A human injury case loss assessment method and device, electronic equipment and storage medium

Abstract

The application provides a human injury case damage assessment method and device, electronic equipment and a storage medium. The method comprises the following steps: identifying a user-uploaded injury image, and determining a first injury label corresponding to the injury image; predicting a first item list required for an injured person examination and a predicted medical expense corresponding to the first item list based on the first injury label and first injured person information corresponding to the injury image; determining a compensation expense based on the first injury label, the first injured person information and a pre-constructed rule base, wherein the rule base comprises compensation standards corresponding to a work loss period, a nursing period and a nutrition period respectively; and generating a damage assessment scheme according to the first item list, the predicted medical expense and the compensation expense, and feeding back the damage assessment scheme to the user. The whole process from case entry to compensation payment is automatically and intelligently processed, the damage assessment standard is more unified, the human injury case damage assessment efficiency is improved, and the cost required for damage assessment of small human injury cases of vehicle insurance is reduced.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence and insurance technology, and in particular to a method, device, electronic device and storage medium for assessing damages in personal injury cases. Background Technology

[0002] Minor personal injury claims in auto insurance (typically referring to claims with payouts below a certain threshold, such as RMB 5,000 or less) account for a very high percentage (over 70%) of all personal injury claims in auto insurance. While the payouts for these cases are relatively small, current loss assessment methods still rely heavily on claims adjusters' manual judgment and operations in areas such as injury evaluation, determination of examination items, and cost correlation. This requires a significant investment of human resources and has low economic efficiency. Therefore, there is an urgent need for a more intelligent, efficient, and cost-effective method for assessing personal injury claims. Summary of the Invention

[0003] This application aims to at least partially address one of the technical problems in the related art.

[0004] Therefore, the first objective of this application is to propose a method for assessing damages in personal injury cases, which integrates technologies such as multimodal large models, historical data mining and analysis, and rule engines to solve problems such as the objectivity of injury assessment, the rationality of recommended examination items, the accuracy of cost calculation, and the standardization of three-stage compensation, ultimately achieving improved efficiency in damage assessment, reduced costs, and unified compensation standards.

[0005] The second objective of this application is to propose a device for assessing damages in personal injury cases.

[0006] The third objective of this application is to propose an electronic device.

[0007] The fourth objective of this application is to provide a computer-readable storage medium.

[0008] The fifth objective of this application is to provide a computer program product.

[0009] To achieve the above objectives, the first aspect of this application proposes a method for assessing damages in personal injury cases, comprising:

[0010] The system identifies the injury images uploaded by the user and determines the first injury tag corresponding to the injury image. The first injury tag includes a location identifier and a severity level. Based on the first injury label and the first patient information corresponding to the injury image, a first list of items required for the patient's examination and the predicted medical costs corresponding to the first list of items are predicted. Based on the first injury label, the first injured person information, and a pre-built rule base, the compensation amount is determined. The rule base includes the compensation standards corresponding to the lost work period, nursing period, and nutritional period, respectively. Based on the first item list, the predicted medical expenses, and the compensation expenses, a damage assessment plan is generated and fed back to the user.

[0011] To achieve the above objectives, a second aspect of this application provides a device for assessing damages in personal injury cases, comprising: The recognition module is used to recognize the injury images uploaded by the user and determine the first injury label corresponding to the injury image. The first injury label includes a location identifier and a severity level. The prediction module is used to predict a first list of items required for the patient's examination and the predicted medical costs corresponding to the first list of items, based on the first injury label and the first patient information corresponding to the injury image. The calculation module is used to determine the compensation amount based on the first injury label, the first injured person information, and a pre-built rule base, wherein the rule base includes the compensation standards corresponding to the lost work period, nursing period, and nutritional period, respectively. The generation module is used to generate a damage assessment plan based on the first item list, the predicted medical expenses, and the compensation expenses, and then provide the plan to the user.

[0012] To achieve the above objectives, a third aspect of this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in the first aspect embodiment.

[0013] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the method described in the first aspect.

