Complication risk assessment method and system for multiple myeloma

Abstract

The invention provides a multiple myeloma complication risk assessment method and system, and the method comprises the steps: determining a complication risk feature set of multiple myeloma based on continuously updated literature crawling information; based on the complication risk feature set, determining a monitoring index set of multiple myeloma through a large language model; and performing clinical monitoring based on the monitoring index set, and when the serum total calcium parameter is higher than a first preset threshold value, and / or the number of monitoring indexes deviating from a preset range in the monitoring index set is greater than or equal to a second preset threshold value, triggering complication risk prompt early warning. The method has the beneficial effects that the complication risk feature set is determined and the examination indexes corresponding to the complication are dynamically adjusted by collecting and updating the continuous information of the complication of the multiple myeloma, so that the patient can be helped to find potential complication risks in time based on the regular examination result of the patient; and the timeliness, the accuracy and the clinical practicability of judging the multiple myeloma complications are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of evaluation technology, and in particular relates to a method and system for evaluating the risk of complications of multiple myeloma. Background Art

[0002] Multiple Myeloma (MM) is a complex blood cancer with a variety of complications and is often difficult to detect and diagnose in the early stages. The disease is characterized by the abnormal proliferation of plasma cells, which form tumors in the bone marrow, affecting the production of normal blood cells and leading to a series of serious health problems.

[0003] The complications of multiple myeloma are often not obvious in the early stages, and the symptoms are similar to those of other common diseases, which can easily lead to misdiagnosis or delayed treatment. How to promptly and accurately judge and respond to possible complications of multiple myeloma is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0004] In view of this, the present invention aims to propose a method and system for risk assessment of complications of multiple myeloma, which can achieve timely discovery and monitoring of complications of multiple myeloma.

[0005] The first aspect of the present invention provides a complication risk assessment method for multiple myeloma, comprising: determining a complication risk feature set for multiple myeloma based on continuously updated literature crawling information; determining a monitoring indication set for possible complications through a large language model based on the complication risk feature set, the monitoring indication set including at least a total serum calcium parameter corresponding to a preset sliding window; performing clinical monitoring based on the monitoring indication set, and triggering a complication risk warning when the total serum calcium parameter is higher than a first preset threshold, and / or the number of monitoring indications in the monitoring indication set that deviate from a preset range is greater than or equal to a second preset threshold.

[0006] In a possible implementation of the first aspect above, the complication risk feature set includes possible complications corresponding to multiple myeloma, concurrent probabilities corresponding to possible complications, and precursory indications corresponding to possible complications; the acquisition of literature crawling information includes: updating possible complications corresponding to multiple myeloma based on continuously updated multiple myeloma-related literature and clinical data; determining concurrent probabilities and / or precursory indications corresponding to possible complications based on possible complications and a preset first knowledge database; the first knowledge database stores clinical research data information related to multiple myeloma.

[0007] In a possible implementation of the above first aspect, based on the set of complication risk characteristics, a set of monitoring indicators for possible complications is determined through a large language model, including: generating corresponding complication monitoring prompt words based on semantic analysis of the set of complication risk characteristics; generating a set of monitoring indicators through a first large language model based on the complication monitoring prompt words, where the set of monitoring indicators includes multiple monitoring indicators, and each monitoring indicator corresponds to at least one possible complication; wherein, when generating the set of monitoring indicators, the first large language model at least utilizes the content of a second knowledge database, and the second knowledge database stores at least research data information related to the set of complication risk characteristics.

[0008] In a possible implementation of the above first aspect, when generating the set of monitoring indicators, the first large language model at least utilizes the content of a second knowledge database, including: performing vectorized encryption processing on the complication monitoring prompt words to obtain a first encrypted vector; matching the first encrypted vector with the encrypted knowledge vector corresponding to the research data information stored in the second knowledge database to determine an associated knowledge vector associated with the complication monitoring prompt words; performing vectorized decryption processing on the associated knowledge vector to obtain the associated research data information corresponding to the associated knowledge vector; and updating the complication monitoring prompt words according to the associated research data information, so that the updated complication monitoring prompt words at least include the associated research data information.

[0009] In a possible implementation of the above first aspect, the monitoring indicators include at least one or a combination of multiple of the hemoglobin parameter, serum M protein parameter, plasma viscosity parameter, serum creatinine parameter, urinary microalbumin parameter, urinary β2 microglobulin parameter, urinary calcium excretion parameter, serum free light chain ratio parameter, troponin parameter, serum ferritin parameter, and hemoglobin parameter corresponding to a preset sliding window.

