Deep learning and model personalization in sensor-based human activity recognition

Human activity recognition (HAR) is a line of research whose goal is to design and develop automatic techniques for recognizing activities of daily living (ADLs) using signals from sensors. HAR is an active research filed in response to the ever-increasing need to collect information remotely related to ADLs for diagnostic and therapeutic purposes. Traditionally, HAR used environmental or wearable sensors to acquire signals and relied on traditional machine-learning techniques to classify ADLs. In recent years, HAR is moving towards the use of both wearable devices (such as smartphones or fitness trackers, since they are daily used by people and they include reliable inertial sensors), and deep learning techniques (given the encouraging results obtained in the area of computer vision). One of the major challenges related to HAR is population diversity, which makes difficult traditional machine-learning algorithms to generalize. Recently, researchers successfully attempted to address the problem by proposing techniques based on personalization combined with traditional machine learning. To date, no effort has been directed at investigating the benefits that personalization can bring in deep learning techniques in the HAR domain. The goal of our research is to verify if personalization applied to both traditional and deep learning techniques can lead to better performance than classical approaches (i.e., without personalization). The experiments were conducted on three datasets that are extensively used in the literature and that contain metadata related to the subjects. AdaBoost is the technique chosen for traditional machine learning, while convolutional neural network is the one chosen for deep learning. These techniques have shown to offer good performance. Personalization considers both the physical characteristics of the subjects and the inertial signals generated by the subjects. Results suggest that personalization is most effective when applied to traditional machine-learning techniques rather than to deep learning ones. Moreover, results show that deep learning without personalization performs better than any other methods experimented in the paper in those cases where the number of training samples is high and samples are heterogeneous (i.e., they represent a wider spectrum of the population). This suggests that traditional deep learning can be more effective, provided you have a large and heterogeneous dataset, intrinsically modeling the population diversity in the training process.

Definition of an FHIR-based multiprotocol IoT home gateway to support the dynamic plug of new devices within instrumented environments

Internet of Things (IoT) technologies have become a milestone advancement in the digital healthcare domain, since the number of IoT medical devices is grown exponentially, and it is now anticipated that by 2020, there will be over 161 million of them connected worldwide. Therefore, in an era of continuous growth, IoT healthcare faces various challenges, such as the collection over multiple protocols (e.g. Bluetooth, MQTT, CoAP, ZigBEE, etc.) the interpretation, as well as the harmonization of the data format that derive from the existing huge amounts of heterogeneous IoT medical devices. In this respect, this study aims at proposing an advanced Home Gateway architecture that offers a unique data collection module, supporting direct data acquisition over multiple protocols (i.e.BLE, MQTT) and indirect data retrieval from cloud health services (i.e. GoogleFit). Moreover, the solution propose a mechanism to automatically convert the original data format, carried over BLE, in HL7 FHIR by exploiting device capabilities semantic annotation implemented by means of FHIR resource as well. The adoption of such annotation enables the dynamic plug of new sensors within the instrumented environment without the need to stop and adapt the gateway. This simplifies the dynamic devices landscape customization requested by the several telemedicine applications contexts (e.g. CVD, Diabetes) and demonstrate, for the first time, a concrete example of using the FHIR standard not only (as usual) for health resources representation and storage but also as instrument to enable seamless integration of IoT devices. The proposed solution also relies on mobile phone technology which is widely adopted aiming at reducing any obstacle for a larger adoption.