[0014] To achieve the above objectives, a fifth aspect of this application provides a computer program product including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The personal injury case assessment method, device, electronic equipment, and storage medium provided in this application identify injury images uploaded by users, determine a first injury label describing the injured person's injury location and severity, and then, based on the first injury label and the injured person's basic information, intelligently predict the list of examination items required for the injured person and the medical costs required to complete all examination items in the list. Furthermore, based on the injured person's injury label and basic information, it calculates the three-stage compensation costs for the lost work period, nursing period, and nutritional period according to a rule base. Thus, based on the item list, predicted medical costs, and three-stage compensation costs, it generates a loss assessment plan for the injured person for claims processing. This enables fully automated and intelligent processing from case entry to compensation payment, making the loss assessment standards more unified and reasonable, significantly improving the efficiency of personal injury case assessment, and reducing the cost required for loss assessment in small-amount personal injury cases under auto insurance.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for assessing damages in personal injury cases provided in an embodiment of this application; Figure 2 This is a flowchart illustrating another method for assessing damages in personal injury cases provided in an embodiment of this application. Figure 3 This is a schematic diagram of a damage assessment device for personal injury cases provided in an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0019] The following description, with reference to the accompanying drawings, describes a method and apparatus for assessing damages in personal injury cases according to embodiments of this application.

[0020] Figure 1 This is a flowchart illustrating a method for assessing damages in personal injury cases provided in an embodiment of this application.

[0021] like Figure 1 As shown, the method for assessing damages in this personal injury case includes the following steps: Step 101: Identify the injury images uploaded by the user and determine the first injury tag corresponding to the injury image.

[0022] The first injury label describes the injury condition of the victim in the current personal injury case requiring damage assessment. It can include a location identifier and a severity level. The location identifier refers to the name of the injured body part, such as the forehead, left forearm, left / right thigh, chest, etc. The severity level can be classified according to the actual accuracy required for damage assessment; for example, the severity level can be divided into mild, moderate, and severe.

[0023] In this application, injury images can be obtained by uploading injury photos or video streams via applications such as APP, mini-programs, or remote video.

[0024] In this application, a multimodal large model can be used to identify injury images and determine the first injury label corresponding to the injury image. The multimodal large model can extract image features from the injury image, then interactively align the image features with text prompts, and output a standardized first injury label through a classification layer.

[0025] It should be noted that before identifying injury images, preprocessing can be performed to optimize image quality and improve recognition accuracy. Preprocessing may include, but is not limited to, operations such as brightness adjustment, contrast enhancement, and histogram equalization.

[0026] Step 102: Based on the first injury label and the first injured person information corresponding to the injury image, predict the first list of items required for the injured person's examination and the predicted medical costs corresponding to the first list of items.

[0027] Among them, the first wounded information is the basic information of the wounded corresponding to the injury image, which may include, but is not limited to, name, age, occupation, etc.

[0028] The predicted medical expenses are the sum of the examination fees and medication costs for all examination items in the first item list.

[0029] It should be noted that the information of the first injured person can be entered by the staff handling the case claims, or it can be obtained automatically by identifying uploaded identity documents and other materials. This application does not limit this. All operations such as obtaining, using and processing the injured person's information in this application are carried out with the authorization of the injured person.

[0030] In this application, the first injury label and the first injured person's information can be input into a trained project prediction model and cost analysis model. The model can then intelligently generate suggested examination plans and cost estimates, resulting in a list of first projects and the corresponding predicted medical costs. Step 103: Determine the compensation amount based on the first injury label, the first injured person's information, and the pre-built rule base.

[0031] The rule base includes compensation standards for lost work periods, nursing care periods, and nutritional support periods.

[0032] This application digitizes and structures the assessment standards for the "three periods" (lost work period, nursing period, and nutritional period) issued by local human resources and social security departments and judicial appraisal institutions, forming a computer-executable rule base. The rule base may also include the standard number of days or calculation logic corresponding to different injuries, age groups, and occupational categories.

[0033] In this application, the compensation includes lost wages, nursing care expenses, and nutritional expenses, and one or more of these expenses may be zero. The first injury label and the first injured person information can be matched with different calculation logics in the rule base to determine the corresponding cost calculation standards for the three successfully matched periods, thereby standardizing and automating the calculation of lost wages, nursing care expenses, and nutritional expenses in the compensation.

[0034] Step 104: Based on the first item list, predicted medical expenses and compensation costs, generate a damage assessment plan and provide it to the user.

[0035] In this application, the estimated medical expenses and compensation costs can be aggregated as the amount to be paid in the loss assessment plan. The plan can then be fed back to the user for online confirmation and signing of an electronic settlement agreement, thereby automatically triggering the payment process and closing the case upon confirmation of payment.

