a value of 1; otherwise, it is 0. Then, the processing record time and replacement time are compared sequentially. The processing record time is also converted into total seconds. If the number of seconds is less than the number of seconds for the replacement time, it is determined that the processing is premature, and an order correction flag is written, with a value of 1; otherwise, it is 0. For example, if a record has a replacement time of 36000 seconds, an image generation time of 37000 seconds, and a processing record time of 35000 seconds, then the image lag flag is 1, and the order correction flag is 1. These two flags are bound to the record number. After all records are processed, a time anomaly flag set is formed.

[0041] The sequential structure submodule calls the time anomaly identifier set and the time field set, sorts all records in ascending order according to the replacement time, processing time and image generation time, and organizes and integrates the time and identifier corresponding to each record to generate a sorted time series. The system calls the time anomaly flag set and the time field set. Each record in the set is sorted based on the replacement time, processing time, and image generation time. The replacement time in seconds is used as the first sorting key. If the values ​​are the same, the processing time is used for sorting; if still the same, the image generation time is used for sorting. The sorted results are arranged in ascending order. Each record's three time parameters and corresponding anomaly flag are then processed and combined into a sequential structure entry. For example, record number 002 has a replacement time of 36000 seconds, a processing time of 36100 seconds, an image generation time of 36200 seconds, an image lag flag of 0, and a sequence correction flag of 0. Therefore, it lists the following in chronological order: "Replacement time 36000 seconds, processing time 36100 seconds, image time 36200 seconds, lag 0, correction 0". After processing all records sequentially, the results are integrated into a sorted time series.

[0042] Specifically, such as Figure 2 , 6 As shown, the identity verification module includes: The identity extraction submodule obtains the record content in the sorted time series, extracts the patient identity tags in the records, filters the product name, replacement time and excrement description fields from the records filled in by the same patient in the same time period, numbers and archives each group of information according to the patient identity, and generates the same patient field group. After retrieving the records from the sorted time series, the system sequentially reads the patient identification tag for each record. This tag is a unique string or number identifying the patient, such as "Patient A001". After obtaining the identification tag, records entered by that patient within the same time period are filtered out. This involves accessing the change time field, product name field, and excrement description field from the records. The determination of the same time period is based on the change time, grouping records whose change times differ by no more than a set range into the same time period group. For example, if the range is set to 5 minutes (300 seconds), when two records change... If the timestamps are 10:05:00 and 10:08:10 respectively, they are grouped into the same time period. Then, extract the product name, replacement time, and excrement description from each record within this time period. Combine this information into a set of field data. Finally, number and archive these field groups according to the patient's identity tag. For example, if patient A001 has two records in this time period, namely "Ostomy bag A, 10:05:00, brownish color, thick texture" and "Ostomy bag A, 10:08:10, brownish color, thick texture", these two records are merged and archived into the same patient field group.

[0043] The content comparison submodule is based on the same patient field group, calls the corresponding fields in multiple groups, compares the product name, replacement time and excrement description content. When any content in the field is inconsistent, it is recorded as a field difference item. The number associated with the field difference item is marked with a collaborative record identifier, and a field difference identifier group is generated. After obtaining the patient field groups for the same segment, the product name, replacement time, and excrement description fields are retrieved group by group. A one-to-one comparison is performed on each field within the same group. The comparison first checks if the product name field is completely identical. If a record's product name differs from other records (e.g., one record says "Ostomy Bag A" and another says "Ostomy Bag B"), this field is identified as a difference and recorded. Next, the replacement time field is checked, converted to seconds, and then compared precisely. For example, a record of 10:05:00 corresponds to 36000 seconds. One entry corresponds to 10:06:00 and 36360 seconds. If the values ​​are not equal, it is considered a discrepancy. Finally, the excrement description field is checked, and the text content is judged to be consistent. For example, if one entry is "brownish in color and thick in texture" and another entry is "light yellow in color and thin in texture", then this field is a discrepancy. The detected discrepancies are bound to the corresponding patient identity tags, and collaborative record tags are added to the record numbers of these discrepancies. For example, if there is a discrepancy between numbers 001 and 002, then both numbers are marked with collaborative record tags. All discrepancies and their tags form a field discrepancy tag group.

[0044] The difference annotation submodule extracts the record numbers with collaborative record identifiers based on the field difference identifier group, establishes a structure for storing the two field groups in the field difference item separately, retains the corresponding identity tags, and uniformly classifies each group of records into a structured collection to generate a collaborative information group. After obtaining the field difference identifier group, all record numbers with collaborative record identifiers are extracted. For example, if the identifier group contains numbers 001, 002, and 005, the corresponding field difference items for these records are read sequentially. The content of the same field name in the difference items is stored in two independent field group structures. For example, the difference content of the product name field is stored in the product name field group, and the difference content of the excrement description field is stored in the excrement description field group. While storing, the original patient identity label is retained for each record to ensure that the source information of the difference fields is clear. For example, if the product names of numbers 001 and 002 are different, the two names are respectively entered into the product name field group with the identity label A001. If the excrement descriptions are different, the two description contents are respectively entered into the excrement description field group with the same identity label. After all the difference content is stored, these field groups are uniformly grouped into a structured set. Each record in the set contains the patient identity, difference field name, difference field content, and collaborative record identifier, thereby generating a collaborative information group.

