Hematologic tumor peaceful treatment and protection intervention method based on potential profile analysis

Abstract

The invention discloses a hematologic tumor calming treatment and protection intervention method based on potential profile analysis, and relates to the technical field of medical treatment and health information analysis, and the method comprises the following steps: S1, constructing a multi-dimensional intervention feature data set; s2, constructing a profile analysis model according to the feature data set; s3, performing category stability verification according to a section analysis result; s4, screening an intervention reference template according to a stability verification result; s5, matching individual features according to the intervention reference template to form a preset intervention path; s6, constructing a profile-driven path execution and feedback control system by using an intervention preset path; by setting a profile response function modeling model and combining a multi-intervention feature matrix to construct a high-resolution profile function, the problem that an existing profile division model is insufficient in continuous variable processing capacity is effectively solved, the profile structure expression capacity and individual heterogeneity description precision are improved, and the profile division efficiency is improved. The method is suitable for the fine layered modeling requirement under the driving of multi-dimensional intervention features.

Description

Technical Field

[0001] The present invention relates to the technical field of medical and health information analysis, and particularly to a palliative care intervention method for hematological malignancies based on latent profile analysis. Background Art

[0002] With the extension of the disease progression cycle of hematological malignancy patients and the increasing demand for palliative care, how to formulate targeted intervention measures based on individual differences has become an important research direction. Traditional intervention paths are mostly set based on the obvious characteristics of patients, failing to fully identify the potential heterogeneity of patient groups in intervention response and needs, and there are significant differences in intervention efficacy.

[0003] 1. In the prior art, existing models are usually established based on classification features, ignoring the subtle changes and synergistic relationships of intervention features on a continuous scale, with rough profile stratification granularity and unable to meet the needs of high-resolution intervention modeling; 2. In the prior art, the current profile verification method cannot effectively reflect whether the profile has behavioral stability and intervention response consistency during the execution of the intervention path, lacking a structured profile response function model, affecting the usability and decision support value of the profile in the actual intervention path. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a palliative care intervention method for hematological malignancies based on latent profile analysis to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a palliative care intervention method for hematological malignancies based on latent profile analysis, including the following steps: S1. Construct a multi-dimensional intervention feature data set; S2. Construct a profile analysis model according to the feature data set; S3. Perform category stability verification according to the profile analysis results; S4. Screen intervention reference templates according to the stability verification results; S5. Match individual characteristics according to the intervention reference template to form a preset intervention path; S6. Use the preset intervention path to construct a profile-driven path execution and feedback control system.

[0006] Further optimizing the technical solution, the feature data set in step S1 is represented by a standardized intervention feature matrix, and the matrix is : ; where represents the The value of the patient on the intervention feature dimension; The intervention feature set is determined according to the palliative care practice for hematological malignancies and mainly includes the following three categories: Physiological intervention features: pain score, physiological tolerance score, sleep disorder index, and the data are from clinical scales or vital sign monitoring systems; Psychological intervention features: anxiety index, emotional fluctuation range, and the data are from standardized psychological assessment tools;

[0007] To further optimize this technical solution, in step S1, all intervention features are processed by Z-score standardization through a standardized formula to eliminate the bias caused by the difference in the numerical scale of the dependent variable itself: ; where : the th original intervention feature value of the th patient; : the mean value of the th intervention feature among all patients; : the standard deviation of the th intervention feature.

[0008] To further optimize this technical solution, in step S1, data transformation is performed on the intervention variables with discrete features or discontinuous distributions through an isomorphic transformation mechanism, and the discrete variable continuous mapping formula is used: ; : the slope parameter of the mapping function, reflecting the variable sensitivity; : the mapping center point, set according to the clinical evaluation threshold.

[0009] To further optimize this technical solution, in step S2, the profile analysis model is constructed using a cointegration mapping-based latent profile modeling mechanism, and its profile cointegration mapping model is: ; where : the basic profile mean value of the th feature in the th latent profile; : the th profile, the th variable for the The response coefficient of a cointegration mapping function; : Profile The th cointegration mapping function, reflecting the non-linear structural relationship between features; : Profile The individual residual term of the th variable within.

[0010] To further optimize this technical solution, the cointegration mapping function is a non-linear structural mapping function for the standard feature set , used to identify the potential variable interaction mechanism within the profile, and is defined as follows: ; where is the response factor of the th term of the cointegration function in the rd profile to the th variable.

