An IoT Platform for Epilepsy Monitoring and Supervising

Table 1

Main epilepsy detection platforms and solutions published so far. A description with some concerns with respect to the relevant factors is detailed. We suggest [10, 55] for further reading.

Ref.	Description
[41]	An EEG sensory device linked to a Smartphone, which performs some processing of the data, is used together with CC services. However, this is a very short communication and does not provide details of the solution.
[6]	This is mostly a research study focused only on the detection of tonic-clonic epileptic seizures. A WD without wireless communication stored data and some machine learning methods were performed for obtaining offline models.
[7]	In EpiCare, the Android app runs on the Smartphone device, gathering data from a EEG electrodes cap. This hardware makes this solution very efficient but rather uncomfortable. The main thing is that the project is focused on SUDEP; thus it is feasible to have a cap while sleeping. A mixture of CC and MCC solution is proposed, suggesting that an intelligent task delivering and allocation should be performed but without proposing any viable technique.
[34]	MyEpiPal is an app that monitors the patient, simplifies the intercommunication between the caregiver and the patient, and allows the self-management. This means that although it makes use of the sensory within the Smartphone, the main goal is to give support to the patient in everyday life. It is not an epilepsy detection platform itself, although the measurements can help in the prediction of the occurrence of a seizure, which is the reason that it is included in this comparison.
[39]	This research details the design of an ad hoc epilepsy detection ECG wireless intelligent sensor, including several detection algorithms, linked to a local computer connected to a network. Several relevant factors were analyzed, the ergonomic issues and the battery life among them. The WD communicates with the local computer in order to deliver alarms, to receive configuration commands, or to start/stop HR recording to be downloaded to the computer. A very detailed explanation of the requirements and of the hardware decisions is included.
[31, 32]	An EEG cap linked to a Smartphone is arranged to send the gathered data to CC services. Whenever a seizure is detected on the cloud, GPS locations are shared through the notification system. Neither the ergonomic issues nor the battery life and autonomy and the economic costs of the platform have been analyzed.
[37]	The UK Epilepsy Society published an app for Android and iPhone as a Web-log of the seizures, medication monitoring, and so on. This solution is a standalone solution, and the data gathered with this tool is not shared with any health service.
[38]	A standalone mobile solution is proposed using a commercial Smartwatch (MIO Alpha) linked to an Android phone. The app is responsible for gathering and processing the data, as well as the modules for the epileptic seizure detection. The solution does not consider intelligent modules, just some thresholds; this technique is by no means valid in this type of detection.