[0036] It should be noted that after generating the loss assessment plan, an anti-fraud rule engine can be used to review the plan before it is pushed to the user. This ensures efficient claims processing while effectively identifying and blocking potential fraud and unreasonable charges, achieving the dual goals of cost reduction, efficiency improvement, and risk control.

[0037] In this embodiment, by recognizing the injury images uploaded by the user, a first injury tag describing the injured person's injury location and severity is determined. Based on the first injury tag and the injured person's basic information, the system intelligently predicts the list of examination items required for the injured person and the medical costs required to complete all examination items in the list. According to the injured person's injury tag and basic information, the system calculates the compensation costs for the three periods of lost work time, nursing care time, and nutritional care time based on a rule base. Thus, based on the list of items, predicted medical costs, and three-period compensation costs, a loss assessment plan for the injured person is generated for claims processing. This can realize the full-process automation and intelligent processing from case entry to compensation payment, making the loss assessment standards more unified and reasonable, significantly improving the efficiency of loss assessment for personal injury cases, and reducing the cost required for loss assessment of small personal injury cases in auto insurance.

[0038] This embodiment provides another method for assessing damages in personal injury cases. Figure 2 This is a flowchart illustrating another method for assessing damages in personal injury cases provided in an embodiment of this application.

[0039] like Figure 2 As shown, the method for assessing damages in this personal injury case may include the following steps: Step 201: Identify the injury images uploaded by the user and determine the first injury tag corresponding to the injury image.

[0040] For a detailed description of step 201 above, please refer to other embodiments of this application, which will not be repeated here.

[0041] Step 202: Input the first injury label and the first injured person information into the target project prediction model to obtain the first project list output by the project prediction model.

[0042] The first item list contains one or more prediction check items arranged from high to low prediction probability.

[0043] It should be noted that in this application, the target project prediction model can be a model based on the Gradient Boosting Decision Tree (GBDT) model and combined with cluster analysis.

[0044] In some possible implementations, during the training of the target project prediction model, multiple training samples can be obtained from a historical database related to personal injury cases. Each training sample contains a second injury label, second injury information, and a second project list consisting of actual examination items.

[0045] The historical database is a collection of case data that has been processed through claims.

[0046] The second injury label describes the injury details of victims in historical personal injury cases. Secondary injury information includes the victim's age, occupation, and other details. The second item list may also include the final confirmed costs of each item in historical claims.

[0047] In this application, valid data can be extracted from a historical claims database to form a training sample set consisting of multiple training samples. After obtaining multiple training samples, categorical features (such as injury location and names of historical examination items) can be encoded, and numerical features (such as age and historical expenses) can be normalized.

[0048] Then, the probability of any two actual inspection items in the second item list appearing simultaneously can be determined. If the probability is greater than a probability threshold, these two actual inspection items are selected as the reference item combination.

[0049] In this application, to avoid recommending redundant or unconventional test combinations, cluster analysis can be performed on frequently co-occurring test items in historical data. Historical test item combinations are categorized into several typical packages (i.e., reference test item combinations), causing the trained test prediction model to tend to recommend test items within the same cluster, thereby improving the rationality and clinical applicability of the recommendations. Therefore, the probability of any two actual test items appearing simultaneously in the second test item list can be determined. If the probability is greater than a probability threshold, these two actual test items are designated as the reference test item combination.

[0050] Then, the initial project prediction model can be trained based on all reference project combinations and multiple training samples to obtain the target project prediction model.

[0051] In this application, an initial item prediction model is trained based on all reference item combinations and multiple training samples, enabling the target item prediction model to effectively learn the complex nonlinear relationship between injury labels and casualty information and examination items. The input to the target item prediction model is a processed feature vector, and the output is the predicted probability of each possible examination item.

[0052] Step 203: Input the first item list into the target cost analysis model to obtain the predicted medical costs.

[0053] In this application, the target cost analysis model can establish a mapping from "examination items" to "standard costs", providing an accurate price basis for loss assessment. Therefore, the first item list can be input into the target cost analysis model to obtain the predicted medical costs output by the target cost analysis model.

