Health data are privacy-sensitive data. As a result, privacy is key to wider adoption of telemonitoring technologies. We aim to incorporate privacy by design in our framework protecting patients and healthcare providers from a range of possible dangers. Privacy can be enhanced not only by providing control to the subjects over their data, but also by working under the hood to ensure that the risk of other types of threats is reduced.
For example, some of the risk-assessment and intervention decisions can be based on the relative parameters of the subject to the population. Therefore, delegation of computation can help protect the privacy of the population by allowing the analysis to be performed on a trusted server without having to share any direct population data with the subject’s monitoring node. Our framework natively supports delegation of computation through request jobs.
One particularly interesting problem in health telemonitoring is the diagnosis inference by an unauthorized (and/or untrusted) party. To protect against such an attack, we have devised a privacy framework that enables private disclosure of information.
Stay tuned… More to come!