Phenylketonuria regional intelligent screening system and storage medium

Abstract

The application discloses a phenylketonuria regional intelligent screening system and a storage medium, and belongs to the technical field of medical software development. The system comprises a data acquisition module, which is used for acquiring biological sample information, gene variation data and blood phenylalanine concentration of a subject; an intelligent analysis module, which is internally provided with a regional mutation characteristic database and is configured to calculate an individualized comprehensive risk index according to the gene variation data, the blood phenylalanine concentration and population grouping information to which the individual belongs; and a screening management module, which is used for performing hierarchical decision-making according to the comprehensive risk index and managing sample tracking, positive recall and follow-up tasks associated with the screening process. The application deeply integrates regional genetic characteristics and multi-source screening data through a software algorithm, realizes the precision and process optimization of phenylketonuria screening, and significantly improves the screening efficiency and prevention and control effect in high-incidence areas.

Description

Technical Field

[0001] This invention relates to the field of medical software, specifically to a regionalized intelligent screening system and storage medium for phenylketonuria. Background Technology

[0002] Newborn disease screening is a key public health program for preventing birth defects. Phenylketonuria (PKU) is a common autosomal recessive genetic disorder, and early screening and intervention can significantly improve prognosis.

[0003] Current PKU screening commonly uses tandem mass spectrometry to detect phenylalanine (Phe) in the blood for initial screening, and gene sequencing is performed to confirm the diagnosis for those who test positive. However, this method has the following drawbacks:

[0004] (1) The screening algorithm uses a nationally unified fixed threshold and does not integrate regional mutation spectrum features, which is not accurate enough for high-incidence areas;

[0005] (2) The screening process is fragmented, and sample tracking, positive recall and follow-up rely on manual labor, which is inefficient and prone to loss;

[0006] (3) The lack of regional data closed loop and dynamic iteration mechanism makes it difficult to support carrier screening and primary prevention.

[0007] Therefore, there is an urgent need for a regionalized, intelligent, and fully closed-loop phenylketonuria screening system. Summary of the Invention

[0008] The present invention aims to overcome the above-mentioned defects of the prior art and provide a regional intelligent screening system and storage medium for phenylketonuria. It achieves accurate risk assessment through regional mutation database and multimodal data fusion, and automates the entire screening process management. It is suitable for newborn screening and carrier screening of people of childbearing age.

[0009] 1. To achieve the above objectives, the present invention provides the following technical solution: a regionalized intelligent screening system for phenylketonuria, characterized in that: the system includes...

[0010] The data acquisition module is configured to acquire the subject's biometric identification information, gene variation data, blood phenylalanine concentration data, and the individual's population group information.

[0011] An intelligent analysis module is communicatively connected to the data acquisition module, and the intelligent analysis module includes:

[0012] A regional mutation signature database is used to store allele frequency data of specific mutations in the phenylalanine hydroxylase gene in different population groups within the target screening area;

[0013] The risk assessment unit is configured to: call the regional mutation feature database, and calculate the comprehensive risk index of phenylketonuria for the subject based on the pathogenicity weight of the mutation site in the gene mutation data, the allele frequency of the mutation site in the corresponding population group of the subject, and the deviation of the blood phenylalanine concentration data from the reference range.

[0014] The screening management module, which is communicatively connected to the intelligent analysis module, includes:

[0015] The decision-making unit is configured to compare the comprehensive risk index with a preset dynamic threshold and generate a screening conclusion that includes the risk level and preliminary recommendations.

[0016] The process management unit is configured to trigger and manage a series of tasks, including positive recall, genetic counseling appointments, and follow-ups, based on the screening results and the biosample identification information.

[0017] Preferably, the formula for calculating the comprehensive risk index by the risk assessment unit is:

[0018]

[0019] Wherein, CRI is the comprehensive risk index, W_i is the pathogenicity weight of the i-th detected gene variant, F_i is the allele frequency of the i-th gene variant in the subject population corresponding to the regional mutation frequency database, Phe is the measured serum phenylalanine concentration, M is the reference median value of serum phenylalanine concentration in the corresponding population, and α is the metabolic concentration influence coefficient.