[0054] It should be noted that in this application, a target cost analysis model that can be constructed based on time series prediction and combined with statistics and rule engines can be built according to whether the prices of different inspection items are stable.

[0055] In some possible implementations, when modeling the target cost analysis model, reference data on the medical costs required for the examination items can be collected based on at least one data source.

[0056] In this application, at least one data source may include, but is not limited to: the fee schedule within the Hospital Information System (HIS), the regional medical insurance payment catalog and prices, and historical claims data from claims companies and other partners.

[0057] Then, the reference data can be analyzed to determine the cost type for each examination item. If the cost type is type one, the reference data can be used to determine the pattern of how the medical costs for the examination items change over time; or, if the cost type is type two, the statistical values ​​of all costs corresponding to the examination items in the reference data can be determined.

[0058] In this application, the first type refers to examination items whose costs fluctuate regularly over time, such as some laboratory tests. Based on historical price time series data from reference data, time series analysis and forecasting methods such as Autoregressive Integrated Moving Average (ARIMA) can be used to determine the price forecast for this first type of examination item over a future period, reflecting the dynamic changes in costs.

[0059] In this application, the second type refers to inspection items with relatively stable costs. Therefore, when calculating the cost of this second type of inspection item, the statistical values ​​(such as mode, median, specific quantiles, etc.) of all costs corresponding to this inspection item in the reference data can be calculated as a standard reference value.

[0060] Then, a target cost analysis model can be constructed based on the patterns or statistical values ​​corresponding to all inspection items.

[0061] It should be noted that the target cost analysis model can also integrate specific cost rules (such as regional maximum price limits and medical insurance reimbursement ratios) to ensure that the output cost recommendations comply with policy requirements.

[0062] In this application, the model that correlates injury with examination items and costs, trained based on a large amount of historical data, can provide damage assessment recommendations that are more in line with clinical practice and market price levels, thereby improving the scientific nature and accuracy of the damage assessment results.

[0063] Step 204: Determine the compensation amount based on the first injury label, the first injured person information, and the pre-built rule base.

[0064] In some possible implementations, the region to which the injured person belongs can be determined based on the first injured person's information. The rule base is then used to filter the region based on the following criteria: the first criterion for lost work time, the second criterion for nursing care, and the third criterion for nutritional support.

[0065] It should be noted that the calculation of expenses such as lost wages, nursing care expenses, and nutritional expenses is directly linked to statistical data such as the economic development level, residents' income, and consumption expenditure in different regions. Therefore, there are differences in the calculation standards for lost wages, nursing care expenses, and nutritional expenses in different regions. Selecting and applying the correct compensation standards according to the region where the case is located can ensure that the loss assessment results are legal, fair, and accurate.

[0066] Then, based on the primary standard and the severity level in the primary injury label, the number of days of lost work time for the injured worker can be determined. Based on the primary injured worker's information or the corresponding income level in the region, and the number of days, the primary compensation for lost work time can be calculated.

[0067] The first criterion may include a mapping relationship between severity level and number of days.

[0068] In this application, the standard number of lost workdays can be determined based on the severity level in the first injury label and the mapping relationship in the first standard. Then, based on the occupational category and / or relevant income proof in the first injured person's information, or the local average wage standard, the first compensation for the lost workdays can be automatically calculated.

[0069] Subsequently, based on the first injury label and the first injured person's information, and in conjunction with the second and third standards respectively, the second expense for nursing care compensation and the third expense for nutritional compensation can be calculated. The sum of the first expense, the second expense, and the third expense is determined as the compensation amount.

[0070] In this application, the nursing period and nutritional period can also be compensated according to their corresponding second and third standards, based on the first injury label and the first injured person's information. The sum of the first, second, and third expenses should be determined as the amount of compensation for the injured person in the current personal injury case for the three periods.

[0071] Step 205: Based on the first item list, predicted medical expenses and compensation costs, generate a damage assessment plan and provide it to the user.

[0072] For a detailed description of step 205 above, please refer to other embodiments of this application, which will not be repeated here.

[0073] In some possible implementations, medical invoices uploaded by users can be parsed to determine the list of third-party examinations actually performed on the injured person and the actual medical expenses.

[0074] Then, if the third item list does not match the first item list, and / or the difference between the actual medical expenses and the predicted medical expenses is greater than the difference threshold, the loss assessment plan can be updated based on the third item list and the actual medical expenses.

