Ontological Foundations for Tracking Data Quality through the Internet of Things

Area of Science:

• Health Informatics
• Data Science
• Ontology Engineering

Background:

• The Internet of Things (IoT) in Health promises improved patient records via sensor data.
• However, device fallibility can introduce errors into these records.
• Current quality control methods (inspection, testing, maintenance) are insufficient alone.

Purpose of the Study:

• To establish constant data quality monitoring for IoT in Health systems.
• To enhance the accuracy and completeness of longitudinal patient records.
• To leverage ontological principles for robust data validation.

Main Methods:

• Developing analytics procedures based on realism-based ontologies.
• Utilizing ontological principles of patients, bodily features, and measurement processes.
• Implementing unique identification for patients, caregivers, devices, and measurements.
• Analyzing patient data, including erroneous representations.

Main Results:

• Proposed a set of categories for analytics procedures to reason with.
• Highlighted the importance of unique identification for all entities in measurements.
• Demonstrated that 'metadata' can be treated as data about first-order entities.
• Established a framework for continuous data quality monitoring in healthcare IoT.

Conclusions:

• Continuous data quality monitoring through ontological analytics is crucial for reliable IoT in Health data.
• Unique identification of all measurement components enhances data integrity.
• Rethinking 'metadata' as first-order data strengthens data management and analysis.