[0020] Preferably, after the risk assessment unit receives a data packet from a subject, the algorithm executes the following steps:

[0021] a. Genetic risk score calculation: For each variant in the genetic variant data, query its corresponding allele frequency (F) in the local area and among the subjects. Combine this with the known pathogenicity weight of the variant (W, which can be preset according to the ACMG guidelines) to calculate the risk score (WF) contributed by the variant. Summate the contribution values ​​of all variants to obtain the total genetic risk score.

[0022] b. Calculation of metabolic risk score: The metabolic risk score is calculated based on the degree of deviation of the subject's blood Phe concentration (Phe) from the reference median (M) of the same age group. The formula is α|Phe-M| / M, where α is an adjustable coefficient.

[0023] c. Synthesis of the Comprehensive Risk Index (CRI): The genetic risk score and the metabolic risk score are added together to obtain the final CRI. This index is a continuous variable; the higher the value, the greater the risk of disease.

[0024] Preferably, the decision unit compares the calculated CRI value with a plurality of preset dynamic thresholds.

[0025] Preferably, the population grouping in the regional mutation feature database is based on the long-term accumulated epidemiological and genetic data of the target screening area, and is divided according to the gene mutation spectrum characteristics.

[0026] Preferably, the process management unit is further configured to build and maintain a regional phenylketonuria information database, which is used to structurally store the genotype, phenotype and treatment follow-up data of confirmed patients, and the regional mutation feature database in the intelligent analysis module is periodically updated based on the regional phenylketonuria information database.

[0027] Preferably, the process management unit further includes a visualization cockpit subunit, configured to present the administrator with a heat map of the real-time screening coverage, initial screening positivity rate, diagnosis rate, recall rate, and geographical distribution of patients in the target screening area.

[0028] Preferably, this system adopts a distributed architecture, including a cloud server deployed at a regional screening center and multiple clients deployed at various grassroots blood collection points; the cloud server runs the core services of the intelligent analysis module and the screening management module, and the clients are used to run some functions of the data acquisition module and to synchronize data and interact with the cloud server.

[0029] The present invention also provides a computer-readable storage medium that, when executed by a processor, implements the functions of the regional intelligent screening system for phenylketonuria as described in any one of claims 1 to 8.

[0030] This invention provides a regionalized intelligent screening system and storage medium for phenylketonuria. It has the following beneficial effects:

[0031] 1. More precise risk assessment: By incorporating regional mutation frequency characteristics into the risk assessment model through software algorithms, the screening conclusion is upgraded from "whether it is positive" to "risk quantification and classification", which is more in line with the concept of precision medicine and is particularly suitable for regions with unique genetic characteristics.

[0032] 2. Automated and collaborative process management: The system transforms discrete workflows into online automated task flows, reducing manual intervention and errors, and ensuring the quality and efficiency of closed-loop screening management.

[0033] 3. Data-driven self-evolution: The system accumulates high-quality clinical data through the "Regional PKU Information Database" and uses this data to update the "Regional Mutation Characteristic Database", enabling the system to continuously optimize and provide data support for public health decision-making.

[0034] 4. Flexible expansion of application scenarios: The same system can seamlessly support two core application scenarios, "newborn disease screening" and "carrier screening of reproductive age", by configuring different thresholds and rules, which effectively promotes the forward shift of PKU prevention and control. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the present invention; Detailed Implementation

[0036] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0037] Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0038] Please see Figure 1 This invention provides a technical solution: a regional intelligent screening system for phenylketonuria (PKU). The system adopts a distributed architecture with cloud servers and multiple terminals, including a data acquisition module, an intelligent analysis module, and a screening management module. It is also equipped with a regional mutation characteristic database and a regional PKU information database, enabling intelligent management of the entire process of newborn screening and carrier screening among the reproductive-age population. The modules communicate via encrypted network to achieve data synchronization, command interaction, and access control.

[0039] The data acquisition module serves as the system entry point, completing the standardized collection and binding of multi-source data. It is used to obtain the subject's biometric identification information, gene mutation data, blood phenylalanine concentration data, and the individual's population group information. The gene mutation data acquired by the data acquisition module comes from targeted sequencing results of high-frequency mutations in exon 7 of the phenylalanine hydroxylase gene, with high-frequency mutations including at least R243Q and R241C.