[0010] In a possible implementation of the above first aspect, the method further includes: when the serum total calcium parameter is higher than a first preset threshold, and / or the weighted fusion parameter corresponding to the set of monitoring indicators is greater than or equal to a third preset threshold, triggering a complication risk prompt warning, where the weighted fusion parameter is determined based on the monitoring indicators in the set of monitoring indicators and the weighted fusion coefficients corresponding to the monitoring indicators; the weighted fusion coefficients corresponding to the monitoring indicators are positively correlated with the concurrent probabilities of the possible complications corresponding to the monitoring indicators.

[0011] The second aspect of the present invention provides a complication risk assessment system for multiple myeloma, which is applied to the method provided in the foregoing first aspect, and includes: a complication update unit for determining a set of complication risk characteristics of multiple myeloma based on continuously updated literature crawling information; a complication monitoring unit for determining a set of monitoring indicators for possible complications through a large language model based on the set of complication risk characteristics, and the set of monitoring indicators at least includes serum total calcium parameters corresponding to a preset sliding window; a risk prompt unit for performing clinical monitoring based on the set of monitoring indicators, and when the serum total calcium parameter is higher than a first preset threshold, and / or the number of monitoring indicators deviating from the preset range in the set of monitoring indicators is greater than or equal to a second preset threshold, triggering a complication risk prompt warning.

[0012] The third aspect of the present invention provides an electronic device, including a processor and a memory communicatively connected to the processor and used for storing instructions executable by the processor, and the processor is used to execute a complication risk assessment method for multiple myeloma.

[0013] The fourth aspect of the present invention provides a server, including at least one processor and a memory communicatively connected to the processor, and the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to enable the at least one processor to execute a complication risk assessment method for multiple myeloma.

[0014] The fifth aspect of the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, it implements a complication risk assessment method for multiple myeloma.

[0015] Compared with the prior art, a complication risk assessment method and system for multiple myeloma of the present invention have the following beneficial effects: By continuously collecting and updating information on complications of multiple myeloma, determining a set of complication risk characteristics and dynamically adjusting the corresponding examination indicators for complications, based on the regular examination results of patients, it can help patients timely discover potential complication risks, and significantly improve the timeliness, accuracy and clinical practicability of the judgment of multiple myeloma complications. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0017] Figure 1 It is a schematic flowchart of a complication risk assessment method for multiple myeloma shown in an embodiment of the present invention;

[0018] Figure 2A flowchart showing the process of obtaining a set of complication risk features according to an embodiment of the present invention;

[0019] Figure 3 A flowchart showing the process of determining a set of monitoring indicators for possible complications according to an embodiment of the present invention. Detailed implementation manners

[0020] To more clearly illustrate the technical solutions of the embodiments of this specification, the embodiments will be introduced in detail below with reference to the accompanying drawings. Obviously, the content described below is some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, the technical solutions or means disclosed in this specification can also be applied to other scenarios based on this technical content.

[0021] Figure 1 A flowchart showing a method for assessing the complication risk of multiple myeloma according to an embodiment of the present invention, which may specifically include:

[0022] Step 110: Based on continuously updated literature crawling information, determine a set of complication risk features of multiple myeloma.

[0023] In some embodiments, considering that different patients with multiple myeloma have different clinical manifestations, and the differences in the physical conditions of different patients will also lead to different complications corresponding to multiple myeloma. In order to obtain the possible complications corresponding to multiple myeloma in a general statistical sense, and to be aware of and establish a corresponding risk warning mechanism in a timely manner when new possible complications corresponding to multiple myeloma appear, it is possible to continuously crawl the updated research literature and / or clinical data related to multiple myeloma as literature crawling information, and determine the required set of complication risk features by screening and extracting the valid information in the literature crawling information.

[0024] In some embodiments, the set of complication risk features may include the possible complications corresponding to multiple myeloma, the concurrent probabilities corresponding to the possible complications, and the precursor indicators corresponding to the possible complications. In some embodiments, a possible complication refers to another disease or symptom that occurs or appears during the disease development or treatment process of a patient diagnosed with multiple myeloma, which may be caused by the multiple myeloma disease itself, or may be caused by the drugs or treatment measures during the treatment process of multiple myeloma and the changes in the patient's physical condition.

