A settlement-invoice-based grouping recommendation system and method

Abstract

The present application relates to a kind of settlement based on grouping recommendation system and method of list, belong to DRG grouping technical field.The present application discloses grouping recommendation system includes in situ fuzzy grouping module, ectopic fuzzy grouping module, QY (ambiguity) grouping module, missing code recommendation module, grouping fit degree module, and grouping recommendation method steps are as follows: in situ fuzzy grouping is carried out, and grouping range is expanded;Ectopic fuzzy grouping is carried out, and grouping range is further expanded;QY (ambiguity) grouping supplement is carried out, and grouping is carried out with the main diagnosis code of surgical coding collocation, obtains the possible existence of real grouping;Missing code grouping supplement is carried out, and in situ fuzzy grouping is carried out, obtains the possible existence of real grouping;The fit degree of S1-S4 all groupings is calculated respectively, and grouping fit degree value is arranged in descending order form and is arranged in list.The present application can recommend the most suitable grouping to doctor by analyzing the position of ICD coding, classification relationship.

Description

Technical Field

[0001] This invention relates to the field of DRG grouping technology, and more specifically to a grouping recommendation system and method based on ICD codes on hospital billing statements. Background Technology

[0002] Diagnosis Related Groups (DRGs) are a patient-centered case combination system and an effective tool for controlling healthcare costs, refining hospital management, and evaluating medical outcomes. Their primary data source is the billing statement. The International Classification of Diseases (ICD) is an international standard for coding and classifying different types of diseases and health-related issues. Key techniques for DRG enrollment include ensuring the accuracy of ICD codes, their correct placement, and the absence of omissions.

[0003] In current practice, the following problems exist:

[0004] 1. Most hospitals lack a deep understanding of ICD coding and do not have dedicated coders; ICD codes are filled out solely by clinicians. During the filling process, clinicians are easily influenced by clinical rules such as clinical specialties, laboratory results, and medical records, neglecting the principle of "resource efficiency" that ICD coding should follow. This leads to incorrect ICD code selection, and may specifically result in the following common misunderstandings leading to enrollment errors:

[0005] (1) A major surgery was performed during hospitalization, accompanied by an internal medicine disease. Since the final discharge department was internal medicine, the internal medicine disease was selected as the primary diagnosis, which resulted in the inability to enter the surgical group.

[0006] (2) A major surgery was performed during hospitalization, but according to the operation sequence, the surgery code with less resource consumption, such as "oxygen inhalation", was selected as the main surgery, which resulted in the inability to enter the surgical group;

[0007] (3) The code was selected as the primary diagnosis or primary surgery based solely on clinical understanding. Because the code lacked key information such as approach, material, and location, it was not recognized by the grouping scheme, resulting in the inability to be correctly enrolled.

[0008] 2. Most hospital coders' understanding of ICD coding remains limited to the rules on the medical record cover sheet. They fail to fill in ICD codes according to the specifications for completing the billing statement, leading to coding errors and inability to enroll patients. A common example is that the medical record cover sheet rules do not allow the selection of direct symptoms leading to death, such as "heart failure," as the primary diagnosis, while some groups in the national grouping scheme support and explicitly require the use of such primary diagnoses for correct enrollment.

[0009] 3. Most of the hospital information systems do not maintain special ICD codes of National Medical Insurance Version 2.0 for settlement list, but use a mapping method to map the hospital internal codes or national clinical version codes to the ICD codes of National Medical Insurance Version 2.0, and in the mapping process, there may be information loss or information mismatch, mapping to a coding with inconsistent connotation or a coding not recognized by the grouping scheme, resulting in incorrect grouping.

[0010] 4. The existing ICD coding and grouping scheme have weak relevance, and the grouping scheme is relatively complex, so that the hospital staff cannot intuitively determine whether the existing settlement list is correctly grouped, and cannot intuitively determine how to increase, delete or modify the ICD coding to correctly group. SUMMARY

[0011] In view of the low grouping accuracy in the prior art, the present application provides a grouping recommendation method based on a settlement list, which analyzes the position and classification relationship of the ICD coding and recommends the most suitable grouping to the doctor.

[0012] The present application also provides a grouping recommendation system based on a settlement list, comprising an in-situ fuzzy grouping module, an ectopic fuzzy grouping module, a QY (ambiguity) grouping module, a missing coding recommendation module and a grouping fitness module.

