A Personalized Liver Cancer Screening and Monitoring Method and System

Abstract

This invention relates to the interdisciplinary field of medical informatics, health economics, and public health management, and discloses a personalized liver cancer screening and monitoring method and system. The method includes: acquiring real-world clinical data of patients with chronic hepatitis B; calculating the aMAP score for each patient using an aMAP risk prediction model based on the real-world clinical data, and classifying patients into different risk groups according to the aMAP score and a preset score threshold; constructing a personalized screening and monitoring decision tree model, and setting differentiated screening and monitoring strategies according to risk groups; constructing a decision tree Markov analysis model to conduct cost-effectiveness analysis on different screening and monitoring strategies, and obtaining the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results. This provides quantitative support for health decision-making, promotes the scientific and refined allocation of resources, helps improve screening compliance and early diagnosis rates among people with chronic liver disease, thereby improving the prognosis of liver cancer and reducing the burden on public health.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of medical informatics, health economics, and public health management, and specifically to a personalized liver cancer screening and monitoring method and system. Background Technology

[0002] Hepatocellular carcinoma (HCC) is the most common type of primary liver cancer, ranking sixth among the most common cancers globally and third among cancer-related deaths, making it a significant public health issue worldwide. In my country, HCC-related deaths account for nearly 50% of global cases, highlighting the severe burden of liver cancer. Despite significant progress in antiviral therapy and vaccination, unequal access to healthcare resources for HCC screening, early diagnosis, and treatment persists globally. Furthermore, the fact that most patients are diagnosed at an advanced stage contributes to the generally poor prognosis of HCC.

[0003] Numerous studies have suggested that liver cancer surveillance can improve early detection rates, radical treatment acceptance rates, and overall survival, making it an effective means of improving liver cancer prognosis. Most international and domestic guidelines recommend using abdominal ultrasound (US) combined with serum alpha-fetoprotein (AFP) testing every six months as a liver cancer screening and surveillance method. However, some studies have pointed out that the application of this single screening mode has many limitations. The interpretation of US results is easily affected by the operator's experience, technique, and attention to detail, as well as the thickness of the patient's subcutaneous fat, and the detection rate for early-stage hepatocellular carcinoma (HCC) is only 47%. AFP results are also affected by many factors, including pregnancy, active liver disease, germ cell tumors of the gonads, and gastrointestinal tumors. According to data, approximately 30% of liver cancer patients consistently show false-negative AFP results. In real-world clinical cohorts, the risk of HCC in patients with chronic liver disease is not uniform, and the current single traditional screening method may lead to poor prediction of individual HCC risk. Therefore, there is an urgent need to build an individualized HCC screening and monitoring system to provide an effective means for the widespread adoption of HCC screening and monitoring in the real world, and to help improve the accessibility and compliance of HCC screening and monitoring among the public.

[0004] In recent years, screening has sparked numerous controversies regarding overdiagnosis and treatment. Therefore, evaluating the effectiveness and cost-effectiveness of screening programs is crucial for weighing the pros and cons and providing objective scientific evidence for public health decision-making. While most published cost-effectiveness studies on HCC surveillance confirm the cost-effectiveness of the standard biannual US combined with AFP screening, few cost-effectiveness analyses exist for individualized HCC screening strategies guided by predictive models. Furthermore, the data included in these studies is mostly simulated, with limited data from real-world cohorts. Consequently, results based on simulated data may differ significantly from real-world findings, lacking individual heterogeneity and containing overly idealized assumptions that do not align with real-world scenarios, thus hindering objective evaluation. Summary of the Invention

[0005] In view of this, the present invention provides a personalized liver cancer screening and monitoring method and system to address the shortcomings of current HCC screening and monitoring strategies, such as insufficient consideration of individual risk heterogeneity, limited monitoring methods, and reliance on simulation data for cost-effectiveness evaluation. The present invention provides a simulation modeling and optimization strategy for personalized screening and monitoring of hepatocellular carcinoma.

[0006] In a first aspect, the present invention provides a personalized liver cancer screening and monitoring method, comprising:

[0007] Acquire real-world clinical data of chronic hepatitis B patients from a multicenter cross-sectional study within a preset historical timeframe;

[0008] Based on real-world clinical data, the aMAP risk prediction model is used to calculate the aMAP score for each patient, and patients are divided into different risk groups according to the aMAP score and preset score thresholds.