[0011] To further optimize this technical solution, when the profile analysis model is used, it includes the following steps: Initialize the number of profiles , setting the initial number of profiles; Estimate the profile parameters , and perform the least residual fitting on all samples under the condition of fixing , using the minimization objective function: ; Profile attribution judgment: Substitute each individual into each profile model, calculate the sum of squared residuals, and attribute it to the profile with the smallest residual; Output the profile structure information: Output the mean vector , the mapping response coefficient , and the cointegration function parameter for each profile.

[0012] To further optimize this technical solution, in step S3, according to the profile modeling mechanism in step S2, obtain the profile mapping function: ; where the profile classification basis is the corresponding to the maximum response value; Introduce the profile response deviation function for perturbation response analysis, with the initial input being , considering the patient feature perturbation , where is the perturbation amplitude of the th intervention feature: ; The further response deviation function: ; where ; that is, the original profile response value; This function reflects the response change degree of the th patient under the perturbation of the intervention feature at the profile.

[0013] To further optimize this technical solution, in step S3, the classification stability of patients under the response deviation of multiple profiles is represented by an innovative profile stability index function: ; where : represents the profile category to which the patient belongs without perturbation, that is, the maximum corresponding to ; : represents the intervention perturbation space, each dimension , is the allowable perturbation range; : represents the minimum response difference between the profile classification and the sub-optimal profile under the worst perturbation; When , it indicates that no matter how it is perturbed, the patient is still stably classified into profile , with profile stability; when , it means that there is variability in the profile classification result; when , it represents critical classification, and the profile attribution is sensitive and prone to instability.

[0014] To further optimize this technical solution, the simultaneous usage methods of the model in step S3 include: Construct a perturbation space: For each patient , generate a perturbation space , and random perturbation, gradient perturbation or systematic perturbation paths can be adopted; Calculate the profile deviation function: Recalculate for each perturbed profile , and construct ; Output the stability value : Evaluate the profile stability of each patient; Statistical distribution analysis: The whole Distribution, identify the unstable regions in the profile model, as the input condition for step S4.

[0015] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program instructions are executed by the processor, the steps of a method for palliative care intervention for hematological malignancies based on latent profile analysis as described in the first aspect of the present invention are implemented.

[0016] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program instructions are executed by the processor, the steps of a method for palliative care intervention for hematological malignancies based on latent profile analysis as described in the first aspect of the present invention are implemented.

[0017] Compared with the prior art, the present invention provides a method for palliative care intervention for hematological malignancies based on latent profile analysis, having the following beneficial effects: The method for palliative care intervention for hematological malignancies based on latent profile analysis, by setting up a profile response function modeling model and constructing a high-resolution profile function in combination with a multi-intervention feature matrix, effectively solves the problem of insufficient processing ability of existing profile division models for continuous variables, improves the profile structure expression ability and the accuracy of characterizing individual heterogeneity, and is applicable to the fine stratification modeling requirements driven by multi-dimensional intervention features; introducing a profile behavior response function and a profile intervention feedback offset function, forming a measurable and adjustable profile stability evaluation mechanism, solving the problem of the lack of response feedback on profile stability in existing verification methods, providing stability support for the construction of the path execution system, and improving the continuous effectiveness of the profile adaptation path and the intervention feedback closed-loop control ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic structural diagram of a method for palliative care intervention for hematological malignancies based on latent profile analysis proposed by the present invention; Figure 2 It is a schematic flow diagram of the profile analysis of a method for palliative care intervention for hematological malignancies based on latent profile analysis proposed by the present invention; Figure 3 It is a schematic flow diagram of the category stability verification of a method for palliative care intervention for hematological malignancies based on latent profile analysis proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following provides a detailed description of the specific embodiments of the present invention in conjunction with the accompanying drawings of the specification.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or selectively exclusive embodiments from other embodiments. Embodiment 1