[0013] The in-situ fuzzy grouping module is used for grid search grouping of all ICD codes under the same classification of each ICD code. The diagnostic coding classification information is obtained according to the subcategory classification standard of the International Classification of Diseases and Related Health Problems ICD-10, and the diagnostic coding includes the primary diagnostic coding and the secondary diagnostic coding; the surgical coding classification information is obtained according to the detailed item classification standard of the International Classification of Diseases Ninth Revision Clinical Modification Surgery and Procedures ICD-9-CM-3, and the surgical coding includes the primary surgical coding and the secondary surgical coding.

[0014] The ectopic fuzzy grouping module is used for changing the position of the aforementioned ICD coding and grouping. The position refers to the field where the ICD coding as a value is located, and there are four fields of primary diagnosis, secondary diagnosis, primary surgery and secondary surgery. The primary diagnosis refers to the disease (or health condition) that causes the patient to be hospitalized for treatment during this hospitalization; the secondary diagnosis refers to the disease that coexists with the patient during hospitalization, occurs later or affects the treatment and hospitalization time; the primary surgery refers to the surgery or operation performed on the patient during this hospitalization for the disease diagnosed by the clinician; and the secondary surgery refers to other surgeries or operations performed on the patient during this hospitalization.

[0015] QY (ambiguity) grouping module: grouping with the main diagnosis code matched with the surgery code, this module is used to supplement the grouping of possible QY cases, to make up for the inaccuracy of grouping caused by the absence of the main diagnosis, QY cases refer to cases that cannot be grouped due to the mismatch between the main diagnosis and the main surgery.

[0016] Missing code recommendation module: used to supplement the missing main surgery code and perform in-situ fuzzy grouping.

[0017] Grouping fit module: used to calculate the fit of all groups respectively, and arrange the group fit values in descending order into a list, and output the list, the list is the recommended grouping list from high to low in recommendation degree.

[0018] A grouping recommendation method based on the settlement list of the application has the following steps:

[0019] S1, in-situ fuzzy grouping, expand the grouping range: obtain the codes of the main diagnosis, secondary diagnosis, main surgery and secondary surgery in the settlement list data for grouping, and obtain the possible real grouping;

[0020] S2, perform ex-situ fuzzy grouping, further expand the grouping range: exchange the positions of the main diagnosis and secondary diagnosis codes in the settlement list data, and exchange the positions of the main surgery and secondary surgery codes, change the positions and perform in-situ fuzzy grouping, obtain the possible real grouping; wherein, the position refers to the field where the ICD code as a value is located, there are four fields of main diagnosis, secondary diagnosis, main surgery and secondary surgery, the main diagnosis refers to the disease (or health condition) that causes the patient to be hospitalized for treatment during this hospitalization; the secondary diagnosis refers to the disease that coexists with the patient during hospitalization, occurs later, or affects the treatment and hospitalization time; the main surgery refers to the surgery or operation performed on the patient during this hospitalization for the disease diagnosed by the clinician; the secondary surgery refers to other surgeries or operations performed on the patient during this hospitalization;

[0021] S3, QY (ambiguity) grouping supplement, grouping with the main diagnosis code matched with the surgery code, making up for the inaccuracy of grouping caused by the absence of the main diagnosis, obtaining the possible real grouping;

[0022] S4, missing code grouping supplement, and in-situ fuzzy grouping, obtaining the possible real grouping;

[0023] S5, calculate the fit of all groups S1-S4 respectively, and arrange the group fit values in descending order into a list, and output the list, the list is the recommended grouping list from high to low in recommendation degree.

[0024] The details of each step of S1-S5 are described below.

[0025] S1 in-situ fuzzy grouping specifically includes the following steps:

[0026] S11: Create an empty set drgSet;

[0027] S12: Obtain a set of diagnosis codes in the settlement list data that are the same as the main diagnosis subentry, and make it set dx, the elements of the set are all diagnosis codes that are the same as the main diagnosis subentry;

[0028] S13: Obtain a set of diagnosis codes in the settlement list data that are the same as any one of the primary diagnosis subentry, and make it set dxMinor, the elements of the set are all diagnosis codes that are the same as any one of the primary diagnosis subentry;

[0029] S14: Obtain a set of surgery codes in the settlement list data that are the same as the main surgery subentry, and make it set tx, the elements of the set are all diagnosis codes that are the same as any one of the main surgery subentry;

[0030] S15: Obtain a set of surgery codes in the settlement list data that are the same as any one of the primary surgery subentry, and make it set txMinor, the elements of the set are all diagnosis codes that are the same as any one of the primary surgery subentry;

[0031] S16: Perform grid search grouping according to dx, dxMinor, tx, and txMinor, that is, perform exhaustive combination of all elements in the above four sets, obtain all disease diagnosis related groups (DRGs) that can enter according to the grouping scheme, and combine all obtained groups into the drgSet set, the elements of the set are all disease diagnosis related groups (DRGs) obtained in this process.