[0009] Construct an individualized screening and monitoring decision tree model and set differentiated screening and monitoring strategies according to risk groups;

[0010] A decision tree Markov analysis model was constructed to conduct cost-effectiveness analysis on different screening and monitoring strategies, and the optimal screening and monitoring strategy corresponding to different risk groups was obtained based on the analysis results.

[0011] The personalized liver cancer screening and monitoring method provided in this invention utilizes multi-center cross-sectional data to cover chronic hepatitis B patients from different regions, ethnicities, and disease stages, avoiding the bias of single-center data and making the aMAP score calculation more consistent with the characteristics of the real population. It precisely stratifies different patient groups according to aMAP score thresholds to clarify their risk levels, and constructs differentiated screening and monitoring strategies for different risk levels. By setting differentiated screening and monitoring strategies, it reduces over- and under-screening, achieving a match between resources and patient needs. Using Markov models to simulate disease progression paths under different screening frequencies, it calculates the 30-year healthcare costs and quality-adjusted life years for each risk group using different strategies. By quantifying the health outcomes and cost-benefit of different risk groups receiving personalized screening and monitoring, it can be used to evaluate the long-term cost-effectiveness of screening and monitoring strategies, providing quantitative support for health decision-making and promoting the scientific and refined allocation of resources.

[0012] In one optional implementation, the step of classifying patients into different risk groups based on aMAP scores and preset scoring thresholds includes: classifying patients with aMAP scores ≤ 50 into a low-risk group, patients with aMAP scores 50 < aMAP scores ≤ 60 into a medium-risk group, and patients with aMAP scores > 60 into a high-risk group.

[0013] The different screening and monitoring strategies based on risk groups include: for the low-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 12-18 months; for the medium-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 6-12 months; and for the high-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 3-6 months.

[0014] The embodiments of the present invention, through standardized grouping and differentiated screening and monitoring strategies based on aMAP scores, are more accurate than the traditional "one-size-fits-all" screening model, effectively improving the screening efficiency of high-risk groups, reducing overdiagnosis and resource waste in low-risk groups, and promoting the optimization of screening and monitoring strategies.

[0015] In one optional implementation, the construction of a decision tree Markov analysis model to perform cost-effectiveness analysis on different screening and monitoring strategies, and to obtain the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results, includes:

[0016] The disease metastasis chain in the model is: chronic hepatitis B → compensated cirrhosis → decompensated cirrhosis → hepatocellular carcinoma → death;

[0017] Based on a pre-set prospective follow-up cohort of chronic hepatitis B, the probability of one disease state moving to another within a pre-set time range in the disease metastasis chain is calculated.

[0018] Repeat the above process of calculating the migration probability to calculate the annual migration probability of each disease state in each risk group, and obtain the screening and monitoring results corresponding to each screening and monitoring strategy based on the annual migration probability.

[0019] Based on the health utility value corresponding to each disease state, the cost of screening and monitoring programs, and the annual treatment cost, the cost-effectiveness of each screening and monitoring strategy is calculated.

[0020] By comparing cost-effectiveness, the optimal screening and monitoring strategy for different risk groups can be determined.

[0021] This invention uses a decision tree Markov analysis model to dynamically simulate the disease metastasis process, quantify the probability of disease progression in different risk groups, and comprehensively evaluate the cost-effectiveness of various screening and monitoring strategies by considering health utility values, screening and treatment costs. It identifies the "best cost-effectiveness" screening program for different risk groups, providing evidence-based support for the allocation of medical resources. Simultaneously, it assists in clinical scientific decision-making, improves patient compliance, and has scalability to adapt to medical developments, promoting the economical and efficient management of chronic liver disease.

[0022] In one alternative implementation, the metastasis probability is determined by calculating the cumulative incidence rate tp of the target state in the disease metastasis chain over t years using the Kaplan-Meier method. t and 95% confidence interval;

[0023] Using the formula tp1 = 1 - (1 - tp) t ) 1 / t Convert to annual transition probability;

[0024] Calculate the standard deviation of the 1-year migration probability to obtain the range of values ​​for the migration probability of each risk group.

[0025] This invention uses the Kaplan-Meier method to process complex clinical data, combined with time standardization transformation and probability fluctuation range calculation, to provide accurate and robust disease progression parameters for Markov models, ensuring the scientific validity of cost-effectiveness analysis results, and thus supporting the evidence-based development and clinical promotion of risk stratification screening and monitoring strategies.

[0026] In one alternative implementation, the calculation of the cost-effectiveness of each screening and monitoring strategy includes:

[0027] The incremental cost is obtained by calculating the cost difference between each screening and monitoring strategy and the pre-set control strategy.

