Tumor patient individualized medication dosage optimization method

Abstract

The invention discloses a tumor patient individualized medication dose optimization method, which comprises the following steps: establishing a patient file, calculating an initial dose based on evidence-based knowledge, making a structured follow-up visit plan, integrating data from communities, families and patient reports, analyzing by using a fusion rule and an intelligent algorithm, and generating a dose optimization suggestion. After doctor-patient auditing, a situational medication reminding and recording tool is automatically updated and pushed, and finally continuous loop feedback and iteration of a treatment scheme are achieved; according to the invention, long-term follow-up visit and dosage decision are bound, and medication reminding, patient education and clinical support are combined, so that real-time, accurate and individualized adjustment of the medication dosage is effectively realized in a basic scene, and the treatment safety and compliance are improved.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and in particular to a method for optimizing individualized drug dosages for cancer patients. Background Technology

[0002] In personalized treatment of cancer patients, clinicians usually formulate initial medication plans based on static initial data such as the patient's body surface area, liver and kidney function and specific gene test results, and with reference to authoritative treatment guidelines. Some hospital information systems or mobile health applications also have medication reminders and individual follow-up appointment functions to improve patient compliance.

[0003] Existing solutions often focus on initial dose calculation or isolated reminder functions, lacking a systematic approach that establishes a closed loop between "long-term, dynamic follow-up monitoring data" and "dosage re-optimization." Patient condition changes, toxic reactions, and simple testing data outside the hospital, especially in the community, are difficult to collect effectively and structurally and fed back into dosage adjustment decisions. This leads to delayed dosage adjustments and prevents truly personalized treatment based on overall response. Furthermore, common medication reminder tools are isolated from patients' clinical records, follow-up plans, and dosage adjustment logic. Reminders only inform when to take the medication but do not connect with integrated aspects of education and treatment, such as why the dosage is taken, how to record and report adverse reactions, etc. Patient compliance data cannot be automatically fed back for evaluating efficacy and safety, resulting in inefficient long-term medication management.

[0004] Therefore, in response to the problems mentioned above, this invention proposes a method for optimizing individualized drug dosage for cancer patients. Summary of the Invention

[0005] To overcome the lack of dosage optimization mechanisms and medication reminder management based on long-term follow-up in existing technologies, this invention proposes a method for individualized medication dosage optimization for cancer patients, which can realize individualized adjustment and full-cycle management of medication dosage for cancer patients.

[0006] The technical solution of this invention is: a method for optimizing individualized drug dosage for cancer patients, comprising the following steps: S1 collects patients' initial clinical data by combining access to a regional medical information platform with manual entry through electronic forms. This initial clinical data includes at least tumor type, pathological stage, gene testing results, liver and kidney function indicators, body surface area, concomitant medications, and previous treatment history.

[0007] S2, based on the patient's individualized medical record, calls upon the built-in evidence-based medicine knowledge base and pharmaceutical database, and combines the patient's body surface area and liver and kidney function indicators to calculate and generate an initial medication dosage plan. The evidence-based medicine knowledge base and pharmaceutical database integrate pharmacokinetic data based on racial differences, evidence of the impact of specific gene polymorphisms on drug efficacy or toxicity, and dosage recommendations from authoritative clinical practice guidelines. The medication dosage plan includes the drug name, single dose, dosing frequency, route of administration, and duration of the first treatment cycle.

[0008] S3 generates a structured long-term follow-up plan with key time points based on the initial dosage regimen and treatment goals. This follow-up plan includes the next follow-up date, specific laboratory tests that need to be repeated, and imaging assessment time points. At the same time, it generates a matching patient self-symptom record list and converts the list into a questionnaire with graphical symptom options, thereby reducing the patient reporting threshold and improving the degree of data standardization.

[0009] S4. At the time points set in the follow-up plan, follow-up data of patients will be collected through online platforms, community health points, or simple home testing devices. The follow-up data includes: patient self-reported symptoms and signs and treatment compliance records collected through questionnaires, laboratory retest results obtained through community health points, and physiological indicators obtained through simple home testing devices; among which simple home testing devices include at least a certified portable coagulation analyzer.

