New data-driven algorithms for intellingent remote patient monitoring systems from AIMLAb at the Technion Faculty of Biomedical Engineering.
With billions of mobile devices around the world and the low cost of embedded medical sensor, capturing and sharing medical data has never been simpler or quicker. It is now possible to obtain continuous and long-term complex physiological data. However, there have been little achievements in harnessing this “wealth” of physiological data to provide actionable clinical evidence.
Part of the problem stems from the wide variety of data content, the absence of data representation requirements (e.g., resolution, sampling, frequency, and metadata), and the use of comparatively limited datasets across certain trials, which struggle to capture the large range of variability between patients and time
Another problem is the lack of intelligent and accurate algorithms capable of decrypting the information stored in a vast number of data points obtained over time, referred as physiological time series. The advancement of machine learning algorithms in combination with current and novel wearable biosensors provides a once-in-a-life-time opportunity to enhance health screening and, as well as help patient management, especially through remote health monitoring.
Assistant Professor Joachim Behar’s Technion Artificial Intelligence in Medicine Laboratory (AIMLab) builds innovative pattern recognition algorithms to manipulate4 the information encoded within vast datasets of physiological time series. These new data-driven algorithms are being used by the AIMLab to develop innovative intelligent remote patient management systems.