[0028] The difference between the quality-adjusted life years of each screening and monitoring strategy and the quality-adjusted life years of the pre-defined control strategy is calculated to obtain the incremental quality-adjusted life years.

[0029] The ratio of the incremental cost to the incremental quality-adjusted life years is calculated to obtain the incremental cost-effectiveness ratio.

[0030] This invention presents the costs and benefits of different screening and monitoring strategies in concrete numerical form by calculating incremental costs, incremental quality-adjusted life years, and incremental cost-effectiveness ratios. This provides decision-makers with an intuitive and quantifiable reference, enabling a clear comparison of the advantages and disadvantages of various screening and monitoring strategies, thus allowing for more scientific and rational decision-making and avoiding subjective judgment and blind selection. By clearly defining the cost-effectiveness of each screening and monitoring strategy, it helps to allocate limited resources (such as funds and manpower) to the strategy with the best cost-effectiveness, improving resource utilization efficiency, avoiding resource waste, achieving optimal resource allocation, and maximizing the social and economic benefits of resources.

[0031] In one optional implementation, obtaining the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results includes:

[0032] The optimal screening and monitoring strategy is defined as the one with the highest cost-effectiveness ratio among the different screening and monitoring strategies corresponding to each risk group, where the ratio is lower than the willingness-to-pay threshold.

[0033] This invention establishes objective and quantifiable strategy optimization criteria, balancing effectiveness and cost, and avoiding subjective decision-making bias; it precisely optimizes the allocation of medical resources, directing resources towards cost-effective strategies and reducing waste; it provides data support for health policy formulation and guides regional resource allocation; it promotes the standardization of clinical pathways, assists in joint decision-making between doctors and patients, and enhances patient compliance; and it can also adjust strategies based on actual economic conditions to ensure that the plan is both clinically effective and economically feasible.

[0034] Secondly, the present invention provides a personalized liver cancer screening and monitoring system, the system comprising:

[0035] The real-world data acquisition module is used to acquire real-world clinical data of chronic hepatitis B patients from a multicenter cross-sectional study within a preset historical time frame.

[0036] The risk group classification module is used to calculate the aMAP score for each patient based on real-world clinical data using the aMAP risk prediction model, and to classify patients into different risk groups according to the aMAP score and preset score thresholds.

[0037] The screening and monitoring strategy personalization module allows users to build individualized screening and monitoring decision tree models and set differentiated screening and monitoring strategies according to risk groups.

[0038] The screening and monitoring strategy optimization module is used to construct a decision tree Markov analysis model, perform cost-effectiveness analysis on different screening and monitoring strategies to obtain analysis results, and obtain the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results.

[0039] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the personalized liver cancer screening and monitoring method of the first aspect or any corresponding embodiment described above.

[0040] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the personalized liver cancer screening and monitoring method of the first aspect or any corresponding embodiment described above.

[0041] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the personalized liver cancer screening and monitoring method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0042] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the personalized liver cancer screening and monitoring method according to an embodiment of the present invention;

[0044] Figure 2 A comparative chart showing the cost-effectiveness analysis results of eight individualized HCC screening and monitoring strategies versus the standard strategy;

[0045] Figure 3 This is a structural block diagram of a personalized liver cancer screening and monitoring system according to an embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0047] 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 only some embodiments of the present invention, 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.

[0048] Existing HCC monitoring strategies do not adequately consider individual risk heterogeneity, have limited monitoring methods, and rely on simulation data for cost-effectiveness evaluation. To address these issues, this embodiment provides a personalized liver cancer screening and monitoring method, employing a simulation modeling and optimization strategy for personalized hepatocellular carcinoma screening and monitoring. Figure 1 As shown, the process includes the following steps:

[0049] Step S101: Obtain real-world clinical data of chronic hepatitis B patients from a multicenter cross-sectional study within a preset historical time frame.

[0050] Specifically, the real-world data obtained in this embodiment of the invention, which is a multi-center cross-sectional study of chronic hepatitis B patients, consists of data from 44,852 patients with chronic hepatitis B collected from 10 medical institutions in China. This avoids the bias of single-center data and incorporates multiple dimensions such as age, gender, total bilirubin, albumin, and platelet count, dynamically reflecting the patient's liver reserve function and pathological status. Compared with traditional single indicators, it provides a more comprehensive prediction of the risk of liver cancer. All data processing procedures comply with the requirements of data de-identification and privacy protection.