[0075] Subsequently, the target item prediction model and target cost analysis model can be updated based on the first injury label, the first injured person information, the third item list, and the actual medical expenses.

[0076] In this application, if a user submits a medical bill, the bill content can be automatically parsed using Optical Character Recognition (OCR) and Natural Language Processing (NLP) technologies to extract the list of the third items actually examined by the injured person and the actual medical expenses. This list is then cross-validated with the first item list recommended by the model and the predicted medical expenses in the above embodiments. If the actual situation does not match the prediction, the damage assessment plan can be updated, and the first injury label and the first injured person's information can be used as new training samples for iterative updates of the target item prediction model and the target cost analysis model.

[0077] In this embodiment, by using a target item prediction model and a target cost analysis model to replace human experience in judging the items to be examined and the required medical expenses, the subjective differences between different claims adjusters are effectively eliminated, so that cases with similar injuries can obtain consistent and fair compensation plans, and reduce claims disputes.

[0078] It should be noted that this application integrates a variety of cutting-edge technologies, such as multimodal large models, historical data mining, time series analysis, and rule engines, and applies them to the specific scenario of small-amount personal injury claims assessment in auto insurance, which can achieve truly intelligent decision-making.

[0079] The following example, using a case where the injured party suffered a minor contusion to the left forearm and received no treatment, details the method for assessing damages in personal injury cases provided in this application.

[0080] First, a report and data entry are conducted. The user reports the incident through the app, uploading a photo of the swelling and redness on their left forearm as an image of the injury. The claims personnel select "no treatment."

[0081] Then, injury recognition is performed. After preprocessing the injury image, features are extracted by the Vision Transformer (ViT) model in the multimodal large model, the Segment Anything Model (SAM) model segments the injured area, and the Bootstrapped Language-Image Pretraining (BLIP-2) model combines prompt words to perform multimodal alignment. Finally, a standardized injury label is output (injured location: left forearm, severity: mild).

[0082] Next, the examination items and costs are inferred. The injury label is input into the pre-trained target item prediction model, which recommends "X-ray examination" (probability 95%) and outputs the standard cost reference value of 150 yuan for this examination in the current region and historical data.

[0083] Then, the cost calculation for the three phases is performed. According to the rule base, it can be determined that "mild contusion of the left forearm" does not require lost work, nursing care, or nutritional support period, so the cost for the three phases is 0 yuan.

[0084] The final loss assessment plan is generated: medical expenses 150 yuan, three-phase expenses 0 yuan, totaling 150 yuan. After the client confirms the loss assessment plan, they sign the agreement online and complete the payment.

[0085] To achieve the above embodiments, this application also proposes a device for assessing damages in personal injury cases.

[0086] Figure 3 This is a schematic diagram of a damage assessment device for personal injury cases provided in an embodiment of this application.

[0087] like Figure 3 As shown, the personal injury case damage assessment device 30 includes: The recognition module 301 is used to recognize the injury images uploaded by the user and determine the first injury label corresponding to the injury image. The first injury label includes the location identifier and the severity level. The prediction module 302 is used to predict the first list of items required for the examination of the injured person and the predicted medical costs corresponding to the first list of items, based on the first injury label and the first injured person information corresponding to the injury image. The calculation module 303 is used to determine the compensation amount based on the first injury label, the first injured person information and the pre-built rule base. The rule base includes the compensation standards corresponding to the lost work period, nursing period and nutrition period respectively. The generation module 304 is used to generate a damage assessment plan based on the first item list, predicted medical expenses and compensation expenses, and then provide the plan to the user.

[0088] Furthermore, in one possible implementation of this application embodiment, the prediction module 302 may specifically be used for: Input the first injury label and the first injured person information into the target project prediction model to obtain the first project list output by the project prediction model. The first project list contains one or more prediction check items arranged from high to low prediction probability. Input the first item list into the target cost analysis model to obtain the predicted medical costs.

[0089] Furthermore, in one possible implementation of this application embodiment, the personal injury case damage assessment device 30 further includes a training module, which can be specifically used for: Multiple training samples were obtained from a historical database related to personal injury cases. Each training sample contained a second injury label, second injury information, and a list of second items consisting of actual examination items. Determine the probability that any two actual inspection items in the second item list will appear simultaneously. If the probability is greater than the probability threshold, any two actual inspection items will be determined as the reference item combination. Based on all reference project combinations and multiple training samples, the initial project prediction model is trained to obtain the target project prediction model.

