| Joint Industry-Academic Research Presentation with Professor Daesun Son of Hallym University… Sharing Future Clinical Data Operation Models
| JNPMEDI’s ‘AI Converter Engine’ Presented as Core Architecture… Simultaneously Ensuring Data Integrity and Traceability
| Aligned with Global Clinical Regulatory Trends Shifting Towards Real-Time Data-Based Review Systems
JNPMEDI, an AI platform-based consulting firm for drug and innovative medical device development, announced on the 23rd that its research paper has been selected for the official poster session at the ‘2026 CDISC EU Interchange’.
The 2026 CDISC EU Interchange is a global conference hosted by CDISC, a non-profit international organization that develops and manages international standards for clinical trial data. It is a venue where pharmaceutical companies, CROs, regulatory agencies, and technology firm representatives from all over the world participate to discuss the latest trends in clinical trial data management and regulatory submission.
The ‘Poster Session’ for which JNPMEDI was selected is an interactive presentation format where research findings are shared, and direct Q&A and discussions with participants take place on-site. Being selected for the poster session is significant as it signifies that the technical validity of the architecture has been recognized by CDISC, the most authoritative body governing global clinical data standards. Through this, JNPMEDI has further strengthened its position as a ‘First Mover’ designing future standards for global regulatory science, rather than just a solution vendor.
The paper, titled ‘Beyond Snapshots: Realizing “Zero-Submission” through a Continuous Regulatory Data Ecosystem in 2041’, was completed through joint industry-academic research between JNPMEDI and Professor Daesun Son of Hallym University. It was conducted using a future scenario and technical feasibility study methodology. In particular, it is attracting significant attention from industry stakeholders as it aligns with the latest global clinical regulatory trends, such as the US FDA's RTOR (Real-Time Oncology Review), which are shifting toward real-time data-based review systems.
Currently, despite the adoption of CDISC as the international clinical data standard, the pharmaceutical industry still relies on a ‘Static Snapshot’ method, where data from a specific point in time is processed and submitted in file format. The manual mapping and repetitive data reprocessing involved in this process are major causes of delays in the overall clinical schedule. Furthermore, the time gap between data generation and review hinders real-time decision-making.
To solve these issues, the research team proposed a ‘Continuous Regulatory Data Ecosystem’. This structure moves away from the method of delivering data, allowing regulatory agencies to directly access the approved data ecosystem to verify necessary information.
As a way to implement this, JNPMEDI presented an architecture based on the ‘AI Converter Engine’. This architecture consists of three core engines: ▲Intelligent Standardization Engine, ▲Metadata-Driven Analytics Engine, and ▲Continuous Pipeline.
First, the ‘Intelligent Standardization Engine’ uses AI to semantically analyze heterogeneous source data generated in hospitals and other facilities immediately upon collection. At the same time, it can perform real-time automatic mapping and conversion to the international standard SDTM domain without additional manual work.
The ‘Metadata-Driven Analytics Engine’ automatically generates analysis data (ADaM) and outputs (TLF) by interpreting statistical analysis plans when they are entered in a machine-readable metadata format. A key feature of this engine is that it automatically records the Data Lineage of all conversion processes to ensure complete traceability.
The ‘Continuous Pipeline’ implements end-to-end automation where downstream data structures, from SDTM to ADaM and TLF, are sequentially and automatically updated when source data is modified. By minimizing human intervention, it ensures Data Integrity and prevents human errors at the source.
Through this poster selection, JNPMEDI will officially promote its differentiated technology to the global market. At the same time, it has secured an opportunity to directly introduce its AI-based clinical data automation technology to companies and stakeholders attending the CDISC conference and continue discussions on practical cooperation.