[0051] Step S102: Calculate the aMAP score for each patient using the aMAP risk prediction model based on real-world clinical data, and classify patients into different risk groups according to the aMAP score and preset score thresholds.

[0052] Specifically, the aMAP risk prediction model was created by Professor Hou Jinlin and Professor Fan Rong's team at Nanfang Hospital, Southern Medical University. It is used to predict the long-term risk of liver cancer in patients with chronic liver disease. Based on a large global cohort, this model is the world's first cross-disease, cross-ethnic prediction model for liver cancer in patients with chronic liver disease.

[0053] aMAP score = {(0.06 × age + 0.89 × sex + 0.48 × (0.66 × log10 total bilirubin - 0.085 × albumin) - 0.01 × platelets) + 7.4} / 14.77 × 100. Where age is in years, total bilirubin is in μmol / L, albumin is in g / L, and platelets are in 10³ / mm³. For sex, the value is 1 for males and 0 for females.

[0054] This invention utilizes multicenter cross-sectional data to cover chronic hepatitis B patients from different regions, ethnicities, and disease stages, making the aMAP score calculation more consistent with the characteristics of the real population (such as age distribution, differences in liver function indicators, etc.).

[0055] Furthermore, the preset scoring thresholds in this embodiment of the invention are determined based on large-scale cohort studies. The risk group classification thresholds can be adjusted according to the incidence of liver cancer and the level of medical resources in different regions. In one embodiment, 50 and 60 points are used as scoring thresholds to classify patients into a low-risk group (0-50 points), a medium-risk group (50-60 points), and a high-risk group (60-100 points) for liver cancer. This provides an intuitive risk assessment scale for clinicians, making it easier for doctors to quickly locate the patient's risk level, avoid subjective judgment bias, and improve the efficiency of diagnosis and treatment.

[0056] Step S103: Construct an individualized screening and monitoring decision tree model and set differentiated screening and monitoring strategies according to risk groups.

[0057] Specifically, the personalized screening and monitoring decision tree model is a core tool for developing precise screening plans based on patient risk characteristics. By integrating multi-dimensional clinical data and risk prediction model outputs, this model can tailor screening frequencies, technology combinations, and intervention thresholds for patients in different risk groups, achieving resource optimization and maximizing effectiveness. In this embodiment of the invention, after inputting the aMAP score into the decision tree model's data input layer, different risk groups are divided according to the score thresholds, and different screening and monitoring strategies are set for each risk group, including:

[0058] 1. For the low-risk group, abdominal ultrasound and serum alpha-fetoprotein testing, combined with abdominal CT scans, are conducted every 12-18 months. This not only meets the monitoring needs of the low-risk population but also reduces the economic burden (reducing examination costs) and psychological stress (avoiding excessive medical anxiety) for patients.

[0059] 2. The intermediate-risk group is required to undergo abdominal ultrasound and serum alpha-fetoprotein testing every 6-12 months, combined with abdominal CT scans; this aims to strike a balance between early detection of liver cancer and control of screening costs for the intermediate-risk population.

[0060] 3. The high-risk group will undergo abdominal ultrasound and serum alpha-fetoprotein (AFP) testing every 3-6 months, combined with abdominal CT scans. This high-frequency testing aims to increase the early detection rate of liver cancer (early treatment survival rate exceeds 70%), reducing the high costs and worsening prognosis associated with late-stage treatment, given the high risk of developing the disease in this population.

[0061] It should be noted that abdominal ultrasound and serum alpha-fetoprotein (AFP) testing, combined with abdominal CT scan, are performed when at least one of the following conditions is initially detected: abnormality on abdominal ultrasound (e.g., large nodules) or positive serum AFP. The abdominal CT scan is then used for further confirmation.

[0062] The current guidelines recommend a standard screening and monitoring strategy (abdominal ultrasound combined with serum alpha-fetoprotein screening every 6 months for non-HCC risk stratification). Compared with the traditional "one-size-fits-all" screening model, the embodiments of this invention are more accurate, effectively improve the screening efficiency of high-risk groups, reduce overdiagnosis and resource waste in low-risk groups, concentrate resources on high-benefit groups, reduce social medical costs, promote the optimization of screening and monitoring strategies, establish a replicable stratified management model, and provide a practical template for the standardized diagnosis and treatment of chronic liver diseases.

[0063] In one embodiment, the present invention provides low, medium and high-risk groups based on aMAP scores to undergo US combined with AFP screening every 12 or 18 months, every 6 or 12 months, and every 3 or 6 months, respectively, forming a total of 8 individualized HCC screening and monitoring strategies. These strategies are compared with the standard screening and monitoring strategies recommended by current guidelines, as shown in Table 1.