[0010] S5 integrates and analyzes the collected follow-up data with the initial clinical data and initial dosage regimen, and evaluates them using a pre-set dosage adjustment algorithm model. This dosage adjustment algorithm model integrates a Bayesian feedback model based on pharmacokinetic or pharmacodynamic principles and a rule-based expert system. When the analysis results meet the pre-set adjustment conditions, it generates dosage optimization adjustment suggestions. These adjustment conditions include the occurrence of toxic reactions of a specific grade or higher, failure of key efficacy markers to meet expectations, or clinically significant fluctuations in liver and kidney function. The adjustment suggestions include dose increase or decrease, change of dosing interval, or adjunctive supportive treatment.

[0011] S6 automatically generates a diagnostic and treatment support briefing for clinicians to review, based on dosage optimization and adjustment suggestions, relevant evidence-based summaries, and recent key patient data, and pushes it to the terminal of the designated responsible medical staff for final confirmation.

[0012] S7 automatically generates a visual medication reminder schedule based on the confirmed initial dosage regimen or dosage optimization adjustment suggestions, and pushes it to the patient's terminal via at least one of the following methods: SMS, application push, or voice call. The reminder content includes a drug image, precautions for taking the medication, and emergency handling tips. It also provides an electronic medication record interface for patients to record their actual medication situation and subjective feelings. When an abnormal medication event or severe discomfort symptoms are recorded, the system automatically triggers an alert and notifies the responsible medical staff.

[0013] S8. Use the compliance data recorded in the electronic medication record interface and the newly collected follow-up data as input for the next round of analysis. Repeat steps S5 to S7 to achieve cyclical optimization and long-term management of the medication dosage plan.

[0014] S9, throughout the entire process, the system automatically matches and pushes concise science education materials based on the patient's current treatment stage and common misconceptions, and provides brief science tips on the core precautions or possible side effects of the current medication when generating medication reminders or follow-up plans.

[0015] The beneficial effects of this invention are: 1. This invention utilizes continuous data obtained from community follow-ups and home devices to drive a built-in hybrid algorithm model for real-time analysis, proactively identifying dosage adjustment needs and generating suggestions. After confirmation by doctors and patients, the treatment plan is automatically updated and reminders are sent, realizing the change from a fixed plan to an adaptive plan, which greatly improves the safety and sustainability of medication efficacy.

[0016] 2. This invention links each medication reminder generated with the patient's current specific dosage regimen, recent adverse reaction data, and coping knowledge. This ensures that the reminder not only tells when to take the medication but also explains why it should be taken this way and what to do if problems arise. At the same time, every medication record and feedback from the patient can be fed back into the system and influence subsequent decisions, thereby significantly improving management efficiency and patient compliance.

[0017] 3. This invention transforms complex oncology pharmacy management into a task that can be operated by community healthcare workers and participated in by patients' families by combining standardized electronic forms, graphical symptom reports, portable monitoring devices and cloud-based intelligent algorithms. Furthermore, through popular science education throughout the process, it provides a good technical direction for promoting the chronic disease management of cancer. Attached Figure Description

[0018] Figure 1 The diagram shown is a schematic representation of the system framework of the present invention. Figure 2 The diagram shown is a schematic representation of the method flow of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 and Figure 2 This invention provides an embodiment: a method for optimizing individualized drug dosage for cancer patients. The method of this invention is implemented through an oncology medication management ecosystem, with each step supported by a corresponding functional module within the system. This system includes a patient personalized medical record module, an initial dose calculation and recommendation module, a follow-up plan and tool module, a follow-up data collection and integration module, a dose optimization analysis and adjustment suggestion generation module, a diagnosis and treatment auxiliary report and review module, a personalized medication reminder and recording tool module, a cyclical feedback and protocol iteration module, and a science popularization and support module. Specifically: S1, the patient personalized medical record module, first needs to establish a secure data interface with the regional medical information platform. After patient authorization, it automatically retrieves available standardized historical data, such as pathology reports, parsed results of genetic testing PDFs, and historical lab reports. For information that cannot be automatically retrieved or needs to be supplemented in the community, the system provides multi-level electronic forms. The first-level form contains mandatory core fields, completed by community doctors during consultations. The second-level form contains detailed medical history and baseline conditions, which can be filled out by the patient's family on a tablet under the guidance of a nurse. The form design uses logical jumps (e.g., selecting "no history of diabetes" skips related detailed questions) and includes numerous help icons. All entered data undergoes logical validation and storage, ultimately forming a record containing static baseline data and a dynamically updated area. The record's homepage displays key information in a visual interface, allowing medical staff to quickly grasp the overall picture of the patient's condition.