Kwon-ho Jung, CEO of JNPMEDI, stated, “This research holds great industrial value as it presents a technical blueprint that can meet the needs of clinical data management and regulatory authorities by utilizing AI technology.” He added, “We will share JNPMEDI’s technological vision at global conferences and expand cooperation opportunities with domestic and international pharmaceutical companies and CROs.”
| Joint Industry-Academic Research Presentation with Professor Daesun Son of Hallym University… Sharing Future Clinical Data Operation Models
| JNPMEDI’s ‘AI Converter Engine’ Presented as Core Architecture… Simultaneously Ensuring Data Integrity and Traceability
| Aligned with Global Clinical Regulatory Trends Shifting Towards Real-Time Data-Based Review Systems
JNPMEDI, an AI platform-based consulting firm for drug and innovative medical device development, announced on the 23rd that its research paper has been selected for the official poster session at the ‘2026 CDISC EU Interchange’.
The 2026 CDISC EU Interchange is a global conference hosted by CDISC, a non-profit international organization that develops and manages international standards for clinical trial data. It is a venue where pharmaceutical companies, CROs, regulatory agencies, and technology firm representatives from all over the world participate to discuss the latest trends in clinical trial data management and regulatory submission.
The ‘Poster Session’ for which JNPMEDI was selected is an interactive presentation format where research findings are shared, and direct Q&A and discussions with participants take place on-site. Being selected for the poster session is significant as it signifies that the technical validity of the architecture has been recognized by CDISC, the most authoritative body governing global clinical data standards. Through this, JNPMEDI has further strengthened its position as a ‘First Mover’ designing future standards for global regulatory science, rather than just a solution vendor.
The paper, titled ‘Beyond Snapshots: Realizing “Zero-Submission” through a Continuous Regulatory Data Ecosystem in 2041’, was completed through joint industry-academic research between JNPMEDI and Professor Daesun Son of Hallym University. It was conducted using a future scenario and technical feasibility study methodology. In particular, it is attracting significant attention from industry stakeholders as it aligns with the latest global clinical regulatory trends, such as the US FDA's RTOR (Real-Time Oncology Review), which are shifting toward real-time data-based review systems.
Currently, despite the adoption of CDISC as the international clinical data standard, the pharmaceutical industry still relies on a ‘Static Snapshot’ method, where data from a specific point in time is processed and submitted in file format. The manual mapping and repetitive data reprocessing involved in this process are major causes of delays in the overall clinical schedule. Furthermore, the time gap between data generation and review hinders real-time decision-making.
To solve these issues, the research team proposed a ‘Continuous Regulatory Data Ecosystem’. This structure moves away from the method of delivering data, allowing regulatory agencies to directly access the approved data ecosystem to verify necessary information.
As a way to implement this, JNPMEDI presented an architecture based on the ‘AI Converter Engine’. This architecture consists of three core engines: ▲Intelligent Standardization Engine, ▲Metadata-Driven Analytics Engine, and ▲Continuous Pipeline.
First, the ‘Intelligent Standardization Engine’ uses AI to semantically analyze heterogeneous source data generated in hospitals and other facilities immediately upon collection. At the same time, it can perform real-time automatic mapping and conversion to the international standard SDTM domain without additional manual work.
The ‘Metadata-Driven Analytics Engine’ automatically generates analysis data (ADaM) and outputs (TLF) by interpreting statistical analysis plans when they are entered in a machine-readable metadata format. A key feature of this engine is that it automatically records the Data Lineage of all conversion processes to ensure complete traceability.
The ‘Continuous Pipeline’ implements end-to-end automation where downstream data structures, from SDTM to ADaM and TLF, are sequentially and automatically updated when source data is modified. By minimizing human intervention, it ensures Data Integrity and prevents human errors at the source.
Through this poster selection, JNPMEDI will officially promote its differentiated technology to the global market. At the same time, it has secured an opportunity to directly introduce its AI-based clinical data automation technology to companies and stakeholders attending the CDISC conference and continue discussions on practical cooperation.
Kwon-ho Jung, CEO of JNPMEDI, stated, “This research holds great industrial value as it presents a technical blueprint that can meet the needs of clinical data management and regulatory authorities by utilizing AI technology.” He added, “We will share JNPMEDI’s technological vision at global conferences and expand cooperation opportunities with domestic and international pharmaceutical companies and CROs.”