To provide comprehensive, scientific, and precise big data supports for national pediatric cancer prevention and control, the National Center for Pediatric Cancer Surveillance constructed the Surveillance Platform in 2019. Based on stratified and service-oriented design concepts, and a microservices architecture, the platform constructed five layers: document storage, data storage, service, application, and visualization. The platform supported three data reporting methods: automatic collection, file import, and manual entry. It ensured data quality through both rule-based and process-based quality control measures. Additionally, strict data security measures had been established in areas such as security domains, permission management, and data de-identification to ensure the safety and reliability of the monitoring data. As to November 2024, the platform had covered 1 750 surveillance sites(hospitals) and collected information about 6 million pediatric cancer cases, achieving positive results. This practice had laid a solid foundation for the smooth implementation of national pediatric cancer surveillance work and provided scientific evidences for pediatric cancer prevention and control in China. In the future, the platform′s performance needs to be continuously optimized and upgraded. It should further integrate relevant datasets in this field and actively explore and expand new application scenarios with the help of cutting-edge technologies such as big language models.

The enterprise master patient index is an important tool for identifying the diagnosis and treatment records of the same patient in heterogeneous medical data from multiple sources. From June to December 2021, the National Children′s Cancer Monitoring Center screened and determined the enterprise master patient index index system and its recognition logic by literature analysis and expert consultation. Based on the National Children′s Cancer surveillance Platform (hereinafter referred to as the surveillance platform), a corresponding intelligent recognition algorithm system was developed. After multiple rounds of real data verification and adjustment, a enterprise master patient index suitable for multi-source heterogeneous medical data was constructed. From January 2022 to March 2024, the intelligent recognition algorithm system had completed the recognition of 2.46 million pediatric tumor case report cards, established 0.33 million primary indexes and their unique identification codes for malignant tumor patients, and improved the data management and application efficiency of the surveillance platform. The enterprise master patient index based on surveillance platforms had played an important role in the registration and follow-up of pediatric cancer cases and related medical research, which could provide references for the construction of master indexes on other medical big data platforms in China.

The national pediatric cancer surveillance data known as the pediatric cancer case report card(report card), had the characteristics of wide sources, diverse collection methods and a large amount of information. Based on the characteristics of the surveillance data, the National Center for Pediatric Cancer Surveillance (surveillance center) established quality control program for surveillance data according to the relevant norms and standards from China and other countries. The program defined the variables, requirements and rules for the quality control of surveillance data. The surveillance center designed different quality control processes according to the way of data reporting including manual filling/file import and port docking, and formulated a series of supporting measures to achieve the completeness, accuracy and standardization of surveillance data. By analyzing the report cards of patients discharged from hospital from 2021 to 2023, the surveillance center found that the number of problem report cards decreased from 40.6% (202 185 cards / 497 538 cards) before feedback to 31.1% (157 725 cards / 506 817 cards) after feedback. The data quality control program not only improved the quality of surveillance data, but also provided references for the establishment of the data quality control program of other registration systems of medical field.

Building a nationwide representative sibling information database of pediatric cancer is of great significance for the research of pediatric cancer. In October 2022, based on the national pediatric cancer surveillance platform, the National Center for Pediatric Cancer Surveillance(NCPCS) identified and integrated the information of pediatric cancer cases using the patient master index, and then determined and retrieved the diagnosis and treatment information of pediatric cancer siblings through the sibling pair matching algorithm system, to establish the sibling information database. The information database was stored in the sibling database module of the surveillance platform, which realized the dynamic update, retrieval, download, and analysis of sibling information. The database provided data and technical support for the further childhood cancer research among siblings, as well as provided a reference for the construction of research-oriented databases for other disease surveillance systems. As of March 2024, this database had included 2 980 childhood cancer patients, collecting nearly 30 000 related medical records. In the future, NCPCS should further improve the sensitivity of sibling decision logic and expand the functionality of the sibling information database, so as to better meet the diverse needs of clinical and scientific research.

In-depth understanding of the clinical diagnosis and treatment practices of various health workers is of great significance for optimizing the allocation of health workforce. In 2023, based on the surveillance platform of National Center for Pediatric Cancer Surveillance(NCPCS), the NCPCS effectively integrated human resources data with clinical diagnosis and treatment data. By clarifying the conceptual and logical structures of the database, a clinical health workforce database was constructed using a distributed relational database. This database adhered to the data quality control principles of uniqueness, integrity, logic, and validity, and implemented scientific and effective data security protection strategies throughout the entire data life cycle. In practical applications, statistical analyses could be conducted on this database from two dimensions: health workforce and diagnosis-treatment processes, assisting relevant departments and hospitals in the refined management of health workforce allocation and promoting discipline construction. As of May 2024, the database had incorporated 931 hospitals, with the number of various health workers exceeding 640 000. The clinical health workforce database provided references for health administrative departments and hospitals at all levels to grasp the clinical practices of various health workers, and to achieve a clinical-demand-oriented allocation of health workforce.