[0040] The intelligent analysis module is the core of the system, including a regional mutation characteristic database and a risk assessment unit.

[0041] The formula for calculating the comprehensive risk index by the risk assessment unit is as follows:

[0042]

[0043] Wherein CRI is the comprehensive risk index, W_i is the pathogenicity weight of the i-th detected gene variant, F_i is the allele frequency of the i-th gene variant in the subject population corresponding to the regional mutation frequency database, Phe is the measured serum phenylalanine concentration, M is the reference median value of serum phenylalanine concentration in the corresponding population, and α is the metabolic concentration influence coefficient.

[0044] The method for constructing a regional mutation feature database is as follows:

[0045] The regional mutation feature database in this embodiment is systematically constructed based on long-term accumulated phenylketonuria (PKU) population screening and genetic diagnosis data within the target screening area (Ningxia Hui Autonomous Region) through the following steps:

[0046] Step 1: Retrospective case data collection and organization

[0047] Clinical and genetic data of PKU patients diagnosed by neonatal disease screening centers within the target region were collected. Taking Ningxia as an example, data from 327 PKU patients diagnosed between 2015 and 2025 through tandem mass spectrometry screening, urinary pterin profile analysis, tetrahydrobiopterin loading test, and gene sequencing were collected.

[0048] The collected data fields include: the child's unique identifier, place of birth, population grouping information (natural classification based on local genetic epidemiological surveys), initial blood phenylalanine concentration, age at diagnosis, and PAH gene sequencing results (including mutation site, mutation type, and zygotic status).

[0049] Step 2: High-throughput sequencing and mutation identification

[0050] The collected case samples underwent sequencing of the entire exon and exon-intron junction regions of the PAH gene. Direct sequencing of PCR products or capture sequencing was used to cover all 13 exons and flanking intron regions of the PAH gene.

[0051] Bioinformatics software (such as BWA and GATK) was used for sequence alignment and variant calling, and Sanger sequencing was used to verify suspected mutations.

[0052] The detected mutations were compared with the PAH international database (PAHdb) and HGMD professional version to determine their known pathogenicity or novel mutation nature.

[0053] Step 3: Population group definition and frequency calculation

[0054] Based on the differences in the genetic background of the population in the target area (such as differences in mutation spectrum shown by historical migration, regional clustering or epidemiological surveys), the tested population is divided into several natural groups (for example, in Ningxia, groups can be formed based on the differences in mutation spectrum revealed by previous studies, avoiding the use of sensitive words, and using codes such as "Group A" or "Group B" or neutral expressions such as "Local Subgroup 1" or "Local Subgroup 2").

[0055] For each group, the number of alleles detected at each specific mutation site is counted, and then divided by the total number of alleles in that group (i.e., the number of individuals tested × 2) to obtain the allele frequency of that mutation in that group.

[0056] 95% confidence intervals were estimated for the frequency data, and the chi-square test or Fisher's exact test was used to compare the statistical significance of frequency differences between different groups (p<0.05 was considered significant).

[0057] Step 4: Data storage and dynamic update mechanism

[0058] The structured data above is stored in a relational database (such as MySQL or PostgreSQL). The data table design includes: a main mutation table (mutation ID, gene name, exon, nucleotide change, amino acid change, mutation type), a population group frequency table (group ID, mutation ID, allele frequency, sample size, confidence interval), and a pathogenicity rating table (graded according to the ACMG / AMP guidelines).

[0059] Establish a data update mechanism: Whenever a new confirmed case is added or a new local population screening study is completed, data is automatically extracted through the ETL process, and the allele frequencies of the corresponding groups are recalculated and updated to ensure the real-time nature of the database and local representativeness.

[0060] Upon receiving a data packet from a subject, the risk assessment unit's algorithm executes the following steps:

[0061] a. Genetic risk score calculation: For each variant in the genetic variant data, query its corresponding allele frequency (F) in the local area and among the subjects. Combine this with the known pathogenicity weight of the variant (W, which can be preset according to the ACMG guidelines) to calculate the risk score (WF) contributed by the variant. Summate the contribution values ​​of all variants to obtain the total genetic risk score.