[0025] In some embodiments, the concurrent probability corresponding to a possible complication refers to the probability of a certain possible complication occurring in a patient diagnosed with multiple myeloma. In some embodiments, the concurrent probability may not only correspond to possible complications, but also to different groups of multiple myeloma patients. For example, there are differences in the probabilities of the same possible complication occurring in patients of different age groups, and there are also differences in the probabilities of the same possible complication occurring in patients with different underlying diseases, which are not limited herein.

[0026] In some embodiments, the precursor indication corresponding to a possible complication refers to the judgment conditions and judgment methods for determining whether a patient has a possible complication. For example, when the possible complication is hypercalcemia, the corresponding precursor indications may include, but are not limited to, the judgment condition that the total serum calcium parameter is higher than a first preset threshold, where the first preset threshold may be set to 2.75 mmol / L or 2.7 mmol / L, which is not limited herein.

[0027] By crawling continuously updated literature and clinical data related to multiple myeloma, it is possible to timely learn about the occurrence of possible complications corresponding to multiple myeloma in clinical practice.

[0028] Figure 2 It is a schematic flowchart of a process for obtaining a set of complication risk characteristics according to an embodiment of the present invention, which may specifically include:

[0029] Step 210: Update the possible complications corresponding to multiple myeloma based on continuously updated multiple myeloma-related literature and clinical data. In some embodiments, it is possible to crawl multiple myeloma-related literature and clinical data that are updated continuously over time, where the relevant literature and clinical data may specifically include research articles published in various academic journals related to multiple myeloma, academic books published by publishing houses, conference papers containing research results submitted at academic conferences, records of clinical research cases, clinical practice guidelines summarized and updated regularly or irregularly, patent application texts and authorized texts containing research results related to multiple myeloma, etc. By crawling the above-mentioned multiple myeloma-related literature and clinical data regularly or irregularly, a large amount of the latest data information related to multiple myeloma can be obtained. By screening the above information and sorting out the complication-related information, the continuously updated possible complications corresponding to multiple myeloma can be obtained, such as hypercalcemia, pathological bone damage, hyperviscosity syndrome, neurological compression, etc., which are not limited herein. Based on the above step 210, it is possible to timely monitor and update the types of complications corresponding to multiple myeloma newly emerging in the field of clinical research, and it is possible to timely respond to the complication situation of patients when patients present similar situations, greatly improving the life protection of patients.

[0030] Step 220: Based on the possible complications and the preset first knowledge database, determine the concurrent probability and / or precursor indications corresponding to the possible complications. The first knowledge database stores clinical research data information related to multiple myeloma. In some embodiments, when determining the disease type corresponding to the possible complications, the concurrent probability and / or precursor indications corresponding to the possible complications can be determined based on the preset first knowledge database. For example, for hypercalcemia among the possible complications, based on the clinical research data information stored in the first knowledge database, it is known that its concurrent probability as a complication of multiple myeloma in medically developed regions such as Europe and the United States is in the range of 30% to 50%, and the clinical manifestation behaviors include vomiting, nausea, anorexia, polydipsia and polyuria, etc. Based on this, the concurrent probability and precursor indications of the possible complication of hypercalcemia can be determined.

[0031] Step 120: Based on the set of complication risk characteristics, determine the set of monitoring indications for multiple myeloma through a large language model.

[0032] In some embodiments, for the determined set of complication risk characteristics, a large language model can be used to determine the corresponding monitoring indications. The large language model (LLM) is a type of natural language processing model based on deep learning that has been trained with a large amount of text data and has the ability to process and generate natural language text. It can determine the corresponding monitoring indications based on information such as the determined complication type, concurrent probability, and precursor indications. Figure 3 It is a schematic flowchart of a process for determining a set of monitoring indications for possible complications according to an embodiment of the present invention, and specifically may include:

[0033] Step 310: Generate corresponding complication monitoring prompt words based on the semantic analysis of the set of complication risk characteristics. In some embodiments, based on a preset prompt word template and combined with the semantic analysis of the set of complication risk characteristics, standard complication monitoring prompt words for use by the large language model can be generated. The semantic analysis of the set of complication risk characteristics may include appropriate expansion of the possible complications. For example, when the possible complication is pathological bone injury, the corresponding complication monitoring prompt word may be "Generate monitoring indications for the pathological bone injury complication of multiple myeloma, where the pathological bone injury includes bone pain type injuries and fracture type injuries, and determine the monitoring indications from the dimensions of blood test indicators, patient feedback information, and CT image analysis respectively."

