Healthcare has witnessed a technological advancement in improving the quality of care and speeding the process of diagnosing patients due to its intervention with the internet of medical things. IoT in healthcare (H-IoT) plays a significant role in facilitating the process of diagnosing and detecting diseases. Different IoT-based medical sensors are used to measure biometrics and send them to the cloud for more analysis. However, the sensed data are massive and vary in their criticality level in which some sensed data are more critical (health-related data) than others. Moreover, computing such critical data in the cloud encounters some delay which is not preferable in real-time monitoring applications. Thus, this work proposes an IoT-fog-based framework to classify the streamed data according to their criticality level and compute the critical data in the fog to detect abnormalities with low latency and high response time. Before designing the proposed work, an analysis was conducted to explore the real data collected by IoT-based medical apps. The exploration of the data involved downloading and manually analyzing up-to-date privacy policies of eight IoT-based medical apps that provide details about data collection practices. The study showed that the streamed data in H-IoT include medical sensors data, apps registration data (personal information), device information, and other information related to cookies. The proposed work introduced the design of fog-based data classification and the algorithm for such classification. The implementation and evaluation of the proposed framework is future work.
|Number of pages||16|
|Journal||Assisted-Fog-Based Framework for IoT-Based Healthcare Data Preservation|
|Publication status||Published - 2021|