[0045] Specifically, such as Figure 2 , 7 As shown, the intermittent presentation module includes: The interval filtering submodule obtains all record nodes in the collaborative information group, filters all record numbers within the time interval selected by the user, sorts the records that meet the conditions in ascending order according to the time of change, retains the corresponding patient identity and image number, and generates a time interval sequence set. After acquiring all record nodes in the collaborative information group, the system first reads the change time of each record and converts it into a numerical format of hours, minutes, and seconds. For example, 10:15:20 is converted to 36920 seconds. Then, the system extracts record numbers based on the user-defined time interval. The start and end values ​​of the time interval are directly input by the user. For example, the start time is 36000 seconds and the end time is 40000 seconds. The system then filters out record numbers whose change times fall within this range. During the filtering process, the system checks whether the change time value of each record meets the condition that the start value is less than or equal to the change time and the change time is less than or equal to the end value. For example, if a record's change time is 37000 seconds, it meets the condition. The filtered record numbers then proceed to the subsequent sorting step. The records corresponding to these numbers are arranged in ascending order of change time value. For example, after sorting, the records would be numbered 005 (36500 seconds), 003 (37000 seconds), and 002 (39500 seconds). After sorting, the system retains the patient identity information and image number corresponding to each record and combines them according to the sorting results to finally form a time segment sequence set.

[0046] The information assembly submodule extracts the stoma type, replacement information, processing content and image number corresponding to the record based on the time segment sequence set. It combines the four contents into a structured field group, writes them into the card generation template and adds the record timestamp information to generate a stoma information card group. After obtaining the time-segment sequence set, each record number is read sequentially, and the corresponding stoma type, replacement information, processing content, and image number fields are extracted. These four items are combined into a structured field group. For example, if the stoma type is "circular," the replacement information is "replace stoma bag A," the processing content is "clean the skin around the stoma," and the image number is "IMG_105020," then the combination would be [circular, replace stoma bag A, clean the skin around the stoma, IMG_105020]. Then, each structured field group is... The field group adds record timestamp information, which comes from the replacement time field of the record. For example, the timestamp corresponding to the replacement time of 37000 seconds is 10:16:40. This timestamp is inserted into the structured field group to form a complete record unit. Then, all record units are uniformly written into the preset card generation template in the order of the time segment sequence set. The fields in the card template are fixed as ostomy type, replacement information, processing content, image number, and timestamp, and maintain a uniform layout format. After all record units are written into the template, an ostomy information card group is generated.

[0047] The image and text encapsulation submodule calls the stoma information card group, merges all image and text information content according to the card generation time, formats and displays the combined card text and image information in a unified manner, establishes the image and text set corresponding to the card group, and generates a stoma care continuous record set. After acquiring the stoma information card set, the cards are arranged chronologically from earliest to latest based on their creation time. Cards from adjacent time periods are merged into a single group. The text and image information are then extracted from each group. The text information is formatted into a fixed paragraph structure; for example, the first line contains stoma type and replacement information, the second line contains the treatment details, and the third line records the timestamp. Images are arranged according to pre-defined display positions in a template, ensuring consistent text and image placement across all cards in the same group. The combined text and images are then packaged into image-text entries, which are sequentially added to the same set. Each set corresponds to a consecutive set of card information, recording the card's chronological order and the original values ​​of all fields, ultimately forming a continuous stoma care record set. This continuous stoma care record set allows for better analysis of the stoma care path, facilitating better stoma care, reducing the probability of stoma deterioration, avoiding delays in treatment, and alleviating the workload of stoma caregivers. It also enables early detection of stoma complications, providing patients with early warnings and timely and effective medical support, and significantly reducing the workload of clinical nursing staff. 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 timeline-based ostomy treatment management system, characterized in that, The system includes: The initial acquisition module collects stoma type image information and equipment information, extracts and compares the image generation time with the submission time, judges the differences, verifies whether the type and height fields are empty, binds the image number and identity, marks inconsistent times, and generates and organizes image and text data. The content organization module is based on the organized graphic data and collects image number, product replacement, excrement description, processing action, location and executor. It compares the excrement description with the fixed word set, merges the images with the replacement and processing content, matches the location and executor source, and generates a unified description item group. The sequential integration module extracts the replacement time, image time, and processing time based on the unified description entry group, sorts them by time, marks the image as lagging if it is later than the replacement time threshold, and marks the processing as corrected if it is earlier than the replacement, thus generating a sorted time series. The identity verification module references the patient's identity in the sorted time series and compares it with the product name, replacement time, and excrement description filled in by the patient during the same period. If the fields are inconsistent, a collaborative identifier is set, a group of fields to be verified is established and the identity label is retained, and a collaborative information group is generated.

2. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The organized graphic data includes basic stoma information, image information, time status markers, and identity binding relationships. The unified description item group includes product replacement information, excrement characteristics, treatment behavior elements, and executor and location elements. The sorted time series includes replacement time series, image time series, treatment time series, and sequence identifiers. The collaborative information group includes patient identity elements, product record elements, field difference elements, and verification field groups.

3. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The replacement time threshold is used to determine the time difference boundary between the replacement behavior and the order of image and processing time. If the fields are inconsistent, a collaboration flag is set, indicating that when there are differences between the patient record and the collected information in key fields, it is marked as a state that requires collaborative verification.

4. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The initial acquisition module includes: The image acquisition submodule obtains information about the image source device and the image file generation time, acquires skin images, extracts the generation time parameter and submission time parameter recorded in the image file, compares the differences between the two values, determines whether the difference exceeds the image submission time difference threshold, and generates a time consistency label. Based on the time consistency label, the field screening submodule performs null value detection on the input stoma type and stoma height fields, determines whether there is any missing content in either of the two fields, classifies and labels the field status, and generates field integrity labels. The image binding submodule calls the field integrity label and image number to obtain the corresponding data record identity parameter, performs one-to-one matching and binding of the image number and data record identity, performs unified labeling on records with abnormal time consistency labels during the binding process, and generates organized image and text data.

5. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The content organization module includes: The description verification submodule acquires the organized graphic data and collects the excrement description content. It then matches and compares each excrement description with the terms in the fixed descriptive term set, calculates the average matching deviation value of the descriptive terms, and generates an excrement description consistency label. The content merging submodule, based on the excrement description consistency tag, calls the image number, product replacement content and processing action field information, performs a merging judgment on the product replacement content and processing action fields corresponding to the same image number, merges the information of fields with consistent content under the same image number, and generates the content group corresponding to the image number. The information matching submodule retrieves the processing location and executor information based on the content group corresponding to the image number, determines whether the source identifiers of the processing location and executor fields in the same record are consistent, and classifies and integrates items with consistent sources to generate a unified description entry group.

6. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The sequential integration module includes: The time extraction submodule obtains the image number in the unified description entry group, extracts the corresponding replacement time, image generation time and processing record time, classifies and stores the three time information according to the record number, removes records with missing time fields, and generates a time field set. The sequential judgment submodule judges the sequence of image generation time and replacement time based on the time field set. If the image time is later than the replacement time threshold, an image lag indicator is marked. At the same time, the processing record time is compared with the replacement time order. If the processing time is earlier than the replacement time threshold, an order correction flag is written and a time anomaly indicator set is generated. The sequential structure submodule calls the time anomaly identifier set and time field set to sort all records in ascending order based on replacement time, processing time and image generation time. It then organizes and integrates the time and identifier corresponding to each record to generate a sorted time series.

7. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The identity verification module includes: The identity extraction submodule obtains the record content in the sorted time series, extracts the patient identity tags in the records, filters the product name, replacement time and excrement description fields from the records filled in by the same patient in the same time period, numbers and archives each group of information according to the patient identity, and generates the same patient field group. The content comparison submodule, based on the same patient field group, calls the corresponding fields in multiple groups, compares the product name, replacement time and excrement description content, and records any inconsistency in any field as a field difference item. The number associated with the field difference item is marked with a collaborative record identifier to generate a field difference identifier group. The difference labeling submodule extracts the record number with collaborative record identifier based on the field difference identifier group, establishes a structure for storing the two field groups in the field difference item respectively, retains the corresponding identity label, and uniformly classifies each group of records into a structured set to generate a collaborative information group.

8. The timeline-based ostomy treatment management system according to claim 1, characterized in that: The system also includes an intermittent presentation module: The intermittent presentation module, based on the collaborative information group, filters records for the selected time period, generates stoma information cards in chronological order, encapsulates stoma type, replacement, and treatment information with image numbers into a graphic view, and generates a continuous nursing record set.

9. The timeline-based ostomy treatment management system according to claim 8, characterized in that: The stoma care continuous record set includes stoma information cards, replacement record entries, treatment record entries, and a graphic view structure.

10. The timeline-based ostomy treatment management system according to claim 8, characterized in that: The intermittent presentation module includes: The interval filtering submodule obtains all record nodes in the collaborative information group, filters all record numbers within the time interval selected by the user, sorts the records that meet the conditions in ascending order according to the time of change, retains the corresponding patient identity and image number, and generates a time interval sequence set. The information assembly submodule extracts the stoma type, replacement information, processing content and image number corresponding to the record based on the time segment sequence set, combines the four contents into a structured field group, writes it into the card generation template and adds the record timestamp information to generate a stoma information card group. The image and text encapsulation submodule calls the stoma information card group, merges all image and text information content according to the card generation time, formats and displays the combined card text and image information in a unified manner, establishes the image and text set corresponding to the card group, and generates a stoma care continuous record set.

Images