[0034] Step 320: Based on the complication monitoring prompt words, generate a set of monitoring indicators through the first large language model, where the set of monitoring indicators includes multiple monitoring indicators, and each monitoring indicator corresponds to at least one possible complication. It can be understood that the monitoring indicators included in the set of monitoring indicators can correspond to one or more possible complications. When a monitoring indicator corresponds to multiple possible complications, effective monitoring of multiple possible complications can be achieved through the reuse of a single monitoring indicator.

[0035] In some embodiments, in the specific implementation of the foregoing step 320, the first large language model uses at least the content of the second knowledge database when generating the set of monitoring indicators. At least research data information related to the set of complication risk characteristics is stored in the second knowledge database. It can be understood that in order to avoid the problem of hallucination in the processing of the relevant professional knowledge of multiple myeloma by the large language model, that is, the content generated by the large language model seems reasonable on the surface, but the actual content is fictional or inaccurate information, a second knowledge database containing research data information related to the set of complication risk characteristics can be connected to the first large language model. In some embodiments, the foregoing second knowledge database may be the same as the first knowledge database provided in the foregoing embodiments, which is not limited herein. In some embodiments, the research data information related to the set of complication risk characteristics stored in the second knowledge database can be updated regularly or irregularly, so as to effectively avoid the problem of hallucination in the generation process of sub-specialty subject words by the large language model due to newly emerging professional knowledge.

[0036] In some embodiments, considering that there may be some encrypted information content that needs to be kept confidential in the research data information related to the set of complication risk characteristics stored in the second knowledge database, when the first large language model uses the content of the second knowledge database to generate the set of monitoring indicators, a vectorized encryption matching method can be adopted: that is, first perform vectorized encryption processing on the complication monitoring prompt words to obtain a first encrypted vector; then match the first encrypted vector with the encrypted knowledge vector corresponding to the research data information stored in the second knowledge database to determine the associated knowledge vector associated with the complication monitoring prompt words; then perform vectorized decryption processing on the associated knowledge vector to obtain the associated research data information corresponding to the associated knowledge vector; finally, update the complication monitoring prompt words according to the associated research data information, so that the updated complication monitoring prompt words at least include the associated research data information. In some embodiments, the vectorized encryption processing methods used in the process of obtaining the first encrypted vector and the encrypted knowledge vector can be the same. Through the method of encrypted vector matching, the risk of leakage of the encrypted information content that needs to be kept confidential stored in the second knowledge database can be avoided directly in the process of obtaining the monitoring indicators.

[0037] In some embodiments, considering that in actual clinical practice, hypercalcemia is a typical complication of multiple myeloma, and the risk probability is relatively high and the impact on patients is relatively large during the current clinical diagnosis and treatment process, therefore, during the determination of the monitoring indication set, it can at least include the total serum calcium parameter corresponding to a preset sliding window, which is used to characterize this possible complication of hypercalcemia. In some embodiments, when the total serum calcium parameter is higher than the first preset threshold, it can be determined that the patient may have a hypercalcemia complication.

[0038] In some embodiments, the monitoring indications can also include at least one or a combination of multiple ones among the hemoglobin parameter, serum M protein parameter, plasma viscosity parameter, serum creatinine parameter, urinary microalbumin parameter, urinary β2-microglobulin parameter, urinary calcium excretion parameter, serum free light chain ratio parameter, troponin parameter, serum ferritin parameter, and hemoglobin parameter corresponding to a preset sliding window. Among them, when the hemoglobin parameter of male patients is less than 110 g / L or the hemoglobin parameter of female patients is less than 100 g / L (for females) and the serum ferritin parameter is in the normal or high range, it can be determined that the patient may have a multiple myeloma anemia complication different from iron deficiency anemia; when the serum M protein parameter is greater than 30 g / L and the plasma viscosity parameter is greater than 1.8 mPa·s, it can be determined that the patient may have a hyperviscosity syndrome complication; when the total serum calcium parameter is greater than 2.75 mmol / L, the serum creatinine parameter is greater than 177 μmol / L47, and the urinary calcium excretion parameter is greater than 7.5 mmol / 24 h, it can be determined that the patient may have a hypercalcemia complication, which is not limited here. In some embodiments, the monitoring indications can also include a description of the patient's physical condition, such as abnormal physical conditions like thirst, polyuria, nausea, constipation, etc. Through the relevant descriptions of the above embodiments, it is possible to timely detect the possible complication situation of the patient through the setting of the monitoring indications combined with the patient's regular examination items and give a timely treatment response.