[0032] The steps of S2 ex-situ fuzzy grouping are:

[0033] Exchange the position of each secondary diagnosis of all disease diagnosis related groups (DRGs) in the drgSet set with the main diagnosis, and exchange the position of each secondary surgery with the main surgery, perform S1 in-situ fuzzy grouping process once each time, obtain a disease diagnosis related group (DRG), and combine these groups into the drgSet set.

[0034] S3 QY grouping specifically includes the following steps:

[0035] S31: Determine whether there is a valid main surgery code in the settlement list, and whether there is a corresponding surgery group in drgSet, if yes, continue to the next step, otherwise stop;

[0036] S32: If the primary diagnosis and secondary diagnosis conditions in the core diagnosis related group (ADRG) of each group in the set tx of surgery groups corresponding to the valid primary surgery code meet the group entry conditions of the ADRG in the grouping scheme, remove the disease diagnosis related group (DRG) that meets the group entry conditions of the ADRG in the settlement list from the group entry conditions related to the primary surgery, primary diagnosis and secondary diagnosis to form a set, the elements of this set are disease diagnosis related groups (DRGs), perform a set union operation on this set and drgSet, and assign the result of the union operation to drgSet.

[0037] S4 missing code grouping specifically includes the following steps:

[0038] S41: Determine whether the settlement list contains a valid surgery code. If not, continue. If yes, terminate;

[0039] S42: According to the charge item details and the valid surgery code table, match the most effective surgery code details;

[0040] S43: Obtain all surgery code sets of the same item as the aforementioned surgery code from the ICD code table (International Classification of Diseases) to become set txQY, the elements of this set are all surgery codes of the same item as the aforementioned surgery code;

[0041] S44: Place txQY in the primary surgery position of the settlement list, and execute the in-place fuzzy grouping module (S11-S16), perform a set union operation on the output set and drgSet, and assign the operation result to drgSet.

[0042] S5 all-group goodness of fit calculation includes the following steps:

[0043] S51: Obtain the total case cost c0 and case hospitalization days p0 in the settlement list

[0044] S52: For each group in drgSet in the aforementioned S1-S4, obtain the average cost c and average hospitalization days p of the group from the payment parameter table;

[0045] S53: Calculate the distance index q',

[0046]

[0047] S54: Map the distance index to a scalar q between 0 and 1, which is the goodness of fit. The specific mapping method is:

[0048]

[0049] Sort the packets in drgSet in q descending order to form a list, and output the list. The list is a recommended packet list arranged from high to low in recommendation degree.

[0050] The beneficial effects of the present application are:

[0051] 1. Reduce grouping errors caused by doctors being limited by clinical rules such as clinical specialties, test results, medical records, etc.

[0052] 2. Reduce coding errors that prevent grouping because coders are limited by the rules of the front page of the medical record and do not fill in the ICD code according to the filling specifications of the settlement list.

[0053] 3. Reduce grouping errors caused by doctors filling out the settlement list using different versions of ICD codes.

[0054] 4. Provide a groupable set, and each group provides a grouping fit degree for doctors to judge the most suitable group. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 is a recommended system block diagram of the present application.

[0056] Figure 2 is a grouping recommendation method flowchart of the present application. DETAILED DESCRIPTION

[0057] Figure 1 shows the recommended system block diagram of the present application, Figure 2 shows the grouping recommendation method of the present application.

[0058] A grouping recommendation system based on the settlement list of the present application includes an in-situ fuzzy grouping module, an ectopic fuzzy grouping module, a QY (ambiguity) grouping module, a missing code recommendation module, and a grouping fit degree module, wherein:

[0059] The in-situ fuzzy grouping module is used to perform grid search grouping on all ICD codes under the same classification of each ICD code. The diagnostic code classification information is obtained according to the subcategory classification standard of the International Classification of Diseases and Related Health Problems ICD-10, and the diagnostic code includes the primary diagnostic code and the secondary diagnostic code; the surgical code classification information is obtained according to the detailed item classification standard of the International Classification of Diseases Ninth Revision Clinical Modification Surgery and Operation ICD-9-CM-3, and the surgical code includes the primary surgical code and the secondary surgical code.