[0064] Table 1

[0065]

[0066] Step S104: Construct a decision tree Markov analysis model, perform cost-effectiveness analysis on different screening and monitoring strategies to obtain analysis results, and obtain the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results.

[0067] This invention employs TreeAge Pro 2022, R1.2 software to construct a decision tree Markov model. The Markov model's disease progression chain is chronic hepatitis B → compensated cirrhosis → decompensated cirrhosis → HCC → death. Simulation results (HCC incidence, HCC and chronic hepatitis B-related mortality) and cost-effectiveness (cost, incremental cost, quality-adjusted life years, incremental quality-adjusted life years, and incremental cost-effectiveness ratio) of 30 years of HCC screening and monitoring in the chronic hepatitis B population are calculated. The study is conducted from the perspective of healthcare providers, using a 5% discount rate, and setting the Willingness to Pay (WTP) threshold to the 2023 per capita GDP of China (US$12,681).

[0068] Specifically, this embodiment of the invention uses a prospective chronic hepatitis B follow-up cohort (Search-B, NCT02167503) to calculate the probability of disease metastasis, that is, the probability of transitioning from one disease state to another within a certain time range. In the Search-B cohort, the initial number of people in each disease state transition path, the number of people experiencing state transition, and the follow-up duration are calculated within the total population: Chronic Hepatitis B → Compensated Cirrhosis, Chronic Hepatitis B → HCC, Compensated Cirrhosis → Decompensated Cirrhosis, Compensated Cirrhosis → HCC, Compensated Cirrhosis → Death, Decompensated Cirrhosis → HCC, Decompensated Cirrhosis → Death, HCC → Death (the left side of the arrow represents the initial state, and the right side represents the target state).

[0069] Furthermore, this embodiment of the invention employs the Kaplan-Meier method to calculate the probability of disease state transition. It truncates patient data that has not progressed to the next disease state to ensure the accuracy of the transition probability estimate. Specifically, the cumulative incidence rate (tpt) of the target state over t years and its 95% confidence interval (lower bound of tpt - upper bound of tpt) for each of the above paths are calculated using Kaplan-Meier curves. The value of t ranges from 2 to 8 years. The formula tp1 = 1 - (1 - tp) t ) 1 / t The parameter representing the duration of t years is converted into the migration probability (tp1) corresponding to 1 year and its 95% confidence interval (lower bound of tp1 - upper bound of tp1). Then, the standard deviation σ corresponding to the 1-year migration probability is calculated, σ = (upper bound of tp1 - lower bound of tp1) / 3.92, yielding the range of migration probability values ​​for each risk group. The above calculation process is repeated to calculate the annual migration probability of each disease state in the low-risk, medium-risk, and high-risk aMAP groups, as shown in Table 2.

[0070] Table 2

[0071]

[0072] This invention uses the Kaplan-Meier method to process complex clinical data, combining time standardization transformation and probability fluctuation range calculation to provide accurate and robust disease progression parameters for Markov models, ensuring the scientific validity of cost-effectiveness analysis results, and thus supporting the evidence-based formulation and clinical promotion of risk stratification screening and monitoring strategies.

[0073] By setting parameters in the decision tree Markov analysis model including: the annual disease metastasis probability of screening-positive chronic hepatitis B (CHB) individuals (from publicly available literature), the annual disease metastasis probability of screening-negative CHB individuals (from publicly available literature), the annual disease metastasis probability in each risk group of aMAP score, the health utility of each disease state (chronic hepatitis B, compensated cirrhosis, decompensated cirrhosis, hepatocellular carcinoma), the cost of HCC screening and monitoring programs (including alpha-fetoprotein, aMAP score, abdominal ultrasound / CT, from multicenter real-world data), and the annual treatment cost of each disease state (chronic hepatitis B, compensated cirrhosis, decompensated cirrhosis, hepatocellular carcinoma) (from publicly available literature), and running the model, the impact of screening and monitoring strategies on the population was obtained: after simulating 30 years of HCC screening and monitoring based on real-world data, the standard strategy will lead to 12,875 HCC cases, 11,961 HCC-related deaths, and 27,480 chronic hepatitis B-related deaths. Table 3 details the impact of eight individualized screening and monitoring strategies on HCC-related and CHB-related mortality rates.