[0021] S2, the initial dose calculation and recommendation module, incorporates an evidence-based medicine knowledge base—a multi-dimensional, expert-calibrated database. For example, for a specific chemotherapy drug, the knowledge base not only contains standard body surface area calculation formulas and liver function-based dose adjustment tables, but also links to relevant literature on drug-metabolizing enzyme gene polymorphisms. When a doctor develops an irinotecan regimen for a patient with advanced colorectal cancer, the system automatically calculates the standard dose based on the patient's body surface area, bilirubin level, and UGT1A1*28 genotype. First, it calculates the standard dose based on BSA, then queries the liver adjustment coefficient based on bilirubin level, and finally, based on the UGT1A1*28 genotype, it indicates a significantly increased risk of severe neutropenia and diarrhea, recommending a reduction in the initial dose to XX% of the standard dose. It also cites relevant clinical studies as evidence, and the resulting initial dose plan clearly lists the daily / weekly dosage, solvent requirements, infusion time, and pretreatment drugs.

[0022] The S3 follow-up plan and tools module, based on the treatment plan and tumor type generated in S2, calls upon a built-in follow-up template library. This library, developed by experts in various oncology subspecialties, defines key monitoring points for different treatment stages and medications. For example, for patients using a three-week chemotherapy regimen, this module automatically generates a follow-up plan for the first cycle (21 days). Days 1-7 include daily symptom questionnaires (focusing on diarrhea, nausea, and vomiting); day 8 prompts for a complete blood count (focusing on neutrophil count); and day 21, before the start of the next cycle, prompts for scheduled blood counts, liver and kidney function tests, and imaging assessments. Simultaneously, the module generates a patient-generated symptom log, designed as a highly user-friendly questionnaire. For instance, when assessing diarrhea, instead of requiring text input, it provides graphical options ranging from "slightly increased bowel movements, 1-2 times daily" to "watery stools, ≥10 times daily, requiring bed rest." Fatigue assessment uses a subscale ranging from energetic to extremely fatigued, unable to perform any activity. This significantly reduces the difficulty of reporting for patients.

[0023] S4, the follow-up data collection and integration module, gathers data from multiple channels when executing the follow-up plan. The data flow is mainly divided into three categories: For patient-reported data, patients receive and fill out the questionnaire generated in S3 via a mobile app. They can also check in to take their medication and record any discomfort on the electronic medication record interface.

[0024] For the data from community testing sites, patients go to the community health service center for follow-up blood tests and liver and kidney function tests as planned. Community doctors directly enter the test results into the system through the doctor's terminal or import them into the corresponding follow-up task through photo recognition.

[0025] Data collected from simple home devices, such as home blood pressure monitors, thermometers, smart scales, and portable coagulation analyzers, can be synchronized to the system via Bluetooth or manual input.

[0026] S5, the dose optimization analysis and adjustment suggestion generation module has a built-in dose adjustment algorithm model that is a hybrid model containing a large number of "IF-THEN" rules that directly encode clinical guidelines and expert experience.

[0027] This model is based on a Bayesian feedback model of pharmacokinetics / pharmacodynamics. Taking the management of oral targeted drugs as an example, the system uses the patient's weight, liver function, and concomitant medications as prior information, combined with the distribution of the drug's PK parameters in the population, to establish an initial PK model for the patient. When the patient regularly reports their blood pressure and liver function retest results, the system uses a Bayesian method to feed back these individual PD data to update the patient's PK parameter estimate, thereby more accurately predicting the drug's exposure in the body. If the predicted AUC is consistently higher than the upper limit of the target therapeutic window and is accompanied by grade 2 or higher hypertension, the system will determine that the adjustment conditions are met and generate a suggestion to "consider adjusting the dose from 5 mg once daily to 5 mg every other day" to reduce exposure and mitigate toxicity.

[0028] This module continuously scans the integrated follow-up data. Once one or more indicators trigger preset adjustment conditions, it automatically runs the analysis model and generates a report recommending dose optimization. The report details the triggering data, the basis for the analysis, the specific adjustment plan, and an analysis of the expected risks and benefits after the adjustment.