[0060] Etiological ambiguity grouping module: used to change the position of the aforementioned ICD code and grouping. Wherein, the position refers to the field in which the ICD code as a value is located, there are four fields of primary diagnosis, secondary diagnosis, primary operation, secondary operation, primary diagnosis refers to the disease (or health condition) that causes the patient to be hospitalized for medical treatment during this period; secondary diagnosis refers to the disease that coexists with the patient during hospitalization, occurs later, or affects the treatment received and the length of hospital stay; primary operation refers to the operation or operation performed on the patient during this hospitalization period for the disease diagnosed by the clinician; secondary operation refers to other operations or operations performed on the patient during this hospitalization.

[0061] QY (ambiguity) grouping module: grouping with the primary diagnosis code matched with the operation code, this module is used to supplement the grouping of QY cases that may exist, to make up for the inaccuracy of grouping caused by the absence of primary diagnosis, QY cases refer to cases that cannot be grouped due to the mismatch between primary diagnosis and primary operation.

[0062] Missing code recommendation module: used to supplement the missing primary operation code and perform in-situ fuzzy grouping.

[0063] Grouping fit module: used to calculate the fit of all groups respectively, and arrange the grouping fit values in descending order into a list, and output the list, the list is the recommended grouping list from high to low.

[0064] A grouping recommendation method based on the settlement list of the present application, before initialization, the primary diagnosis code, secondary diagnosis code set, primary operation code, secondary operation code set, age, gender, hospitalization days, and charge item details of the settlement list are obtained.

[0065] The grouping recommendation method steps are as follows:

[0066] According to the obtained data, the in-situ fuzzy grouping module is used to perform in-situ fuzzy grouping, including the following steps:

[0067] S11: create an empty set drgSet;

[0068] S12: obtain the diagnosis code set same as the primary diagnosis subentry in the settlement list data, become set dx, the elements of the set are all diagnosis codes same as the primary diagnosis subentry;

[0069] S13: obtain the diagnosis code set same as any secondary diagnosis subentry in the settlement list data, become set dxMinor, the elements of the set are all diagnosis codes same as any secondary diagnosis subentry;

[0070] S14: Obtain the set of surgery codes in the settlement list data that are the same as the main surgery details, and make it set tx, the elements of the set are all diagnosis codes that are the same as any main surgery details;

[0071] S15: Obtain the set of surgery codes in the settlement list data that are the same as the minor surgery details, and make it set txMinor, the elements of the set are all diagnosis codes that are the same as any minor surgery details;

[0072] S16: According to dx, dxMinor, tx, txMinor, perform grid search grouping, that is, perform exhaustive combination of all elements in the above four sets, obtain all disease diagnosis related groups (DRGs) that can enter according to the grouping scheme to obtain the real grouping that may exist, and combine all the obtained groups into the drgSet set, the elements of the set are all disease diagnosis related groups (DRGs) obtained in this process.

[0073] According to the drgSet set obtained in S1, use the ectopic fuzzy grouping module to perform ectopic fuzzy grouping to obtain the real grouping that may exist, including the following steps:

[0074] Exchange the position of each minor diagnosis of all disease diagnosis related groups (DRGs) in the drgSet set with the main diagnosis, and exchange the position of each minor surgery with the main surgery, perform S1 in-place fuzzy grouping process once each time to obtain a disease diagnosis related group (DRG), and combine these groups into the drgSet set.

[0075] Use the QY grouping module to perform QY grouping supplement to compensate for the inaccuracy of grouping caused by the absence of main diagnosis, and obtain the real grouping that may exist, including the following steps:

[0076] S31: Determine whether there is a valid main surgery code in the settlement list, and whether there is a corresponding surgery group in drgSet, if yes, continue to the next step, otherwise, stop;

[0077] S32: If the main surgery code corresponds to a surgery group set tx, and the main diagnosis and minor diagnosis conditions in the core disease diagnosis related group (ADRG) of each group in the set tx meet the group entry conditions of the ADRG in the grouping scheme, remove the disease diagnosis related groups (DRGs) that meet the group entry conditions of the ADRG in the grouping scheme from the settlement list, and form a set of disease diagnosis related groups (DRGs) that meet the group entry conditions of the ADRG in the grouping scheme, the elements of this set are disease diagnosis related groups (DRGs), perform set union operation on this set and drgSet, and assign the result of the set union operation to drgSet.