[0074] Table 3

[0075]

[0076] After introducing eight personalized screening and surveillance strategies (AH), the incidence and mortality rates of HCC both decreased significantly. Specific results are as follows:

[0077] 1. HCC incidence: Strategies A through H reduced HCC cases by 5.25% to 6.07%.

[0078] 2. HCC-related mortality: All strategies resulted in a decrease, ranging from 15.07% to 15.60%.

[0079] 3. Mortality related to chronic hepatitis B: All strategies effectively reduced the risk of death from chronic hepatitis B by 30.82% to 31.93%, highlighting the potential of these strategies to improve survival outcomes in patients with HCC and chronic hepatitis B.

[0080] The cost-effectiveness analysis of different screening and monitoring strategies was conducted using a decision tree Markov analysis model. The analysis included: cost, incremental cost, quality-adjusted life years, incremental quality-adjusted life years, and incremental cost-effectiveness ratio, as shown in Table 4.

[0081] Table 4

[0082]

[0083] Specifically, in this embodiment of the invention, the cost difference between each screening and monitoring strategy and the cost difference between the preset control strategy (current screening and monitoring strategy) is calculated to obtain the incremental cost; and the quality-adjusted life years (MAUs) of each screening and monitoring strategy and the MAUs of the preset control strategy are calculated to obtain the incremental MAUs; finally, the ratio of the incremental cost to the incremental MAUs is calculated to obtain the incremental cost-effectiveness ratio (ICER).

[0084] ICER is the ratio of the cost difference to the effect difference between an intervention strategy and a control strategy. It represents the cost required to gain one more QALY for an individualized HCC screening and monitoring strategy (Strategy AH) guided by the aMAP score compared to the baseline screening strategy (Base-case). The specific calculation formula is as follows:

[0085]

[0086] A positive ICER value indicates the incremental cost required to increase quality-adjusted life years or reduce blind years; a negative ICER value indicates that the optimal screening and surveillance strategy has a lower cost while increasing quality-adjusted life years or reducing blind years compared to other strategies.

[0087] This invention, through calculating incremental cost, incremental quality-adjusted life years, and incremental cost-effectiveness ratio, presents the costs and benefits of different screening and monitoring strategies in concrete numerical form. This provides decision-makers with an intuitive and quantifiable reference, enabling a clear comparison of the advantages and disadvantages of various screening and monitoring strategies, thus allowing for more scientific and rational decision-making and avoiding subjective judgment and blind selection. Clearly defining the cost-effectiveness of each screening and monitoring strategy helps allocate resources to the strategy with the best cost-effectiveness, improving resource utilization efficiency, avoiding resource waste, achieving optimal resource allocation, and maximizing the social and economic benefits of resources. Figure 2 As shown, compared to the control strategy, the individualized HCC screening and monitoring strategy increases the cost of HCC screening and monitoring by $449-878 million, generating an additional 48,603.89-103,961.00 QALYs. It successfully optimized the incremental cost-effectiveness ratio (ICER) of HCC screening and monitoring to $8,444.92-9,246.61 / QALY, which is significantly lower than the WTP threshold ($12,681), demonstrating cost-effectiveness.

[0088] Finally, the strategy with the best cost-effectiveness ratio that is lower than the willingness-to-pay threshold among the different screening and monitoring strategies corresponding to each risk group is selected as the optimal screening and monitoring strategy. According to the data in Table 4 of this embodiment, strategy H is the optimal screening and monitoring strategy, ensuring that the screening and monitoring strategy has both clinical effectiveness and economic feasibility.

[0089] The personalized liver cancer screening and monitoring method provided in this invention is based on real-world data modeling, which enhances the applicability and feasibility of the model in clinical settings. It effectively overcomes the problems of idealized assumptions and lack of individual heterogeneity in simulated data, making it closer to real clinical situations and enhancing the reliability and extrapolation of model predictions. Based on the aMAP score, risk stratification of the chronic hepatitis B population is performed, enabling individualized adjustments to HCC screening strategies. This effectively improves screening efficiency for high-risk groups and reduces overdiagnosis and resource waste in low-risk groups. By quantifying the health outcomes and cost-benefit of different risk groups under personalized screening and monitoring, it can be used to evaluate the long-term cost-effectiveness of screening and monitoring strategies, providing quantitative support for health decision-making, promoting the scientific and refined development of resource allocation, helping to improve screening compliance and early diagnosis rates in the chronic liver disease population, thereby improving liver cancer prognosis, reducing the burden on public health, and promoting the development of chronic liver disease management towards a more cost-effective direction.