[0090] Furthermore, in one possible implementation of this application embodiment, the personal injury case damage assessment device 30 further includes a construction module, which can be specifically used for: Based on at least one data source, collect reference data on the medical costs required for the examination items; Analyze the reference data to determine the cost type for each inspection item; When the cost type is the first type, the pattern of how the medical costs required for the examination items change over time is determined based on the reference data; If the cost type is type 2, determine the statistical values ​​of all costs corresponding to the inspection items in the reference data; Based on the patterns or statistical values ​​corresponding to all inspection items, a target cost analysis model is constructed.

[0091] Furthermore, in one possible implementation of this application embodiment, the personal injury case damage assessment device 30 further includes an update module, which can be specifically used for: When a user uploads a medical bill, the medical bill is parsed to determine the list of the third items that the injured person actually underwent and the actual medical expenses. If the third item list does not match the first item list, and / or the difference between the actual medical expenses and the predicted medical expenses is greater than the difference threshold, the loss assessment plan shall be updated based on the third item list and the actual medical expenses. Based on the first injury label, the first injured person information, the third item list, and the actual medical expenses, update the target item prediction model and the target cost analysis model.

[0092] Furthermore, in one possible implementation of this application embodiment, the calculation module 303 may specifically be used for: Based on the information of the first injured person, determine the region to which the injured person belongs; In the rule base, filter the first standard for the associated period of absence from work, the second standard for the associated period of care, and the third standard for the associated period of nutrition. The number of days of lost work period for the injured person is determined based on the first standard and the severity level in the first injury label. The first rule contains the mapping relationship between severity level and number of days. The first expense for lost work period compensation is calculated based on the information of the first injured person or the income corresponding to the region, and the number of days. Based on the first injury label and the first injured person information, the second expense for nursing care compensation and the third expense for nutritional compensation are calculated by combining the second and third standards respectively. The sum of the first expense, the second expense, and the third expense shall be determined as the compensation.

[0093] It should be noted that the foregoing explanation of the embodiment of the method for assessing damages in personal injury cases also applies to the personal injury case assessment device of this embodiment, and will not be repeated here.

[0094] In this embodiment, by recognizing the injury images uploaded by the user, a first injury tag describing the injured person's injury location and severity is determined. Based on the first injury tag and the injured person's basic information, the system intelligently predicts the list of examination items required for the injured person and the medical costs required to complete all examination items in the list. According to the injured person's injury tag and basic information, the system calculates the compensation costs for the three periods of lost work time, nursing care time, and nutritional care time based on a rule base. Thus, based on the list of items, predicted medical costs, and three-period compensation costs, a loss assessment plan for the injured person is generated for claims processing. This can achieve fully automated and intelligent processing from case entry to compensation payment, making the loss assessment standards more unified and reasonable, significantly improving the efficiency of loss assessment for personal injury cases, and reducing the cost required for loss assessment of small personal injury cases in auto insurance.

[0095] To implement the above embodiments, this application also proposes an electronic device, including: a processor and a memory communicatively connected to the processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to implement the method provided in the foregoing embodiments.

[0096] To implement the above embodiments, this application also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided in the foregoing embodiments.

[0097] To implement the above embodiments, this application also proposes a computer program product, including a computer program that, when executed by a processor, implements the methods provided in the foregoing embodiments.

[0098] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0099] It should be noted that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. Furthermore, such collection / sharing should only be conducted after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes authorization of relevant user information before the user uses the function. In addition, any necessary steps must be taken to protect and safeguard access to such personal information data and ensure that others with access to personal information data comply with their privacy policies and procedures.

[0100] This application is intended to provide an implementation scheme for users to selectively prevent the use or access to their personal information data. Specifically, this disclosure is intended to provide hardware and / or software to prevent or block access to such personal information data. Once personal information data is no longer needed, risks can be minimized by restricting data collection and deleting data. Furthermore, where applicable, such personal information is de-identified to protect user privacy.