[0078] Using the missing code recommendation module, missing code grouping is supplemented, and in-situ fuzzy grouping is performed to obtain possible real groups, including the following steps:

[0079] S41: Determine whether the settlement list contains valid operation codes. If not, continue to the next step. If yes, terminate;

[0080] S42: According to the charge item details and the valid operation code table, the most effective operation code details are matched;

[0081] S43: From the ICD code table (International Classification of Diseases, this embodiment uses the national medical insurance version 2.0 ICD code dictionary table), obtain all operation code sets of the same item as the aforementioned operation code, become set txQY, the elements of this set are all operation codes of the same item as the aforementioned operation code;

[0082] S44: Place txQY in the main operation position of the settlement list, and perform S1 in-situ fuzzy grouping process, perform set union operation on the output set and drgSet, and assign the operation result to drgSet.

[0083] Using the grouping fit degree module, the fit degrees of all groups in drgSet obtained in the aforementioned four steps are calculated respectively:

[0084] S51: Obtain the total cost c0 of the case in the settlement list and the case hospitalization days p0

[0085] S52: For each group in the aforementioned S1-S4 drgSet, obtain the average cost c and the average hospitalization days p of the group from the payment parameter table;

[0086] S53: Calculate the distance index q',

[0087]

[0088] S54: Map the distance index to a scalar q between 0 and 1. This scalar q is the fit degree. The specific mapping method is:

[0089]

[0090] The groups in the drgSet are sorted in descending order of q to form an output list, which is a recommended group list arranged in descending order of recommendation degree, and doctors can select according to the recommended group list. The group recommendation method provides various groupable sets, and the group fit degree of each group is calculated for doctors to determine the most suitable group. Moreover, the application can reduce group errors caused by the limitations of clinical specialists, test results, medical records and other clinical rules, reduce coding errors caused by the limitations of the coding staff filling in the ICD code according to the filling specifications of the settlement list, reduce group errors caused by the doctors filling in the settlement list using different versions of the ICD code, and greatly improve the group rate.

Claims

1. A settlement-invoice-based grouping recommendation method, characterized by: The steps are as follows: S1. Perform in-situ fuzzy grouping and expand the grouping range: Obtain the codes of primary diagnosis, secondary diagnosis, primary surgery, and secondary surgery in the settlement list data and group them to obtain the possible true groups; S2. Perform anatomical fuzzy grouping to further expand the grouping range: exchange the positions of the primary and secondary diagnosis codes in the settlement list data, and at the same time exchange the positions of the primary and secondary surgery codes. After changing the positions, perform S1 in-situ fuzzy grouping to obtain the possible true groups. S3. Perform QY (ambiguous) grouping supplementation, grouping with the primary diagnosis code paired with the surgical code to make up for the inaccuracy of grouping caused by the lack of primary diagnosis and obtain the possible true grouping; S4. Perform missing coding grouping and perform in-situ fuzzy grouping to obtain the possible true groups; S5. Calculate the fit of all groups from S1 to S4. The steps for calculating the fit are as follows: S51: Obtain the total cost c0 and the number of days of hospitalization p0 from the settlement statement. S52: For each group in the drgSet obtained in S1~S4 above, obtain the average cost c and average length of hospital stay p of the group from the payment parameter table; S53: Calculate the distance index q'. ； S54: Map the distance index to a scalar q between 0 and 1. This scalar q represents the degree of fit. The specific mapping method is as follows: ； Sort the groups in drgSet in descending order of q to form a list, and output the list to obtain a list of recommended groups arranged from high to low recommendation level.