[0062] b. Calculation of metabolic risk score: The metabolic risk score is calculated based on the degree of deviation of the subject's blood Phe concentration (Phe) from the reference median (M) of the same age group. The formula is `α|Phe-M| / M`, where α is an adjustable coefficient.

[0063] c. Synthesis of the Comprehensive Risk Index (CRI): The genetic risk score and the metabolic risk score are added together to obtain the final CRI. This index is a continuous variable; the higher the value, the greater the risk of disease.

[0064] Decision Unit: Based on the calculated CRI value, multiple preset dynamic thresholds are compared. For example, a "high-risk threshold" and a "concern threshold" are set. If CRI > high-risk threshold, the conclusion is "high risk, strong recommendation for diagnosis"; if CRI is between the two thresholds, the conclusion is "medium risk, short-term follow-up or clinical evaluation recommended"; if CRI is below the concern threshold, the conclusion is "low risk". For carrier screening, the threshold setting is more complex, for example, setting an extremely high risk threshold for cases where "both partners carry the pathogenic variant".

[0065] Process Management Unit: This is a workflow engine. Based on the conclusions of the decision-making unit, it automatically triggers subsequent tasks. For example, for "high-risk" newborns, it automatically sends recall notifications (SMS / APP push) to the responsible doctor and family, and creates a "Confirmation Follow-up" task in the system. For "high-risk" couples, it automatically generates genetic counseling appointment links. Simultaneously, this unit maintains the "Regional PKU Information Database," continuously archiving the complete information chain of all confirmed patients, providing a data source for database updates and scientific research analysis.

[0066] The regional mutation feature database is the core basis for the risk assessment of this invention. It is used to store allele frequency data of specific mutations in the phenylalanine hydroxylase gene in different population groups within the target screening area. The population groups in the regional mutation feature database are based on the long-term accumulated epidemiological and genetic data of the target screening area and are divided according to the gene mutation spectrum characteristics.

[0067] The screening management module makes tiered decisions based on CRI and drives the automated execution of the entire process. It consists of two parts: a decision-making unit and a process management unit.

[0068] The process management unit is also configured to build and maintain a regional phenylketonuria (PKU) information database. This database is used to structurally store the genotype, phenotype, and treatment follow-up data of diagnosed patients. The regional mutation characteristic database in the intelligent analysis module is periodically updated based on this regional PKU information database. The process management unit also includes a visualization dashboard subunit, configured to present administrators with a heatmap of real-time screening coverage, initial screening positivity rate, diagnostic rate, recall rate, and patient geographic distribution in the target screening area.

[0069] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor of a server, workstation, or mobile device, can implement all or part of the functional modules of the system described above.

[0070] In this embodiment, the system is deployed in a provincial maternal and child health care institution or a regional newborn screening center, and adopts a B / S architecture:

[0071] Cloud server: Runs data access service, intelligent analysis service, workflow engine service, and data storage service;

[0072] Grassroots client: Deployed at blood collection points and community health service centers for information entry and sample registration.

[0073] Laboratory terminal: It interfaces with tandem mass spectrometry detection systems and gene sequencing analysis systems to automatically acquire detection data.

[0074] Management side: Provides a visual dashboard for regional screening quality monitoring, statistical analysis, and task scheduling.

[0075] The screening procedure for phenylketonuria in newborns is as follows:

[0076] Newborns have their blood drawn at local blood collection points, their information is entered, and a unique sample ID is generated.

[0077] Laboratory tests detect blood phenylalanine concentration and PAH gene mutations; data is automatically uploaded.

[0078] The intelligent analysis module retrieves the frequency of regional mutations, calculates the CRI, and identifies areas as high-risk.

[0079] The screening management module automatically sends recall notifications and creates follow-up tasks;

[0080] Once the child has been diagnosed, they will begin treatment and long-term follow-up.