[0039] Step 130: Perform clinical monitoring based on the monitoring indication set. When the total serum calcium parameter is higher than the first preset threshold, and / or the number of monitoring indications deviating from the preset range in the monitoring indication set is greater than or equal to the second preset threshold, trigger a complication risk warning.

[0040] In some embodiments, when the number of monitoring indications deviating from the preset range in the monitoring indication set is greater than or equal to the second preset threshold, it indicates that the patient has deviations in multiple monitoring indications, and the risk of possible complications is relatively high. A complication risk warning can be triggered in a timely manner for a timely treatment response.

[0041] In some embodiments, in addition to providing a complication risk prompt and warning based on the number of monitoring indicators deviating from a preset range, a complication risk prompt and warning can also be provided based on the weighted fusion parameters corresponding to the set of monitoring indicators. Specifically, when the weighted fusion parameters corresponding to the set of monitoring indicators are greater than or equal to a third preset threshold, a complication risk prompt and warning can be triggered, where the weighted fusion parameters are determined based on the monitoring indicators in the set of monitoring indicators and the weighted fusion coefficients corresponding to the monitoring indicators. It can be understood that the higher the concurrent probability of a possible complication, the greater the weighted proportion of the monitoring indicators corresponding to the complication, and the higher the value of the corresponding weighted fusion coefficient, that is, the weighted fusion coefficient corresponding to the monitoring indicator is positively correlated with the concurrent probability of the possible complication corresponding to the monitoring indicator.

[0042] Some embodiments of the present invention also provide a complication risk assessment system for multiple myeloma, which is applied to the method provided in the foregoing embodiments, and specifically may include: a complication update unit for determining a set of complication risk characteristics of multiple myeloma based on continuously updated literature crawling information; a complication monitoring unit for determining a set of monitoring indicators of possible complications through a large language model based on the set of complication risk characteristics, where the set of monitoring indicators at least includes a serum total calcium parameter corresponding to a preset sliding window; and a risk prompt unit for performing clinical monitoring based on the set of monitoring indicators, and triggering a complication risk prompt and warning when the serum total calcium parameter is higher than a first preset threshold, and / or the number of monitoring indicators deviating from the preset range in the set of monitoring indicators is greater than or equal to a second preset threshold. It can be understood that each functional module in the above complication risk assessment system is implemented using the same process steps as the method provided in the foregoing embodiments, and is not limited herein.

[0043] Some embodiments of the present invention also provide an electronic device, including a processor and a memory communicatively connected to the processor and used to store instructions executable by the processor, where the processor is used to execute a complication risk assessment method for multiple myeloma.

[0044] Some embodiments of the present invention also provide a server, including at least one processor and a memory communicatively connected to the processor, where the memory stores instructions executable by the at least one processor, and when the instructions are executed by the processor, the at least one processor is caused to execute a complication risk assessment method for multiple myeloma.

[0045] Some embodiments of the present invention also provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, a complication risk assessment method is implemented.

[0046] In summary, through the technical solution provided by the present invention, it is possible to continuously collect and update information on the complications of multiple myeloma, determine the complication risk feature set and dynamically adjust the inspection indicators corresponding to the complications. Based on the regular examination results of the patients, it can help the patients to timely detect potential complication risks, significantly improving the timeliness, accuracy and clinical practicability of the judgment of multiple myeloma complications.

[0047] The above content is a further detailed description of the present disclosure in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present disclosure is only limited to these descriptions. For those of ordinary skill in the technical field to which the present disclosure belongs, without departing from the concept of the present disclosure, several simple deductions or substitutions can still be made, which should all be regarded as belonging to the protection scope of the present disclosure.

Claims

1. A method for assessing the risk of complications of multiple myeloma, characterized in that, Including: Based on continuously updated literature crawling information, determine the set of complication risk characteristics of the multiple myeloma; Based on the set of complication risk characteristics, determine the set of monitoring indicators for possible complications through a large language model, where the set of monitoring indicators at least includes the total serum calcium parameter corresponding to a preset sliding window; Based on the set of monitoring indicators, conduct clinical monitoring. When the total serum calcium parameter is higher than the first preset threshold, and / or the number of monitoring indicators deviating from the preset range in the set of monitoring indicators is greater than or equal to the second preset threshold, trigger a complication risk prompt warning.