2. The method of claim 1, wherein: The steps for S1 in-situ fuzzy grouping are as follows: S11: Create an empty collection drgSet; S12: Obtain the set of diagnostic codes that are the same as the main diagnostic subcategory in the settlement list data, and call it set dx. The elements of the set are all diagnostic codes that are the same as the main diagnostic subcategory. S13: Obtain the set of diagnostic codes that are the same as any sub-category to be diagnosed in the settlement list data, and call it set dxMinor. The elements of the set are all diagnostic codes that are the same as any sub-category to be diagnosed. S14: Obtain the set of surgical codes that are the same as the main surgical details in the settlement list data, and call it set tx. The elements of the set are all diagnostic codes that are the same as any main surgical details. S15: Obtain the set of surgical codes that are the same as any one of the required surgical details from the settlement list data, and call it the set txMinor. The elements of the set are all the diagnostic codes that are the same as any one of the required surgical details. S16: Perform grid search grouping based on dx, dxMinor, tx, and txMinor. That is, exhaustively combine all elements in the above four sets to obtain all accessible disease diagnosis related groups (DRGs) according to the grouping scheme, and merge all obtained groups into the drgSet set. The elements of the set are all disease diagnosis related groups (DRGs) obtained in this process. 3.The method of claim 2, wherein: The steps for S2 fuzzy grouping are: For each minor diagnosis in all disease diagnosis related groups (DRGs) in the drgSet set, swap the position of the minor diagnosis with the major diagnosis. At the same time, swap the position of each minor surgery with the major surgery. Perform the S1 in-situ fuzzy grouping process once for each swap to obtain a disease diagnosis related group (DRG). Then merge these groups into the drgSet set.

4. The method of claim 3, wherein: The steps for supplementing the S3 QY group are as follows: S31: Determine if there is a valid primary surgery code in the settlement list and if there is no corresponding surgical group in drgSet. If so, continue to the next step; otherwise, stop. S32: If, in the surgical group set tx corresponding to the valid primary surgery code, the primary and secondary diagnostic conditions in the core disease diagnosis related group (ADRG) of each group have met the inclusion conditions of the ADRG in the grouping scheme, the disease diagnosis related groups (DRGs) that meet the inclusion conditions of the ADRGs in the settlement list, excluding the inclusion conditions related to the primary surgery, primary diagnosis, and secondary diagnosis, are formed into a set. The elements of this set are disease diagnosis related groups (DRGs). This set is then subjected to a set union operation with drgSet, and the result of the union operation is assigned to drgSet.

5. The method of claim 4, wherein: The S4 missing coding grouping specifically includes the following steps: S41: Determine if a valid surgical code exists on the settlement list. If not, continue to the next step; otherwise, terminate. S42: Based on the billing details and the valid surgical code table, the most effective surgical code details are matched; S43: Obtain the set of all surgical codes that are in the same subcategory as the aforementioned surgical codes from the ICD coding table, which is called set txQY. The elements of this set are all surgical codes that are in the same subcategory as the aforementioned surgical codes. S44: Place txQY in the main operation position of the settlement list, and execute the in-situ fuzzy grouping process of S1. Perform a set union operation on the output set and drgSet, and assign the result to drgSet.

6. A group recommendation system based on a settlement list, characterized in that: A grouping recommendation method based on a settlement list as described in any one of claims 1-5 includes an in-situ fuzzy grouping module, an out-of-situ fuzzy grouping module, a QY (ambiguous) grouping module, a missing code recommendation module, and a grouping matching degree module, wherein: In-situ fuzzy grouping module: used to perform grid search grouping of all ICD codes under the same category for each ICD code; among them, the diagnostic code classification information is obtained according to the subcategory classification standard of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), and the diagnostic codes include primary diagnostic codes and secondary diagnostic codes; the surgical code classification information is obtained according to the detailed classification standard of the International Classification of Diseases, Ninth Revision, Clinical Revision, Surgical and Procedures (ICD-9-CM-3), and the surgical codes include primary surgical codes and secondary surgical codes; The eccentric fuzzy grouping module is used to change the position of the aforementioned ICD codes and group them. The position refers to the field in which the ICD code, as a value, is located. There are four fields: primary diagnosis, secondary diagnosis, primary surgery, and secondary surgery. Primary diagnosis refers to the disease (or health condition) determined by a medical institution as the main reason for the patient's hospitalization. Secondary diagnosis refers to diseases that existed concurrently with the patient's hospitalization, occurred later, or affected the treatment received and length of hospitalization. Primary surgery refers to the surgery or procedure performed on the patient during their hospitalization for the condition for which the clinician made the primary diagnosis. Secondary surgery refers to other surgeries or procedures performed on the patient during their hospitalization. QY (Ambiguous) Grouping Module: Grouping is done using the primary diagnosis code paired with the surgical code. This module is used to supplement the grouping of possible QY cases and make up for the inaccuracy of grouping caused by the lack of a primary diagnosis. Missing Code Recommendation Module: Used to supplement missing major surgical codes and perform in-situ fuzzy grouping; Grouping Matching Module: This module calculates the matching degree of all groups from S1 to S4, arranges the matching degree values ​​in descending order into a list, and outputs the list, which is the recommended grouping list from high to low recommendation degree.

Images