[0081] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0082] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A regionalized intelligent screening system for phenylketonuria, characterized in that: The system includes The data acquisition module is configured to acquire the subject's biometric identification information, gene variation data, blood phenylalanine concentration data, and the individual's population group information. An intelligent analysis module is communicatively connected to the data acquisition module, and the intelligent analysis module includes: A regional mutation signature database is used to store allele frequency data of specific mutations in the phenylalanine hydroxylase gene in different population groups within the target screening area; The risk assessment unit is configured to: call the regional mutation feature database, and calculate the comprehensive risk index of phenylketonuria for the subject based on the pathogenicity weight of the mutation site in the gene mutation data, the allele frequency of the mutation site in the corresponding population group of the subject, and the deviation of the blood phenylalanine concentration data from the reference range. The screening management module, which is communicatively connected to the intelligent analysis module, includes: The decision-making unit is configured to compare the comprehensive risk index with a preset dynamic threshold and generate a screening conclusion that includes the risk level and preliminary recommendations. The process management unit is configured to trigger and manage a series of tasks, including positive recall, genetic counseling appointments, and follow-ups, based on the screening results and the biosample identification information.

2. The regionalized intelligent screening system for phenylketonuria according to claim 1, characterized in that: The formula used by the risk assessment unit to calculate the comprehensive risk index is as follows: Wherein, CRI is the comprehensive risk index, W_i is the pathogenicity weight of the i-th detected gene variant, F_i is the allele frequency of the i-th gene variant in the subject population corresponding to the regional mutation frequency database, Phe is the measured serum phenylalanine concentration, M is the reference median value of serum phenylalanine concentration in the corresponding population, and α is the metabolic concentration influence coefficient.

3. The regionalized intelligent screening system for phenylketonuria according to claim 2, characterized in that: After receiving a data packet from a subject, the risk assessment unit performs the following steps: a. Gene risk score calculation: Traverse each variant in the gene variant data, query its corresponding allele frequency (F) in the local area and the subject, combine it with the known pathogenicity weight (W) of the variant, and calculate the risk score (WF) of the variant contribution according to the ACMG guidelines. Add up the contribution values ​​of all variants to obtain the total gene risk score. b. Calculation of metabolic risk score: The metabolic risk score is calculated based on the degree of deviation of the subject's blood Phe concentration (Phe) from the reference median (M) of healthy people of the same age. The formula is α|Phe-M| / M, where α is an adjustable coefficient. c. Synthesis of Comprehensive Risk Index (CRI): The genetic risk score and the metabolic risk score are added together to obtain the final CRI. This index is a continuous variable, and the higher the value, the greater the risk of disease.

4. The regionalized intelligent screening system for phenylketonuria according to claim 3, characterized in that: The decision-making unit compares the calculated CRI value with multiple preset dynamic thresholds.

5. The regionalized intelligent screening system for phenylketonuria according to claim 1, characterized in that: The population grouping in the regional mutation feature database is based on the long-term accumulated epidemiological and genetic data of the target screening area, and is divided according to the gene mutation spectrum characteristics.

6. The regionalized intelligent screening system for phenylketonuria according to claim 1, characterized in that: The process management unit is also configured to build and maintain a regional phenylketonuria information database, which is used to structurally store the genotype, phenotype and treatment follow-up data of confirmed patients, as well as the regional mutation feature database in the intelligent analysis module, and to perform periodic iterative updates based on the regional phenylketonuria information database.

7. The regionalized intelligent screening system for phenylketonuria according to claim 1, characterized in that: The process management unit also includes a visualization cockpit subunit, which is configured to present the administrator with a heat map of the real-time screening coverage, initial screening positivity rate, diagnosis rate, recall rate, and geographical distribution of patients in the target screening area.

8. The regionalized intelligent screening system for phenylketonuria according to claim 1, characterized in that: This system adopts a distributed architecture, including a cloud server deployed in the regional screening center and multiple clients deployed at various grassroots blood collection points. The cloud server runs the core services of the intelligent analysis module and the screening management module, while the clients are used to run some functions of the data acquisition module and to synchronize data and interact with the cloud server.

9. A computer-readable storage medium, characterized in that, When the computer-readable storage medium is executed by a processor, it implements the functions of the regional intelligent screening system for phenylketonuria as described in any one of claims 1 to 8.

Images