[0029] In step S6, the diagnostic and treatment support report and review module integrates the adjustment suggestions generated in step S5 with the patient's key data and relevant evidence-based data within the current cycle to generate a diagnostic and treatment support report. This report is pushed in real time to the community physician responsible for the patient and / or the oncology specialist at their superior hospital via the doctor's app or web backend. Doctors can directly review the report, "accept" suggestions with one click, make modifications, or reject suggestions with reasons. The doctor then makes the final confirmation.

[0030] S7, the personalized medication reminder and recording tool module, automatically generates medication reminders based on the initial plan in step S2 or the adjusted plan confirmed in S6. The reminders include not only the time but also a picture of the medication, the specific dosage, the method of administration, and prompts based on the patient's reported symptoms. The push notification channel can be set according to patient preferences: more app push notifications for younger patients; SMS or automated voice calls can be added for older patients.

[0031] The electronic medication record interface is designed to be simple, primarily featuring a calendar and pillbox view. Patients can complete the check-in by clicking the corresponding pillbox after each dose. If a check-in is missed, the system will prompt "Did you miss your 8 AM medication?" Patients can choose "Did you take it?", "Missed it", or "Postpone it". Any record of "missed dose" or "severe discomfort" will trigger a yellow or red alert in the background and notify the responsible nurse or doctor via message, enabling proactive intervention.

[0032] In step S8, after a treatment cycle ends, the feedback and iterative treatment module immediately imports the data recorded in step S7 and the new round of follow-up data collected in step S4 into the patient's file, becoming new input for the next round (S5) dose optimization analysis. For example, if the patient's diarrhea subsides in the first cycle but a rash appears in the second cycle, the system will comprehensively analyze the changes in the toxicity profile between the two cycles and may provide a new recommendation: "Diarrhea is under control, the current dose is tolerable, it is recommended to maintain the current dose and add topical medication to treat the rash." This cycle repeats continuously, ensuring the treatment plan always evolves according to the patient's individual response.

[0033] This invention provides an embodiment where, when the system creates a treatment plan for a patient that includes oxaliplatin, the science education module automatically pushes short videos and articles on "Precautions for Oxaliplatin" to the patient's device, focusing on the prevention of neurotoxicity by "avoiding contact with cold stimuli (cold drinks, cold air)." When winter arrives, the system automatically adds a message to the medication reminder: "The weather is getting colder; please keep warm and avoid contact with cold water to prevent oxaliplatin-related neurotoxicity." When a patient reports symptoms of numbness in their hands and feet, the system, while collecting this information, will display a small prompt: "The symptoms you reported may be related to the medication; this is a common side effect. Please see the nursing advice on 'How to relieve numbness in hands and feet'."

[0034] This invention provides a comparative example: This study designed a 6-month simulated retrospective cohort study. The experiment simulated the inclusion of 200 patients with solid tumors who received oral chemotherapy or targeted drug therapy in the community. They were randomly divided into an experimental group (n=100) and a control group (n=100). The experimental group was managed according to the present invention, while the control group was managed according to routine community follow-up management (regular outpatient visits, follow-up examinations according to a fixed schedule, paper or simple electronic reminders). The specific results are shown in the table below.

[0035]

[0036] As shown in the table above, the experimental group using the method of this invention performed significantly better than the control group in all aspects. First, by detecting signs of toxicity early and adjusting the dosage in a timely manner, the experimental group maintained a higher treatment intensity while reducing the incidence of serious adverse reactions by approximately 40-50%. This means that patients can more safely receive effective doses of treatment, which theoretically benefits long-term efficacy. Second, the reminders, convenient recording tools, and contextualized science popularization provided by this invention significantly improved patient compliance and knowledge levels in the experimental group. Third, the experimental group significantly reduced unplanned emergency room visits and frequent trips to large hospitals, allowing for more stable follow-up and monitoring work to be carried out in the community and at home, optimizing the allocation of medical resources, reducing the economic and time burden on patients, and alleviating the pressure on large hospitals. With the system's assistance, community doctors are more efficient in handling medication issues and make more informed decisions.