[0101] In the foregoing descriptions of the embodiments, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0102] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0103] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0104] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0105] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0106] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0107] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0108] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for assessing damages in personal injury cases, characterized in that, Includes the following steps: The system identifies the injury images uploaded by the user and determines the first injury tag corresponding to the injury image. The first injury tag includes a location identifier and a severity level. Based on the first injury label and the first patient information corresponding to the injury image, a first list of items required for the patient's examination and the predicted medical costs corresponding to the first list of items are predicted. Based on the first injury label, the first injured person information, and a pre-built rule base, the compensation amount is determined. The rule base includes the compensation standards corresponding to the lost work period, nursing period, and nutritional period, respectively. Based on the first item list, the predicted medical expenses, and the compensation expenses, a damage assessment plan is generated and fed back to the user.

2. The method for assessing damages in personal injury cases according to claim 1, characterized in that, The step of predicting a first list of items required for the patient's examination and the predicted medical costs corresponding to the first list of items, based on the first injury label and the first patient information corresponding to the injury image, includes: The first injury label and the first injured person information are input into the target project prediction model to obtain the first project list output by the project prediction model. The first project list contains one or more prediction check items arranged from high to low prediction probability. The first item list is input into the target cost analysis model to obtain the predicted medical costs.

3. The method for assessing damages in personal injury cases according to claim 2, characterized in that, The training process of the target project prediction model includes: Multiple training samples were obtained from a historical database related to personal injury cases. Each training sample contained a second injury label, second injury information, and a list of second items consisting of actual examination items. Determine the probability that any two actual inspection items in the second item list will appear simultaneously. If the probability is greater than the probability threshold, any two actual inspection items will be determined as a reference item combination. Based on all the reference project combinations and the multiple training samples, the initial project prediction model is trained to obtain the target project prediction model.

4. The method for assessing damages in personal injury cases according to claim 2, characterized in that, The process of constructing the target cost analysis model includes: Based on at least one data source, collect reference data on the medical costs required for the examination items; Analyze the reference data to determine the cost type for each inspection item; When the cost type is the first type, the pattern of how the medical costs required for the examination items change over time is determined based on the reference data; If the cost type is the second type, determine the statistical values ​​of all costs corresponding to the inspection items in the reference data; Based on the patterns or statistical values ​​corresponding to all inspection items, the target cost analysis model is constructed.

5. The method for assessing damages in personal injury cases according to any one of claims 1-4, characterized in that, Also includes: When the user uploads medical bills, the medical bills are parsed to determine the list of the third items that the injured person actually underwent and the actual medical expenses. If the third item list does not match the first item list, and / or the difference between the actual medical expenses and the predicted medical expenses is greater than the difference threshold, the loss assessment plan shall be updated based on the third item list and the actual medical expenses. Based on the first injury label, the first injured person information, the third item list, and the actual medical expenses, update the target item prediction model and the target cost analysis model.

6. The method for assessing damages in personal injury cases according to any one of claims 1-4, characterized in that, The determination of compensation based on the first injury label, the first injured person information, and a pre-built rule base includes: Based on the first information about the injured person, the region to which the injured person belongs is determined; In the rule base, filter the first standard associated with the period of absence from work, the second standard associated with the period of care, and the third standard associated with the period of nutrition for the region. The number of days of lost work period for the injured person is determined based on the first standard and the severity level in the first injury label. The first standard includes a mapping relationship between the severity level and the number of days. Based on the first injured person's information or the income corresponding to the region, and the number of days, calculate the first expense for the lost work period compensation; Based on the first injury label and the first injured person information, and in conjunction with the second standard and the third standard respectively, the second expense of the nursing period compensation and the third expense of the nutritional period compensation are calculated. The sum of the first expense, the second expense, and the third expense is determined as the compensation expense.

7. A device for assessing damages in personal injury cases, characterized in that, include: The recognition module is used to recognize the injury images uploaded by the user and determine the first injury label corresponding to the injury image. The first injury label includes a location identifier and a severity level. The prediction module is used to predict a first list of items required for the patient's examination and the predicted medical costs corresponding to the first list of items, based on the first injury label and the first patient information corresponding to the injury image. The calculation module is used to determine the compensation amount based on the first injury label, the first injured person information, and a pre-built rule base, wherein the rule base includes the compensation standards corresponding to the lost work period, nursing period, and nutritional period, respectively. The generation module is used to generate a damage assessment plan based on the first item list, the predicted medical expenses, and the compensation expenses, and then provide the plan back to the user.

8. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

Images