To provide comprehensive, scientific, and precise big data supports for national pediatric cancer prevention and control, the National Center for Pediatric Cancer Surveillance constructed the Surveillance Platform in 2019. Based on stratified and service-oriented design concepts, and a microservices architecture, the platform constructed five layers: document storage, data storage, service, application, and visualization. The platform supported three data reporting methods: automatic collection, file import, and manual entry. It ensured data quality through both rule-based and process-based quality control measures. Additionally, strict data security measures had been established in areas such as security domains, permission management, and data de-identification to ensure the safety and reliability of the monitoring data. As to November 2024, the platform had covered 1 750 surveillance sites(hospitals) and collected information about 6 million pediatric cancer cases, achieving positive results. This practice had laid a solid foundation for the smooth implementation of national pediatric cancer surveillance work and provided scientific evidences for pediatric cancer prevention and control in China. In the future, the platform′s performance needs to be continuously optimized and upgraded. It should further integrate relevant datasets in this field and actively explore and expand new application scenarios with the help of cutting-edge technologies such as big language models.

The enterprise master patient index is an important tool for identifying the diagnosis and treatment records of the same patient in heterogeneous medical data from multiple sources. From June to December 2021, the National Children′s Cancer Monitoring Center screened and determined the enterprise master patient index index system and its recognition logic by literature analysis and expert consultation. Based on the National Children′s Cancer surveillance Platform (hereinafter referred to as the surveillance platform), a corresponding intelligent recognition algorithm system was developed. After multiple rounds of real data verification and adjustment, a enterprise master patient index suitable for multi-source heterogeneous medical data was constructed. From January 2022 to March 2024, the intelligent recognition algorithm system had completed the recognition of 2.46 million pediatric tumor case report cards, established 0.33 million primary indexes and their unique identification codes for malignant tumor patients, and improved the data management and application efficiency of the surveillance platform. The enterprise master patient index based on surveillance platforms had played an important role in the registration and follow-up of pediatric cancer cases and related medical research, which could provide references for the construction of master indexes on other medical big data platforms in China.

The national pediatric cancer surveillance data known as the pediatric cancer case report card(report card), had the characteristics of wide sources, diverse collection methods and a large amount of information. Based on the characteristics of the surveillance data, the National Center for Pediatric Cancer Surveillance (surveillance center) established quality control program for surveillance data according to the relevant norms and standards from China and other countries. The program defined the variables, requirements and rules for the quality control of surveillance data. The surveillance center designed different quality control processes according to the way of data reporting including manual filling/file import and port docking, and formulated a series of supporting measures to achieve the completeness, accuracy and standardization of surveillance data. By analyzing the report cards of patients discharged from hospital from 2021 to 2023, the surveillance center found that the number of problem report cards decreased from 40.6% (202 185 cards / 497 538 cards) before feedback to 31.1% (157 725 cards / 506 817 cards) after feedback. The data quality control program not only improved the quality of surveillance data, but also provided references for the establishment of the data quality control program of other registration systems of medical field.

Building a nationwide representative sibling information database of pediatric cancer is of great significance for the research of pediatric cancer. In October 2022, based on the national pediatric cancer surveillance platform, the National Center for Pediatric Cancer Surveillance(NCPCS) identified and integrated the information of pediatric cancer cases using the patient master index, and then determined and retrieved the diagnosis and treatment information of pediatric cancer siblings through the sibling pair matching algorithm system, to establish the sibling information database. The information database was stored in the sibling database module of the surveillance platform, which realized the dynamic update, retrieval, download, and analysis of sibling information. The database provided data and technical support for the further childhood cancer research among siblings, as well as provided a reference for the construction of research-oriented databases for other disease surveillance systems. As of March 2024, this database had included 2 980 childhood cancer patients, collecting nearly 30 000 related medical records. In the future, NCPCS should further improve the sensitivity of sibling decision logic and expand the functionality of the sibling information database, so as to better meet the diverse needs of clinical and scientific research.

In-depth understanding of the clinical diagnosis and treatment practices of various health workers is of great significance for optimizing the allocation of health workforce. In 2023, based on the surveillance platform of National Center for Pediatric Cancer Surveillance(NCPCS), the NCPCS effectively integrated human resources data with clinical diagnosis and treatment data. By clarifying the conceptual and logical structures of the database, a clinical health workforce database was constructed using a distributed relational database. This database adhered to the data quality control principles of uniqueness, integrity, logic, and validity, and implemented scientific and effective data security protection strategies throughout the entire data life cycle. In practical applications, statistical analyses could be conducted on this database from two dimensions: health workforce and diagnosis-treatment processes, assisting relevant departments and hospitals in the refined management of health workforce allocation and promoting discipline construction. As of May 2024, the database had incorporated 931 hospitals, with the number of various health workers exceeding 640 000. The clinical health workforce database provided references for health administrative departments and hospitals at all levels to grasp the clinical practices of various health workers, and to achieve a clinical-demand-oriented allocation of health workforce.