[0023] Referring to Figures 1 to 3 , this is the first embodiment of the present invention. This embodiment provides a method for palliative care intervention for hematological malignancies based on latent profile analysis, including the following steps: S1. Construct a multi-dimensional intervention feature dataset; In step S1, the feature dataset is represented by a standardized intervention feature matrix, and the matrix is : ; Where represents the value of the th patient on the th intervention feature dimension; The intervention feature set is determined according to the practice of palliative care for hematological malignancies and mainly includes the following three categories: Physiological intervention features: pain score, physiological tolerance score, sleep disorder index, and the data are from clinical scales or vital sign monitoring systems; Psychological intervention features: anxiety index, emotional fluctuation range, and the data are from standardized psychological assessment tools; Treatment process indicators: blood index fluctuation coefficient during the intervention period, complication occurrence frequency, and the data are from laboratory test results and process event records; In step S1, all intervention features are processed by the standardization formula for Z-score standardization to eliminate the bias caused by the difference in the numerical scale of the dependent variable itself: ; Where, : the th patient's The original intervention eigenvalue; : The mean value of the th intervention feature among all patients; : The standard deviation of the th intervention feature; In the step S1, the intervention variables with discrete features or discontinuous distributions are transformed through an isomorphic transformation mechanism, and the discrete variable continuous mapping formula is used: ; : The slope parameter of the mapping function, reflecting the variable sensitivity; : The mapping center point, set according to the clinical evaluation threshold.

[0024] S2. Construct a profile analysis model based on the feature dataset; In the step S2, the co-integration mapping type latent profile modeling mechanism is adopted to construct the profile analysis model, and its profile co-integration mapping model is: ; Among them, : The basic profile mean value of the th feature in the th latent profile; : The response coefficient of the th variable in the th profile to the th co-integration mapping function; : The th co-integration mapping function of the profile , reflecting the non-linear structural relationship between features; : The individual residual term of the th profile within the th variable; The co-integration mapping function is a non-linear structural mapping function for the standard feature set , used to identify the potential variable interaction mechanism inside the profile, and is defined as follows: ; Among them, is the response factor of the th term of the co-integration function in the th profile to the th variable; When the profile analysis model is used, it includes the following steps: Initialize the number of profiles , and set the initial number of profiles; Estimate profile parameters , and perform minimum residual fitting on all samples under the condition of fixing , using the minimization objective function: ; Profile attribution judgment: Substitute each individual into each profile model, calculate the sum of squared residuals, and assign it to the profile with the minimum residual ; Output profile structure information: Output the mean vector , mapping response coefficient , cointegration function parameters ; S3. Verify the class stability according to the profile analysis results; In the step S3, according to the profile modeling mechanism of step S2, obtain the profile mapping function: ; Among them, the profile classification basis is the maximum response corresponding to value; Introduce the profile response deviation function for perturbation response analysis, and the initial input is , considering the patient characteristic perturbation , where is the perturbation amplitude of the th intervention characteristic: ; Further response deviation function: ; Among them, ; that is, the original profile response value; This function reflects the response change degree of the th patient under the th profile under the intervention characteristic perturbation; In the step S3, represent the classification stability of the patient under the multi-profile response deviation through the innovative profile stability index function: ; Among them, : represents the profile category to which the patient belongs without perturbation, that is, the maximum corresponding to ; : represents the intervention perturbation space, each dimension , is the allowable disturbance range; : represents the minimum response difference between the profile classification in the worst disturbance case and the sub-optimal profile; When , it indicates that regardless of the disturbance, the patient is still stably classified into profile , with profile stability; when , it means that there is variability in the profile classification result; when , it represents critical classification, where the profile attribution is sensitive and prone to instability; The simultaneous usage mode of the model in step S3 includes: Construct a disturbance space: for each patient , generate a disturbance space , and random disturbance, gradient disturbance or systematic disturbance path can be adopted; Calculate the profile deviation function: for each profile after disturbance recalculate , construct ; Output the stability value : evaluate the profile stability of each patient; Statistical distribution analysis: the overall distribution can be further analyzed to identify the unstable regions in the profile model, which are used as the input conditions for step S4.

[0025] S4. Screen the intervention reference template according to the stability verification result; After completing the profile stability assessment, in this step, according to the mean score in the consistency factor matrix, high-stability profiles are screened out as the standard intervention reference categories. The profiles in this reference group are used to establish the structural benchmark for intervention implementation, and no longer undergo modeling processing. Only the profile mean vector and the key value ranges of each index are saved in the form of a parameter table for subsequent intervention parameter matching and path formulation.

