Information fusion of heterogeneous sensors for enriched personal healthcare activity logging

A literature review of current technologies on health data integration for patient-centered health management

“…Their mashup server uses open APIs of the commercial sensors and custom REST APIs of their mobile phone context logging service to connect various sensing data including location, weather, step count, sleep and so on, and manual logging data including food, exercise, mood and so on. Seo et al 49 introduced an information fusion method with sensor and activity models that classifies and aggregates data from heterogeneous physical, activity and social sensors. The proposed system is platform-independent compared with previous systems for a similar purpose, as it accesses to data using only REST Web APIs.…”

“…8 In addition, besides the health and fitness data, also the data related to social activities can be retrieved and integrated from services using Web APIs. 5,49 Although besides the Web API, there are other necessary connectivity protocols such as Bluetooth, 8,32 platform-bound APIs, 4,39,46 and so on, 31,44 it is better to simplify the types of connectivities for the matter of integration. Otherwise, it introduces complexity to the integration process, for example, maintaining different communication technologies, dealing with different Bluetooth communication modes and protocols.…”

“…Aboelfotoh et al 23 Yes Structural High Azarm et al 24 No Structural High Bentley et al 25 No Structural High Carbonaro et al 26 Yes Semantic Medium Chung and Park 27 No Structural High Corredor et al 28 Yes Structural Medium Fox et al 29 Yes Structural High Garai et al 30 Yes Structural High Gay and Leijdekkers 8 Yes Structural High Georgi et al 31 No Structural Medium Hong et al 32 Yes Structural High Jabbar et al 33 No Semantic Medium Kim et al 34 Yes Semantic Medium Kim and Chung 35 No Semantic Medium Kumar et al 4 No Structural Medium Mandel et al 36 Yes Semantic Medium Marceglia et al 37 Yes Structural High Mezghani et al 38 No Semantic Medium Pagkalos and Petrou 5 No Semantic Medium Pang et al 39 Yes Structural High Potter et al 40 No Semantic Medium Pustišek 41 No Structural High Puustjärvi and Puustjärvi 42 Yes Semantic Medium Rhayem et al 43 No Semantic Medium Roehrs et al 44 Yes Structural Medium Ruggeri and Briante 45 No Structural High Ryu et al 46 No Structural High Santos et al 47 No Structural High Schweitzer et al 48 Yes Structural Medium Seo et al 49 No Structural High Triantafyllidis et al 50 No Structural High Urbauer et al 51 Yes Structural High ontology, so that the integrated data are interpretable. The interoperability level relates to that if the approach involves Semantic Web technologies, or if any interoperability standard or medical coding system is applied.…”

“…Three studies tried to aggregate the health-related data from multiple sensors and Web services in a mashup way. 25,29,49 Bentley et al 25 presented a Health Mashups system, which can perform statistical analysis on the data across multiple sources and display observations as a natural language feed in a mobile application. Their mashup server uses open APIs of the commercial sensors and custom REST APIs of their mobile phone context logging service to connect various sensing data including location, weather, step count, sleep and so on, and manual logging data including food, exercise, mood and so on.…”

Meaningful Integration of Data from Heterogeneous Health Services and Home Environment Based on Ontology

“…The integration was implemented by leveraging a RESTful Web service on the application layer of the system to enable data sharing among applications. Seo et al proposed an information fusion method, which is platform independent that accesses data by using only REST Web APIs [32]. The sensor and activity models of the information fusion method classify and aggregate data from different types of sensors including physical, activity, and social sensors.…”