2. The complication risk assessment method for multiple myeloma according to claim 1, wherein The set of complication risk characteristics includes the possible complications corresponding to the multiple myeloma, the concurrent probabilities corresponding to the possible complications, and the precursor indicators corresponding to the possible complications; The acquisition of the literature crawling information includes: Based on continuously updated multiple myeloma-related literature and clinical data, update the possible complications corresponding to the multiple myeloma; Based on the possible complications and a preset first knowledge database, determine the concurrent probabilities and / or the precursor indicators corresponding to the possible complications; The first knowledge database stores clinical research data information related to the multiple myeloma.

3. The method for assessing the risk of complications of multiple myeloma according to claim 2, wherein: The determining of the set of monitoring indicators for the possible complications through the large language model based on the set of complication risk characteristics includes: Based on the semantic analysis of the set of complication risk characteristics, generate corresponding complication monitoring prompt words; Based on the complication monitoring prompt words, generate the set of monitoring indicators through a first large language model, where the set of monitoring indicators includes multiple monitoring indicators, and each monitoring indicator corresponds to at least one of the possible complications; Wherein, when generating the set of monitoring indicators, the first large language model at least uses the content of a second knowledge database, and the second knowledge database stores at least research data information related to the set of complication risk characteristics.

4. The complication risk assessment method for multiple myeloma according to claim 3, characterized in that The first large language model at least uses the content of the second knowledge database when generating the set of monitoring indicators, including: Perform vectorized encryption processing on the complication monitoring prompt words to obtain a first encrypted vector; Match the first encrypted vector with the encrypted knowledge vectors corresponding to the research data information stored in the second knowledge database to determine the associated knowledge vector associated with the complication monitoring prompt words; Perform vectorized decryption processing on the associated knowledge vector to obtain the associated research data information corresponding to the associated knowledge vector; Update the complication monitoring prompt words according to the associated research data information, so that the updated complication monitoring prompt words at least include the associated research data information.

5. The complication risk assessment method for multiple myeloma according to claim 3, wherein The monitoring indicators include at least one or a combination of multiple of the hemoglobin parameter, serum M protein parameter, plasma viscosity parameter, serum creatinine parameter, urinary microalbumin parameter, urinary β2-microglobulin parameter, urinary calcium excretion parameter, serum free light chain ratio parameter, troponin parameter, serum ferritin parameter, hemoglobin parameter corresponding to a preset sliding window.

6. The method for assessing the complication risk of multiple myeloma according to claim 5, wherein The method further includes: When the total serum calcium parameter is higher than the first preset threshold, and / or the weighted fusion parameter corresponding to the set of monitoring indicators is greater than or equal to the third preset threshold, the complication risk prompt warning is triggered, where the weighted fusion parameter is determined based on the monitoring indicators in the set of monitoring indicators and the weighted fusion coefficients corresponding to the monitoring indicators; The weighted fusion coefficient corresponding to the monitoring indicator is positively correlated with the concurrent probability of the possible complication corresponding to the monitoring indicator.

7. A complication risk assessment system for multiple myeloma, applied to the complication risk assessment method for multiple myeloma according to any one of claims 1 to 6, characterized in that, Comprising: A complication update unit, configured to determine a set of complication risk characteristics of the multiple myeloma based on continuously updated literature crawling information; A complication monitoring unit, configured to determine a set of monitoring indicators of possible complications through a large language model based on the set of complication risk characteristics, where the set of monitoring indicators at least includes the total serum calcium parameter corresponding to a preset sliding window; A risk prompt unit, configured to perform clinical monitoring based on the set of monitoring indicators, and trigger a complication risk prompt warning when the total serum calcium parameter is higher than the first preset threshold, and / or the number of monitoring indicators deviating from the preset range in the set of monitoring indicators is greater than or equal to the second preset threshold.

8. An electronic device, comprising a processor and a memory communicatively connected to the processor and configured to store executable instructions of the processor, wherein: The processor is configured to execute the complication risk assessment method of multiple myeloma according to any one of claims 1 to 6 above.

9. A server, characterized in that: Comprising at least one processor, and a memory communicatively connected to the processor, where the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to enable the at least one processor to execute the complication risk assessment method of multiple myeloma according to any one of claims 1 to 6.

10. A computer-readable storage medium stores a computer program, characterized in that: When the computer program is executed by the processor, it implements the complication risk assessment method of multiple myeloma according to any one of claims 1 to 6.

Images