[0090] This embodiment also provides a personalized liver cancer screening and monitoring system, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. As used below, the term "module" can be a combination of software and / or hardware that performs a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0091] This embodiment provides a personalized liver cancer screening and monitoring system, such as Figure 3 As shown, it includes:

[0092] The real-world data acquisition module 301 is used to acquire real-world clinical data of chronic hepatitis B patients from a multi-center cross-sectional study within a preset historical time frame.

[0093] The risk group classification module 302 is used to calculate the aMAP score for each patient based on real-world clinical data using the aMAP risk prediction model, and to classify patients into different risk groups according to the aMAP score and preset score thresholds.

[0094] The personalized screening and monitoring strategy setting module 303 allows users to build an individualized screening and monitoring decision tree model and set differentiated screening and monitoring strategies according to risk groups.

[0095] The screening and monitoring strategy optimization module 304 is used to construct a decision tree Markov analysis model, perform cost-effectiveness analysis on different screening and monitoring strategies to obtain analysis results, and obtain the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results.

[0096] In some optional implementations, patients are divided into different risk groups based on their aMAP scores and preset scoring thresholds, including: patients with aMAP scores ≤ 50 are classified as low-risk, patients with aMAP scores ≤ 60 are classified as medium-risk, and patients with aMAP scores > 60 are classified as high-risk.

[0097] The different screening and monitoring strategies based on risk groups include: for the low-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 12-18 months; for the medium-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 6-12 months; and for the high-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 3-6 months.

[0098] In some optional implementations, the screening and monitoring strategy optimization module 304 includes:

[0099] The disease metastasis chain setting unit is used to construct the model as follows: chronic hepatitis B → compensated cirrhosis → decompensated cirrhosis → hepatocellular carcinoma → death;

[0100] The state transition probability calculation unit is used to calculate the transition probability of one disease state to another within a preset time range in the disease metastasis chain, based on preset prospective chronic hepatitis B follow-up cohort data.

[0101] The screening and monitoring result acquisition unit is used to repeat the above-mentioned process of calculating the transfer probability, calculate the annual transfer probability of each disease state in each risk group, and obtain the screening and monitoring results corresponding to each screening and monitoring strategy based on the annual transfer probability.

[0102] The cost-effectiveness analysis unit is used to calculate the cost-effectiveness of each screening and monitoring strategy based on the health utility value corresponding to each disease state, the cost of screening and monitoring programs, and the annual treatment cost.

[0103] The optimal screening and monitoring strategy output unit is used to obtain the optimal screening and monitoring strategy for different risk groups by comparing cost-effectiveness.

[0104] In some alternative implementations, the metastasis probability is determined by calculating the cumulative incidence rate tp of the target state in the disease metastasis chain over t years using the Kaplan-Meier method. t and 95% confidence interval;

[0105] Using the formula tp1 = 1 - (1 - tp) t ) 1 / t Convert to annual transition probability;

[0106] Calculate the standard deviation of the 1-year migration probability to obtain the range of values ​​for the migration probability of each risk group.

[0107] In some alternative implementations, calculating the cost-effectiveness of each screening and monitoring strategy includes:

[0108] The incremental cost is obtained by calculating the cost difference between each screening and monitoring strategy and the pre-set control strategy.

[0109] The difference between the quality-adjusted life years of each screening and monitoring strategy and the quality-adjusted life years of the pre-defined control strategy is calculated to obtain the incremental quality-adjusted life years.

[0110] The ratio of the incremental cost to the incremental quality-adjusted life years is calculated to obtain the incremental cost-effectiveness ratio.

[0111] In some optional implementations, the optimal screening and monitoring strategy output unit specifically includes: taking the strategy with the most cost-effectiveness and an incremental cost-effectiveness ratio lower than the willingness-to-pay threshold among the different screening and monitoring strategies corresponding to each risk group as the optimal screening and monitoring strategy.

[0112] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0113] The personalized liver cancer screening and monitoring system in this embodiment is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0114] This invention also provides a computer device having the above-described features. Figure 3 The individualized liver cancer screening and monitoring system shown.

[0115] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 4As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4 Take a processor 10 as an example.

[0116] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GPA), or any combination thereof.

