Design and development of a medical big data processing system based on Hadoop

Area of Science:

• Health Informatics
• Computer Science
• Data Science

Background:

• Secondary use of medical big data is crucial for healthcare services and clinical research.
• Traditional systems struggle with the Volume, Variety, and Velocity of big data.
• Understanding hospital information system (HIS) user behavior is key for system design and expansion.

Purpose of the Study:

• To investigate a practical Hadoop-based system for processing medical big data.
• To uncover features of hospital information system user behaviors using big data analytics.
• To demonstrate the efficiency of distributed MapReduce algorithms for medical big data.

Main Methods:

• Developed a Hadoop-based system for medical big data processing.
• Utilized Apache Hadoop MapReduce for parallel data processing.
• Implemented a five-node Hadoop cluster to execute distributed MapReduce algorithms.

Main Results:

• The Hadoop-based system effectively processes medical big data.
• Distributed MapReduce algorithms show superior efficiency compared to single nodes.
• Analysis uncovered features of hospital information system user behaviors.

Conclusions:

• Hadoop-based systems offer a scalable and reliable solution for medical big data.
• Distributed algorithms enhance the efficiency of medical big data processing in healthcare.
• Medical big data analytics can lead to more intelligent and user-friendly HIS with personalized recommendations.