Machine learning for robust mobile monitoring
Healthcare systems world-wide are entering a new, exciting phase: ever-increasing quantities of complex, massively multivariate data concerning all aspects of patient care are starting to be routinely acquired and stored, throughout the life of a patient, and increasingly involving mobile settings.
This exponential growth in data quantities far outpaces the capability of clinical experts to cope, resulting in a so-called “data deluge” in which mobile data are largely unexploited. There is huge potential for using advances in machine learning methodologies to exploit the contents of these complex mobile datasets by performing robust, scalable, automated inference to improve healthcare outcomes significantly by using patient-specific probabilistic models, a field in which there is little existing research, and which promises to develop into a new industry supporting the next generation of mobile healthcare technology.
Data integration across spatial scales, from molecular to population level, and across temporal scales, from fixed genomic data to a beat-by-beat electrocardiogram, will be one of the key challenges for exploiting these massive, disparate mobile-associated datasets.
This presentation aims to describe the urgently-needed interaction between machine learning and mobile healthcare technology. We demonstrate how advances in machine learning for healthcare (or “computational health informatics”) can cope with the noise and artefact typically present in mobile-acquired datasets, and how, perhaps for the first time, the resulting fruits of mobile monitoring can be used within clinical practice to improve patient outcomes.