[0117] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0118] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0119] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0120] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0121] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0122] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0123] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A personalized liver cancer screening and monitoring method, characterized in that, include: Acquire real-world clinical data of chronic hepatitis B patients from a multicenter cross-sectional study within a preset historical timeframe; Based on real-world clinical data, the aMAP risk prediction model is used to calculate the aMAP score for each patient, and patients are divided into different risk groups according to the aMAP score and preset score thresholds. Construct an individualized screening and monitoring decision tree model and set differentiated screening and monitoring strategies according to risk groups; A decision tree Markov analysis model was constructed, in which parameters included: the annual disease metastasis probability of screening-positive chronic hepatitis B individuals, the annual disease metastasis probability of screening-negative chronic hepatitis B individuals, the annual disease metastasis probability in each risk group of the aMAP score, the health utility of each disease state, the cost of HCC screening and monitoring programs, and the annual treatment cost of each disease state. Cost-effectiveness analysis was performed on different screening and monitoring strategies to obtain the analysis results, and based on the analysis results, the optimal screening and monitoring strategy corresponding to different risk groups was determined, including: The disease metastasis chain in the model is: chronic hepatitis B → compensated cirrhosis → decompensated cirrhosis → hepatocellular carcinoma → death; Based on a pre-set prospective follow-up cohort of chronic hepatitis B, the probability of one disease state moving to another within a pre-set time range in the disease metastasis chain is calculated. Repeat the above process of calculating the migration probability to calculate the annual migration probability of each disease state in each risk group, and obtain the screening and monitoring results corresponding to each screening and monitoring strategy based on the annual migration probability. Based on the health utility value corresponding to each disease state, the cost of screening and monitoring programs, and the annual treatment cost, the cost-effectiveness of each screening and monitoring strategy is calculated, including: calculating the cost difference between each screening and monitoring strategy and the cost of a preset control strategy to obtain the incremental cost; calculating the difference between the quality-adjusted life years of each screening and monitoring strategy and the quality-adjusted life years of the preset control strategy to obtain the incremental quality-adjusted life years; and calculating the ratio of the incremental cost to the incremental quality-adjusted life years to obtain the incremental cost-effectiveness ratio. By comparing cost-effectiveness, the optimal screening and monitoring strategy for different risk groups can be determined.

2. The method according to claim 1, characterized in that, The patients are divided into different risk groups based on their aMAP scores and preset scoring thresholds, including: patients with aMAP scores ≤ 50 are classified as low-risk, patients with aMAP scores ≤ 60 are classified as medium-risk, and patients with aMAP scores > 60 are classified as high-risk. The different screening and monitoring strategies based on risk groups include: for the low-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 12-18 months; for the medium-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 6-12 months; and for the high-risk group, abdominal ultrasound and serum alpha-fetoprotein testing combined with abdominal CT scans every 3-6 months.

3. The method according to claim 1, characterized in that, The transition probability is determined in the following way: The Kaplan-Meier method was used to calculate the cumulative incidence rate (tp) of the target state in the disease metastasis chain over t years. t and 95% confidence interval; Using the formula tp1 = 1 - (1 - tp) t ) 1 / t Convert to annual transition probability; Calculate the standard deviation of the 1-year migration probability to obtain the range of values ​​for the migration probability of each risk group.

4. The method according to claim 1, characterized in that, The method of obtaining the optimal screening and monitoring strategy for different risk groups by comparing cost-effectiveness includes: The optimal screening and monitoring strategy is defined as the one with the highest cost-effectiveness ratio among the different screening and monitoring strategies corresponding to each risk group, where the ratio is lower than the willingness-to-pay threshold.

5. A personalized liver cancer screening and monitoring system, based on the method of claim 1, characterized in that, include: The real-world data acquisition module is used to acquire real-world clinical data of chronic hepatitis B patients from a multicenter cross-sectional study within a preset historical time frame. The risk group classification module is used to calculate the aMAP score for each patient based on real-world clinical data using the aMAP risk prediction model, and to classify patients into different risk groups according to the aMAP score and preset score thresholds. The screening and monitoring strategy personalization module allows users to build individualized screening and monitoring decision tree models and set differentiated screening and monitoring strategies according to risk groups. The screening and monitoring strategy optimization module is used to construct a decision tree Markov analysis model, perform cost-effectiveness analysis on different screening and monitoring strategies to obtain analysis results, and obtain the optimal screening and monitoring strategy corresponding to different risk groups based on the analysis results.

6. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the personalized liver cancer screening and monitoring method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the individualized liver cancer screening and monitoring method according to any one of claims 1 to 4.

8. A computer program product, characterized in that, Includes computer instructions for causing a computer to perform the individualized liver cancer screening and monitoring method according to any one of claims 1 to 4.

Images