Machine learning algorithms - the possibilities are only limited by imagination
From its definition, a 'field of study that gives computers the ability to learn without being explicitly programmed', machine learning algorithms could sound like something out of a sci-fi film. If you've ever seen 'I, Robot', a film where robots start to think for themselves, you will know what I mean.
However, its applications aren't as futuristic as they sound and are already being applied to real world situations to provide valuable insights. We share just some of the applications and explore the possibilities of analysed data.
The growth of self-help/monitoring apps has really increased over the past 12 months, however not all apps collect and use data in a way that can provide reliable results and some are more for gimmick purposes. Ovuline, a fertility start-up, is not a gimmick app and has recently reported that its machine learning algorithms have helped 50,000 users get pregnant. The app collects and analyses data from its users, as well as other apps such as Fitbit, to predict fertility. As is the case with machine learning algorithms the more data available the more accurate predictions can be.
Machine learning algorithms can also be beneficial in monitoring the condition of machines. Prior to founding Oxehealth Professor Lionel Tarassenko worked with Rolls Royce, where the intelligent algorithms he helped create were used in detecting anomalies in jet engines. By analysing thousands of hours of data each day, collected from multiple sources, on the performance of its engines Rolls Royce were able to detect faults at an early stage and check and fix them before a failure or breakdown occurred. Similar intelligent algorithms can be found in the Oxecam.
The Oxecam, is based on the science of photoplethysmography and utilises sophisticated algorithms to monitor the vital signs of individuals through a camera. The technology has recently been validated in a clinical study of patients at the Oxford Kidney Unit and can be more accurate than traditional contact methods of monitoring. For example, the camera can take into consideration the activity of individuals and through its intelligence modify the parameters of data that are being analysed specific to the individual. The non-contact element of the Oxecam also ensures that certain errors endemic to traditional monitoring can be avoided; for example chest bands, which are used for measuring breathing rate, can easily move out of position and give erroneous readings.
Monitoring the health of patients is just one application of the Oxecam, in reality its possibilities are only limited to your imagination. To find out more or discuss possible applications please contact us.