A cloud-based framework for large-scale traditional Chinese medical record retrieval

Area of Science:

• Medical Informatics
• Health Big Data Analytics
• Traditional Chinese Medicine

Background:

• Electronic medical records (EMRs) are prevalent, increasing the need for effective data retrieval.
• Secondary use of medical records is crucial but faces challenges in large-scale data management.
• Retrieving complete patient information from big data requires robust systems.

Purpose of the Study:

• To propose an efficient and robust cloud-based framework for large-scale Traditional Chinese Medical Records (TCMRs) retrieval.
• To address the challenges of accurately and effectively retrieving relevant TCMRs from big data.
• To enhance the secondary use of TCMRs.

Main Methods:

• Developed a parallel index building method and a distributed search cluster for improved performance and high concurrency.
• Implemented a real-time multi-indexing model for up-to-date retrieval.
• Utilized semantics-based query expansion and a multi-factor ranking model to boost retrieval quality.
• Created a template-based visualization method for displaying medical reports via a web interface.

Main Results:

• Parallel indexing and distributed search significantly improve index building performance and concurrent retrieval.
• The multi-indexing model ensures real-time indexing and retrieval of the latest TCMRs.
• Semantics expansion and multi-factor ranking enhance the overall quality of retrieved records.
• Template-based visualization improves the availability and universality of medical report display.

Conclusions:

• The proposed cloud-based framework offers advantages for secondary use of large-scale TCMRs.
• The system demonstrates improved integration with existing clinical systems.
• The framework is versatile and applicable in various clinical scenarios.