[0026] S5. Match individual characteristics based on the intervention reference template to form an intervention preset path; Based on the profile reconstruction result, establish a profile intervention path matrix , indicating the characteristic intervention preset path of the th profile in the th intervention stage.

[0027] S6. Use the intervention preset path to construct a profile-driven path execution and feedback control system; In this step, a response feedback offset control system is constructed to dynamically evaluate the deviation degree between the execution path and the actual state of the patient, and accordingly perform path fine-tuning to achieve closed-loop execution control of the path; Construct a response feedback offset function: ; Among them, : is a non-linear response offset function, used to reflect the sensitivity of the offset: ; : the th patient's actual response to the th intervention feature in the th stage; : the preset value (from S5) of the th intervention feature in the th stage corresponding to the profile path; : represents the overall offset response degree of the patient in the th stage; Construct a dynamic path adjustment function: ; Among them, : the path feedback adjustment rate (set by path fault tolerance); : the average offset direction estimation term based on the patient group: ; : the th profile patient set; : the number of patients in this profile; Apply the adjustment function to the path construction of the next stage to form a feedback closed-loop execution process: ; The formula in this step includes in the use process: Path execution stage : The system intervenes according to ; Response feedback collection: Obtain the true response of each patient ; Offset function calculation: Obtain , and evaluate the execution offset degree; Path calibration judgment: If the average offset of the group exceeds the threshold, execute path correction; Path adjustment execution: Construct the path of the next stage , and form a closed-loop control. Embodiment 2

[0028] Based on Embodiment 1, this embodiment provides a palliative care intervention system for hematological malignancies based on latent profile analysis, including the following modules: Feature Quantification Module: Corresponding to step S1, it converts into a standardized intervention variable matrix , for subsequent modeling and analysis; Profile Modeling Module: Corresponding to step S2, based on the continuous variable latent profile analysis mechanism, it constructs a patient profile classification model , realizing the identification of the implicit structure of the intervention object and profile division; Profile Stability Verification Module: Corresponding to step S3, based on the profile response function stability verification model , it analyzes the stability, robustness, and internal consistency of the profile division results, providing a basis for the availability of the profile for subsequent path construction; Profile Feature Extraction Module: Corresponding to step S4, it screens, sorts, and analyzes the differences of the key intervention features within each profile, and extracts the core vector of the profile intervention features , as the basis for path construction; Profile Path Construction Module: Corresponding to step S5, it constructs a profile feature intervention path matrix , defining the execution sequence and target values of each intervention stage, providing a specification basis for path execution; Path Execution and Feedback Control Module: Corresponding to step S6, through the feedback response offset function and the dynamic path adjustment function , it establishes a path execution feedback closed-loop mechanism to realize the automatic adjustment and optimization substitution of the path execution process. Example 3

[0029] This example also provides a computer device, applicable to a situation of a palliative care intervention method for hematological malignancies based on latent profile analysis, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement a palliative care intervention method for hematological malignancies based on latent profile analysis as proposed in the above example.

[0030] This example also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements a palliative care intervention method for hematological malignancies based on latent profile analysis as proposed in the above example.

[0031] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, trackball, or touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0032] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0033] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0034] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other suitable processing as necessary, and then storing it in a computer memory.

[0035] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0036] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A palliative care intervention method for hematological tumors based on latent profile analysis, characterized in that, It includes the following steps: S1. Construct a multi-dimensional intervention feature dataset; S2. Construct a profile analysis model according to the feature dataset; S3. Conduct category stability verification according to the profile analysis results; S4. Screen intervention reference templates according to the stability verification results; S5. Match individual characteristics based on the intervention reference templates to form a preset intervention path; S6. Use the preset intervention path to construct a profile-driven path execution and feedback control system.

2. The method for palliative care intervention of hematological tumors based on latent profile analysis according to claim 1, wherein In the step S1, the feature data set is represented by a standardized intervention feature matrix to obtain a continuous variable feature set , and the feature matrix is as follows: ； wherein represents the -th patient's value on the -th intervention feature dimension; Intervention feature set Determined according to the practice of palliative care for hematological malignancies, including the following three categories: Physiological intervention features: pain score, physiological tolerance score, sleep disorder index, and the data are from clinical scales or vital sign monitoring systems; Psychological intervention features: anxiety index, emotional fluctuation range, and the data are from standardized psychological assessment tools; Treatment process indicators: blood index fluctuation coefficient and complication occurrence frequency during the intervention period, and the data are from laboratory test results and process event records.