[0037] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for optimizing individualized drug dosage for cancer patients, characterized in that, Includes the following steps: S1, the system collects and enters the patient's initial clinical data, which includes at least the tumor type, pathological stage, gene testing results, liver and kidney function indicators, body surface area, concomitant medications and previous treatment history; S2, based on the patient's initial clinical data, calls the built-in evidence-based medicine knowledge base and pharmaceutical database, and combines the patient's body surface area and liver and kidney function indicators to calculate and generate an initial medication dosage plan, which includes the drug name, single dose, dosing frequency, route of administration and duration of the first treatment cycle; S3 automatically generates a structured long-term follow-up plan containing key time nodes based on the initial dosage regimen and treatment goals. The follow-up plan includes the next follow-up date, specific laboratory tests that need to be repeated, imaging assessment time points, and a list of the patient's self-reported symptoms. S4. At the time points set in the follow-up plan, follow-up data of patients are collected through online platforms, community medical points or simple home testing devices. The follow-up data includes patient-reported symptoms and signs, simple physiological indicators, laboratory retest results, treatment compliance records and adverse reaction information after medication. S5. The collected follow-up data, initial clinical data and initial medication dosage plan are integrated and analyzed. Using a preset dosage adjustment algorithm model, the current efficacy and toxicity are evaluated. If the preset adjustment conditions are met, dosage optimization adjustment suggestions are generated. The adjustment suggestions include dosage increase or decrease, change of dosing interval or adjunctive supportive treatment. S6, based on the initial medication dosage plan or the dosage optimization and adjustment suggestions in step S5, generates a visual medication reminder schedule with clear time stamps, and pushes it through the patient's terminal. At the same time, it provides an electronic medication record interface for patients or their families to record the actual medication time, dosage and subjective feelings at the time. S7. Use the data recorded in step S6 and the follow-up data collected in step S4 as input for the next round of analysis, and repeat steps S5 to S6 to achieve cyclical optimization and long-term management of the medication dosage plan.

2. The method for optimizing individualized drug dosage for cancer patients according to claim 1, characterized in that: In step S1, the initial clinical data is collected partly automatically through connection to the regional medical information platform, and partly by community medical staff or patients' families under the guidance of standardized electronic forms.

3. The method for optimizing individualized medication dosage for cancer patients according to claim 2, characterized in that: In step S2, the evidence-based medicine knowledge base and pharmaceutical database integrate pharmacokinetic data based on ethnic differences, evidence on the effects of specific gene polymorphisms on drug efficacy / toxicity, and dosage recommendations from authoritative clinical practice guidelines.

4. The method for optimizing individualized drug dosage for cancer patients according to claim 3, characterized in that: In step S4, the simple testing devices include home blood pressure monitors, thermometers, smart scales, and portable coagulation analyzers, whose measurement data can be synchronized to the system via Bluetooth or manual input.

5. The method for optimizing individualized drug dosage for cancer patients according to claim 4, characterized in that: In step S5, the dose adjustment algorithm model integrates a Bayesian feedback model based on pharmacokinetic and pharmacodynamic principles and a rule-based expert system. The preset adjustment conditions include, but are not limited to, the occurrence of toxic reactions of a specific level or above, failure of key efficacy markers to achieve the expected changes, or clinically significant fluctuations in liver and kidney function.

6. The method for optimizing individualized drug dosage for cancer patients according to claim 5, characterized in that: In step S6, the medication reminder schedule can be pushed via SMS, application push, or voice call, and the reminder content includes a picture of the medication, precautions for taking it, and tips for handling emergencies.

7. The method for optimizing individualized drug dosage for cancer patients according to claim 6, characterized in that: In step S6, when the electronic medication record interface records abnormal medication events or severe discomfort symptoms, it triggers an alert and pushes the information to the terminal of the designated responsible medical staff.

8. The method for optimizing individualized drug dosage for cancer patients according to claim 7, characterized in that: After step S5, the dosage optimization and adjustment recommendations, along with relevant evidence-based data and a summary of the patient's recent key data, are used to generate a diagnostic and treatment support brief for final review and confirmation by community doctors or doctors at higher-level hospitals.

9. The method for optimizing individualized drug dosage for cancer patients according to claim 8, characterized in that, The method also includes a primary care oncology science popularization and treatment assistance step: the system matches and pushes concise science popularization materials based on the patient's current treatment stage and common misconceptions, and embeds brief science popularization tips on the core precautions or possible side effects of the current drug each time a medication reminder or follow-up plan is generated.

10. The method for optimizing individualized drug dosage for cancer patients according to claim 9, characterized in that: In step S3, the patient's self-reported symptom list is converted into a questionnaire containing graphical symptom options to lower the reporting threshold for patients and improve the standardization of data.

Images