3. The method for palliative care intervention of hematological tumors based on latent profile analysis according to claim 2, wherein In step S1, all intervention features are processed by Z-score standardization through a standardized formula to eliminate the bias caused by the difference in the numerical scale of the dependent variable itself: ； Among them, : the th patient's th original intervention eigenvalue; : Mean value of the th intervention feature among all patients; : Standard deviation of the th intervention feature.

4. The palliative care intervention method for hematological malignancies based on latent profile analysis according to claim 3, wherein In step S1, data transformation is performed on intervention variables with discrete features or discontinuous distributions through an isomorphic transformation mechanism, and a discrete variable continuous mapping formula is used: ； : The slope parameter of the mapping function, which reflects the variable sensitivity; : The mapping center point is set according to the clinical evaluation threshold.

5. The palliative care intervention method for blood tumors based on latent profile analysis according to claim 1, wherein In step S2, a co-integration mapping type latent profile modeling mechanism is adopted to construct the profile analysis model, and its profile co-integration mapping model is: ； Among them, : the basic profile mean value of the th feature in the th potential profile; : The -th variable in the -th profile's response coefficient to the -th cointegration mapping function; : Cross-section The th cointegration mapping function, reflecting the non-linear structural relationship between features; : Profile The Individual residual term of the variable.

6. The palliative care intervention method for hematological malignancies based on latent profile analysis according to claim 5, wherein The cointegration mapping function is a non-linear structure mapping function for the standard feature set and is used to identify the potential variable interaction mechanism inside the profile and is defined as follows: ； Among them, is the response factor of the th term of the cointegration function in the th profile with respect to the 7. A palliative care intervention method for hematological malignancies based on latent profile analysis according to claim 5, characterized in that, When the profile analysis model is used, it includes the following steps: Initialize the number of profiles , and set the initial number of profiles; Estimated profile parameters , for a fixed , perform a least-residual fit on all samples using the minimization objective function: ； Profile attribution judgment: For each individual Substitute into each profile model, calculate the sum of squared residuals, and assign it to the profile with the smallest residual ; Output profile structure information: Output the mean vector for each profile , mapping response coefficient , cointegration function parameter .

8. A method for palliative care intervention of hematological tumors based on latent profile analysis according to claim 1, characterized in that, In step S3, according to the profile modeling mechanism in step S2, a profile mapping function is obtained: ； Among them, the basis for cross-section classification is the maximum response corresponding to value; The perturbation response analysis is carried out by introducing the sectional response deviation function, and the initial input is , considering the perturbation of patient characteristics , where is the perturbation amplitude of the th intervention characteristic: ； Further response deviation function: ； Among them, ; that is, the original profile response value; This function reflects the degree of response change of the th patient under the profile under the perturbation of intervention characteristics.

9. The method for palliative care intervention of hematological tumors based on latent profile analysis according to claim 8, wherein In step S3, the classification stability of patients under multi-profile response deviation is represented by an innovative profile stability index function: ； Among them, : represents the profile category to which the patient belongs when there is no perturbation, that is, the maximum corresponding to ; : represents the intervention perturbation space, each dimension , is the allowable perturbation range; : represents the minimum response difference between the profile classification and the sub-optimal profile under the worst disturbance; When it indicates that no matter how much perturbation there is, the patient is still stably classified into profile , with profile stability; when , it means that there is variability in the profile classification result; when , it represents critical classification, with the profile attribution being sensitive and prone to instability.

10. A method for palliative care intervention of hematological tumors based on latent profile analysis according to claim 8, characterized in that, The simultaneous usage methods of the model in step S3 include: Construct a perturbation space: For each patient , generate a perturbation space , and adopt a random perturbation, gradient perturbation, or systematic perturbation path; Calculate the profile deviation function: For each profile after perturbation recalculate , construct ; Output stability value : Evaluate the profile stability of each patient; Statistical distribution analysis: Further analyze the whole distribution, identify the unstable regions in the profile model, and use them as the input conditions for step S4.

Images